CA2351456A1 - Method and apparatus for providing differentiated services - Google Patents
Method and apparatus for providing differentiated services Download PDFInfo
- Publication number
- CA2351456A1 CA2351456A1 CA002351456A CA2351456A CA2351456A1 CA 2351456 A1 CA2351456 A1 CA 2351456A1 CA 002351456 A CA002351456 A CA 002351456A CA 2351456 A CA2351456 A CA 2351456A CA 2351456 A1 CA2351456 A1 CA 2351456A1
- Authority
- CA
- Canada
- Prior art keywords
- packet
- queue
- data
- forwarding
- packets
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/6215—Individual queue per QOS, rate or priority
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/622—Queue service order
- H04L47/623—Weighted service order
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/625—Queue scheduling characterised by scheduling criteria for service slots or service orders
- H04L47/627—Queue scheduling characterised by scheduling criteria for service slots or service orders policing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
- H04L49/9031—Wraparound memory, e.g. overrun or underrun detection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
- H04L49/9057—Arrangements for supporting packet reassembly or resequencing
Abstract
A method and apparatus for providing differentiated services using a multi- level queuing mechanism includes checking whether a packet of data is to receive a preferred level of service, and if the packet is not to receive th e preferred level of service, then placing the packet in a first forwarding queue. However, if the packet of data is to receive the preferred level of service then checking whether it is permissible to forward the packet to a device in the network at the current time. If it is permissible to forward t he packet to the device at the current time, then placing the packet in a secon d forwarding queue, otherwise temporarily placing the packet in a holding queu e before placing the packet in the second forwarding queue.
Description
WO 00130307 PCTlUS99/27381 METHOD AND APPARATUS FOR PROVIDING DIFFERENTIATED SERVICES
BACKGROUND OF THE INVENTION
Field of the Invention The present invention pertains to routing data in a network. More particularly, this invention relates to providing differentiated services using a mufti-level queuing mechanism.
Background Computer systems are increasingly becoming commonplace in homes and businesses throughout the world, and are increasingly becoming interconnected via networks. As the number of interconnected computer systems in use increases, so too does the amount of data being transferred by the networks (also referred to as the network "traffic"). Typically, the speed at which data can be transferred over the network decreases as the amount of network traffic increases because more and more data is trying to be transferred over the same amount of network hardware. For some types of data (e.g., phone calls or video conferencing), it is important that the data not be delayed very much during transfer over the network, regardless of the amount of network traffic.
One solution to this network traffic problem that has been presented is the use of "differentiated services" schemes. Generally, the idea of differentiated services is that certain types of data receive preferential treatment over the network, causing it to be transferred over the network faster than other types of data. This preferential treatment can be obtained in a variety of manners, such as by paying a premium price.
However, given the amount of data that is typically being transfi~rred over devices in the network, any mechanism to implement differentiated services should operate relatively quickly so as not to impede the transfer of data. Furthermore, care should be taken to ensure that the data that is receiving preferential treatment does not exceed a preconfigured amount or rate (e.g., the amount or rate that has been paid for). Currently, there are no differentiated services implementations that provide quick and efficient handling of differentiated services data.
BACKGROUND OF THE INVENTION
Field of the Invention The present invention pertains to routing data in a network. More particularly, this invention relates to providing differentiated services using a mufti-level queuing mechanism.
Background Computer systems are increasingly becoming commonplace in homes and businesses throughout the world, and are increasingly becoming interconnected via networks. As the number of interconnected computer systems in use increases, so too does the amount of data being transferred by the networks (also referred to as the network "traffic"). Typically, the speed at which data can be transferred over the network decreases as the amount of network traffic increases because more and more data is trying to be transferred over the same amount of network hardware. For some types of data (e.g., phone calls or video conferencing), it is important that the data not be delayed very much during transfer over the network, regardless of the amount of network traffic.
One solution to this network traffic problem that has been presented is the use of "differentiated services" schemes. Generally, the idea of differentiated services is that certain types of data receive preferential treatment over the network, causing it to be transferred over the network faster than other types of data. This preferential treatment can be obtained in a variety of manners, such as by paying a premium price.
However, given the amount of data that is typically being transfi~rred over devices in the network, any mechanism to implement differentiated services should operate relatively quickly so as not to impede the transfer of data. Furthermore, care should be taken to ensure that the data that is receiving preferential treatment does not exceed a preconfigured amount or rate (e.g., the amount or rate that has been paid for). Currently, there are no differentiated services implementations that provide quick and efficient handling of differentiated services data.
Thus, a need exists for an improved way to implement differentiated services.
WO 00/30307 PCTJUS99/273$I
WO 00/30307 PCTJUS99/273$I
SUMMARY OF THE IN'JENTION
A method and apparatus for providing differentiated services using a mufti-level queuing mechanism is described herein. According to one aspect of the present invention, a method for providing differentiated services in a network includes checking whether a packet of data is to receive a preferred leve;I of service, and if the packet is not to receive the preferred level of service, then placing the packet in a first forwarding queue. However, if the packet of data is to receive the preferred level of service then checking whether it is permissible to forward the packet to a device in the network at the current time. If it is permissible to forward the packet to the device at the current time, then placing the packet in a second forwarding queue, otherwise temporarily placing the packet in a holding queue before placing the packet in the second forwarding queue.
According to another aspect of the present invention, an apparatus for providing differentiated services in a network includes first and second forwarding queues for forwarding packets of data to another device in the network, and also includes a holding queue to temporarily store packets of data prior to being placed in the second forwarding queue. The apparatus further includes a policy meter to check whether a packet of data is to receive a preferred level of service, and, if the packet is not to receive the preferred level of service, then to place the packet in the first forwarding queue.
However, if the packet of data is to receive the preferred level of service, then the policy meter checks whether it is permissible to forward the packet to the device at the current time. If it is permissible to forward the packet to the device at the .current time, then the policy meter places the packet in the second forwarding queue, and. otherwise places the packet in the holding queue.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention is illustrated by way of example and not limitation in the figures of the accompanying drawings, in which like references indicate similar elements and in which:
Figure 1 illustrates a network environment into which one embodiment of the present invention can be incorporated;
Figure 2 is a diagram illustrating a metering and shaping device according to one embodiment of the present invention;
A method and apparatus for providing differentiated services using a mufti-level queuing mechanism is described herein. According to one aspect of the present invention, a method for providing differentiated services in a network includes checking whether a packet of data is to receive a preferred leve;I of service, and if the packet is not to receive the preferred level of service, then placing the packet in a first forwarding queue. However, if the packet of data is to receive the preferred level of service then checking whether it is permissible to forward the packet to a device in the network at the current time. If it is permissible to forward the packet to the device at the current time, then placing the packet in a second forwarding queue, otherwise temporarily placing the packet in a holding queue before placing the packet in the second forwarding queue.
According to another aspect of the present invention, an apparatus for providing differentiated services in a network includes first and second forwarding queues for forwarding packets of data to another device in the network, and also includes a holding queue to temporarily store packets of data prior to being placed in the second forwarding queue. The apparatus further includes a policy meter to check whether a packet of data is to receive a preferred level of service, and, if the packet is not to receive the preferred level of service, then to place the packet in the first forwarding queue.
However, if the packet of data is to receive the preferred level of service, then the policy meter checks whether it is permissible to forward the packet to the device at the current time. If it is permissible to forward the packet to the device at the .current time, then the policy meter places the packet in the second forwarding queue, and. otherwise places the packet in the holding queue.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention is illustrated by way of example and not limitation in the figures of the accompanying drawings, in which like references indicate similar elements and in which:
Figure 1 illustrates a network environment into which one embodiment of the present invention can be incorporated;
Figure 2 is a diagram illustrating a metering and shaping device according to one embodiment of the present invention;
Figure 3 is a flowchart illustrating the steps followed by a policy meter and shaper according to one embodiment of the present invention;
Figure 4 is a flowchart illustrating the steps followed by a shaper in placing a packet into the hold queue according to one embodiment of the present invention;
Figure 5 is a block diagram illustrating an example holding queue according to one embodiment of the present invention; and Figure 6 is a flowchart illustrating the steps followed in forwarding packets from a holding queue to a premium output queue according 1:0 one embodiment of the present invention.
Figure 4 is a flowchart illustrating the steps followed by a shaper in placing a packet into the hold queue according to one embodiment of the present invention;
Figure 5 is a block diagram illustrating an example holding queue according to one embodiment of the present invention; and Figure 6 is a flowchart illustrating the steps followed in forwarding packets from a holding queue to a premium output queue according 1:0 one embodiment of the present invention.
DETAILED DESCRIIPTION
In the following detailed description numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances well known methods, procedures, components, and circuits have not been described in detail so as not to obscure the present invention.
In alternative embodiments, the present invention may be applicable to implementations of the invention in integrated circuits or chip sets, wireless implementations, switching systems products and trarnsmission systems products. For purposes of this application, the terms switching systems products shall be taken to mean private branch exchanges (PBXs}, central office switching systems that interconnect subscribers, toll/tandem switching systems for interconnecting trunks between switching centers, and broadband core switches found at the center of a service provider's network that may be fed by broadband edge switches or access multiplexors, and associated signaling, and support systems and services. The terra transmission systems products shall be taken to mean products used by service providers to provide interconnection between their subscribers and their networks such as Loop systems, and which provide multiplexing, aggregation and transport between a service provider's switching systems across the wide area, and associated signaling and support systems and services.
Some portions of the detailed descriptions which follow are presented in terms of algorithms and symbolic representations of operation;; on data bits within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self consistent sequence of steps leading to a desired result. The steps are those requiring physical manipulations of physical quantities. Usualhr, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like. It should be borne in mind, however, that aI1 of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the following discussions, it is appreciated that throughout the present invention, discussions utilizing terms such as "processing" or "computing" or "calculating" or "determining" or "displaying" or the like, refer to the action and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices.
The present invention makes use of a multi-level queuing structure to provide differentiated services. Packets are forwarded into a network supporting differentiated services by using the multi-level queue structure. A first queue level is a set of forwarding queues into which packets are placed prier to being forwarded into the network. The amount of time a packet remains in the first queue level is dependent on the total number ofpackets in the first level as well as the rate at which the forwarding device is able to provide packets to the network. Additionally, packets that are to receive a premium service level may also be temporarily placed into a second level queue, referred.
to as a hold queue, prior to being placed in the frst queue level. The present invention ensures that packets are transferred from the hold queue into a first level queue at not higher than a preconfigured rate, and thus that the rate of transfer of premium service level packets to the network does not exceed the preconfigured rate.
Figure 1 illustrates a network environment into which one embodiment of the present invention can be incorporated. A network 110 is illustrated interconnecting multiple devices 12I, 122, 123, 124, and 130. Netwa~rk 110 is intended to represent any of a wide variety of conventional networks, such as the Internet, an intranet, a local area network (LAN), a subset of a larger network (e.g., a portion of the Internet), etc. Devices 121-I24 and 130 are intended to represent any of a w de variety of conventional devices that can receive data from and/or transfer data to network 1 I 0. Examples of such devices include "end points", such as workstations and server~~, or connections to other networks, such as routers, bridges, gateways, etc.
Network 110 is configured to provide differentiated services to data packets.
According to one embodiment of the present invention, the differentiated services include three different levels of service: best effort service, assured service, and premium service.
Best effort service is the lowest level, guaranteeing only that the network will do its best to try and deliver packets of data to their destinations. Currently, many networks (such as the Internet) provide best effort service. Assured service is a higher level of service, ensuring that packets of data are unlikely to be dropped from the network so long as the amount of such data stays within a particular profile (e.g., not larger than a particular size or greater than a particular rate). Premium service i;c the highest level of service. Packets of data at the premium service level are given preferential treatment through the network, and are almost certain not to be dropped from the network so long as the amount of such data stays within a particular profile (e.g., premium packets are not sent at greater than a particular rate).
It is to be appreciated that network I 10 includes multiple components, including any combination of routers, switches, bridges, gateways, workstations, servers, etc. to transfer packets of data. In order to support differentiated services, each of the components within network i I O needs to be able to identify and properly treat packets of different service levels. This can include, for example, the components giving priority to higher service level packets by forwarding them prior to any lower service level packets, or using dedicated paths or channels for higher priority packets. According to one embodiment of the present invention, each of the packets of data being transferred through the network includes a packet header that identifies the service level for the packet. Alternate embodiments can identify service levels in other manners, such as by separate signaling or use of dedicated input and output ports.
According to one embodiment of the present invention, packets of data to be sent using premium and assured service levels are required to conform to particular profiles.
For the assured service level, the particular profile identifies a particular "expected capacity" (e.g., not greater than a particular rate, such as 100k bits per second; or a particular burst rate, such as four packets back-to-back). For the premium service level, the particular prof le identifies a particular packet rate that will not be exceeded (e.g., not greater than twenty packets per minute).
Edge devices, such as, shaping/metering device 130, operate to ensure that such profiles are not violated. By doing so at the edge devices at which data is being input to network 110, each component within network 110 is not required to verify that packets of data conform to the requirements for their identified Ieve1 of service.
Thus, in the illustrated embodiment of Figure l, data is received by shaping/metering device 130 from a source 135. It is to be appreciated that source 135 can be an end point (e.g., a server or workstation), or alternatively a connection to another WO 0013030'1 PCT/US99/2'1381 _g_ network (e.g., a bridge or gateway to a company's LAN). Upon receipt of a packet of data from source 135, device 130 analyzes the packet to ensure that the profiles are nofi being violated and forwards packets to network 110. If one of the profiles is being violated, then packets of data may be dropped or re-labeled to be forwarded to network 110 at a different service level. Packets that are dropped are simply ignored by device 130 and are not forwarded to network 110. The operation of device 130 in ensuring that profiles are not being violated is discussed in more detail below with reference to the following figures.
It should be noted that devices 12I-124 may also provide metering and shaping functionality for packets of data received by the devices from other sources (not shown) to be provided to network 110, as well as for packets of data being forwarded to other destinations by the devices.
Figure 2 is a diagram illustrating a metering and shaping device according to one embodiment of the present invention. In the discussion of Figure 2 and subsequent f gores, reference is made to the forwarding and/or transferring of packets by different logic units of device 130. According to one embodiment of the present invention, device 130 includes a data buffer (not shown) having multiple locations for the storage of data packets. When a data packet is received by device 130, the data is stored in a location of the buffer and remains in that location until the data i~ forwarded out of device 130 to network 110. The discussion of "movement" of such packets within device 130 in this embodiment refers to the transferring or forwarding of packet "pointers" which identify the locations of the buffer. However, according to ali:ernate embodiments of the present invention, the actual data itself may be transferred or forwarded between multiple buffers (not shown) within device 130.
As illustrated in Figure 2, data packets are received by device I30 from source 135. Device 130 is capable of handling multiple "flows". A flow refers to a particular input port to output port mapping of device I30. According to one embodiment, each flow is identified by a set of identifying information in the packet header, including the Internet Protocol (IP) source address, IP destination address, source port, destination port, and protocol identifier. These five pieces of information are also collectively referred to as a "5-tuple". A packet of data received by device I:SO is input to classifier 205.
Classifier 205 compares a set of preprogrammed 5-tuples (which can include "wildcards") to the received 5-tuple for the packet and outputs an identifier of a particular flow for the packet. The number of flows within a device, such as a muter, is typically on the order of 30 to 50. However, it is to be appreciated that fewer or greater flows could be included within a device. The use of classifiers and flows within devices is well known to those skilled in the art and thus will not be discussed further except as it pertains to the present invention.
The identifier output from classifier 205 identifies a particular shaper/policy meter to which the packet is forwarded, as well as ultimately a set of forwarding queues and a forwarding engine. For ease of explanation, only a single shaper/policy meter 2I0 and forwarding engine 230 are illustrated in Figure 2. It is to be appreciated, however, that a different shaper/policy meter and forwarding engine (as well as holding queue and forwarding queues) can be included for each of the different flows supported by device 130. Although, it is also to be appreciated that in embodiments where shaper/policy meter 210 are implemented in software, various procedures and sections of software code can be shared by different flows.
The packet is then provided to policy meter and shaper 2I0. Shaper/meter 210 accesses the service type information in the packet header (typically provided by the original source) to determine whether the packet is to receive a preferred level of service.
Based on the level of service the packet is to receive, as well as when previous packets of data have been forwarded by device 130, shaper/meter 210 either forwards the packet to shaper 214 for temporary storage in hold queue 218 or to forwarding engine 230.
Shaper/meter 210 is responsible for ensuring that the configured policies for different levels of service are maintained. By way of example, if the owner of a particular source has contracted with the owner of device 130 that the source will not attempt to send greater than x number of premium level packets per minute to device 130 fox forwarding to nerivork 110, shaper/meter 210 ensures that this "contracted for" rate is not exceeded. In fact, according to one embodiment of the present invention if the contracted for rate is exceeded then some of the premium level packets may be dropped.
Eventually, packets that are not dropped are forwarded to one of the output queues 232 or 234. The packets are then forwarded on to network 110 from queues 232 and 234 by forwarding engine 230. Packets of data at the premium service level are placed in premium output queue 234, while the packets at the a;~sured or best effort service levels are placed in secondary output queue 232. Forwarding engine 230 controls the w0 00/30307 PCT/US99/27381 forwarding of packets to network 1 I 0, with priority being given to packets in premium output queue 234.
According to one embodiment of the present invention, forwarding a packet of data to one of the output queues 232 or 234 is accomplished by sending the packet to forwarding engine 230. The placement of the packet into the proper one of queues 232 and 234, as well as the proper location within that queue, is performed by forwarding engine 230. Forwarding engine 230 may determine the proper one of queues 232 and 234 based on service level information contained in the header of the data packet, or alternatively from parameters received from shaper/meter 210 when the packet is transferred to forwarding engine 230. According to one embodiment of the present invention, each of queues 232 and 234 is a conventional first-in first-out (FIFO) queue.
Forwarding engine 230 can use any of a wide: variety of conventional arbitration algorithms for selecting packets from queues 232 and 234 for output to network 110.
Examples of such algorithms include giving any pacl~et in premium output queue priority over any packet in secondary output queue 232, and a weighted round robin scheme that favors output queue 234. In the illustrated embodiment, forwarding engine 230 outputs data to network I 10 at its fastest possible: speed, which is dependent on the physical connection between device 130 and networlw; 110.
According to one embodiment of the present invention, classifier 205, shaper/meter 210, and forwarding engine 230 are implemented in software. in this embodiment, software instructions to carry out the processes of classifier 205, shaper/meter 210, and forwarding engine 230 are stored in a system memory (not shown) of device 130 of Figure 2 and executed by a processor (not shown) of device 130.
According to one alternate embodiment of the present invention, classifier 205, shaper/meter 210, and forwarding engine 230 are implemented in firmware, such as a ROM or Flash memory of the device 130. According to another alternate embodiment of the present invention, classifier 205, shaper/meter 210, and forwarding engine 230 are implemented in hardware, such as an application specific integrated circuit (ASIC) of device 130.
In the discussion to follow, reference is made to various times, such as "the current time". In the illustrated embodiment, time is referenced to a clock that "ticks" at a particular "tick time". The tick time is a value (e.g., in microseconds) that is calculated by dividing the minimum data packet size (64 bytes according to one implementation) by the maximum output line rate for device 130 (10 Megabits per second or l00 Megabits per second according to one implementation).
Figure 3 is a flowchart illustrating the steps followed by a policy meter and shaper according to one embodiment of the present invention. Shaper/meter 210 of Figure 2 first receives a packet of data (e.g., from classifier 205), step 302. Upon receipt of the packet, policy meter 212 checks whether the packet is to receive premium service, step 304. As discussed above, the packet header identifies the typo; of service the packet is to receive, so policy meter 212 can make this determination based on the packet header information.
If the packet is to receive premium service, then policy meter 212 checks whether it is permissible to send the packet at the present time, step 306: In the illustrated embodiment, policy meter 212 maintains a record of when it is permissible to send the next packet, referred to as the "okay to send" value or time. The okay to send time is updated each time policy meter 212 forwards a newly received premium service packet to either premium output queue 234 or holding queue 218. In step 306, policy meter 2i2 compares the current time to the okay to send time anal determines that it is permissible to send the packet at the current time if the current time is greater than or equal to the okay to send time.
If it is permissible to send the packet at the ciurrent time, then policy meter 212 marks the packet as "premium service", step 308. Thiis marking is, according to one embodiment of the present invention, a predetermined bit pattern in the packet header that is expected by the components of network 110 to identify premium service packets.
Aitennatively, the type of service information in the header that was originally provided to device 130 from the source 135 could be used by the components of network 110, and the marking step not performed.
Policy meter 212 then updates the okay to send time, step 310. The okay to send time is updated to be the current time plus the amount of time necessary to send the packet from the premium output queue 234 to network 110. It should be noted that this amount of time necessary to send the packet is the amount of time necessary to send the packet to network 110 at the "contracted for" rate, which may be longer than the actual time necessary to send the packet to network 110. By way of example, the connection to network 110 may be fast enough to support 50 packets per second, although the contracted for rate is 10 packets per second. Thus, the amount of time necessary to send the packet would be 0.1 seconds (1 ( 10) rather than 0.02 seconds (1 ( 50).
In the illustrated embodiment, the amount of time necessary to send the packet from the premium output queue 234 to network 110 is determined based on the actual size of the packet of data {e.g., number of bytes), including any header and/or tail information. According to one embodiment of the present invention, a lookup table is used that includes both the packet size and a corresponding amount of time.
This lookup table can be pre-computed, based on the known cloclk rate being used by policy meter 212 as well as the output transfer rate of device I30 and the contracted for rate.
In one implementation, the lookup table includes a different: entry for each possible packet size.
According to an alternate implementation, ranges of packet sizes are used (e.g.; less than I00 bytes, between 100 and 200 bytes, between 200 and 400 bytes, etc.).
After updating the okay to send value, policy meter 212 places the packet into the premium output queue 234, step 312. At this point, forwarding engine 230 is responsible for forwarding the packet to network 110.
Returning to step 306, if it is not permissible t:a send the packet at the current time, policy meter 212 checks whether there is sufficient space in hold queue 218 to store an additional packet of data, step 314. According to one implementation, the determination of whether there is sufficient space in hold queue 218 is made based on a maximum amount of time into the future that a packet will be held by shaper/meter 210 for placing into the forwarding queue. Policy meter 212 compares the current okay to send time minus the current time to the maximum amount of time into the future that a packet will be held for placing into the forwarding queue. If the current okay to send value minus the current time exceeds the maximum amount of time, then there is insufficient space in hold queue 218. Otherwise, there is sufficient space in hold queue 218.
If policy meter 212 determines that there is insufficient space in hold queue 218, then policy meter 2I2 drops the data packet, step 316. In other words, the data packet is ignored by device 130 and is not forwarded to network 110. According to an alternate embodiment, rather than dropping the packet, policy meter 212 re-marks the packet (e.g., as best effort, step 334 discussed below), and forward s the packet to network 110 with the new marking.
However, if policy meter 2I2 determines that there is sufficient space in hold queue 218, then policy meter 212 marks the packet as "premium service", step 318, analogous to step 308 discussed above. Policy meter :? 12 then updates the okay to send time, step 320. The okay to send time is updated to be the current okay to send time plus _13_ the amount of time necessary to send the packet from the premium output queue 234 to network 110. It should be noted that, analogous to the discussion above, this amount of time necessary to send the packet is the amount of time necessary to send the packet to network 110 at the "contracted for" rate, not necessarily the actual time necessary to send the packet to network 110.
The packet is then placed in the hold queue, step 322. Policy meter 212 forwards the packet as well as the current okay to send time to shaper 214, which in turn places the packet into the hold queue.
Figure 4 is a flowchart illustrating the steps followed by shaper 214 in placing a packet into the hold queue according to one embodinnent of the present invention. Shaper 214 f rst checks whether the difference between the okay to send time and the current time is less than a threshold amount, step 402. If the difference is less than the threshold amount, then shaper 214 places the packet in premiu~:n output queue 234, step 312 of Figure 3. In one implementation, this threshold amount is on the order of 10-12 clock ticks. This comparison to a threshold amount is made to prevent shaper 214 from expending the time to place a packet into the hold queue if it is going to be forwarded out of the queue very quickly anyway. If the difference in step 402, however, is not less than the threshold amount, then shaper 2I4 masks the okay to send time by the size of the queue, step 404. This masking generates a pointer into the hold queue based on the okay to send time.
Figure 5 is a block diagram illustrating an example holding queue according to one embodiment of the present invention. As illustrated, holding queue 500 is a "circular" queue having storage capacity for y packets of data. The masking of the okay to send time by the size of the queue causes the pointer to wrap around queue 500.
Returning to Figure 4, shaper 214 f nds the first available location in the holding queue using the pointer (from step 404), step 406. According to one implementation, the first available location is found by first accessing the .queue location identified by the pointer generated in step 404. If that queue location is available (that is, no other packets are currently being stored there), then the finding process is complete.
Otherwise, the pointer is incremented by one and the newly pointed to queue location is checked for availability. If the newly pointed to queue location is available, then the finding process is complete, otherwise the incrementing and checking continues until an available location is found. It is to be appreciated that, because the determination was made in step 314 of Figure 3 that there is sufficient space in the hold queue, eventually an available location will be found in step 406.
Upon finding the first available location within the queue, shaper 214 places the packet into that queue location, step 408.
According to one alternate embodiment of the present invention, shaper 214 does not perform step 402. In this alternate embodiment, shaper 214 always places packets into holding queue 218 without regard for the difference between the okay to send time and the current time. According to another alternate embodiment of the present invention, different processes are used to find an available location in holding queue 218.
By way of example, different starting locations can be used for starting the search for the first available queue location, or different incrementa~tion values can be used.
Returning to Figure 3, once the packet is placed in the hold queue, shaperlrneter 210 waits until it is permissible to send the packet to network 110, step 324.
Once it is okay to send the packet, sender 216 of Figure 2 places the packet into premium output queue 234, step 312.
Figure 6 is a flowchart illustrating the steps followed in forwarding packets from holding queue 218 to premium output queue 234 according to one embodiment of the present invention. An interrupt is generated every tick time, step 602, by a component (not shown) of the system in which the present invention is embodied. In response to the interrupt, sender 2I 6 accesses the current location of :hold queue 218, step 604. Sender 216 maintains a pointer into hold queue 218 that is incremented by one in response to each interrupt, continually wrapping around the queue. Sender 216 checks whether there is a packet in the current location of hold queue 218, :>tep 606. If there is a packet, then sender 216 forwards the packet to premium output queue 234, step 608.
Otherwise, the process simply ends, and then repeats at the next interrupt.
Returning to step 304 of Figure 3, if the packet is not to receive premium service, then policy meter 2i2 checks whether the packet is to receive assured service, step 326.
As discussed above, the packet header identifies the t~rpe of service the packet is to receive, so policy meter 212 can make this determination based on the packet header information. If the packet is to receive assured service, then policy meter 2I2 checks whether the packet size is within a permitted burst size, step 328. The permitted burst size is a preconfigured maximum size or rate which packets at the assured service level are not to exceed. Typical values for the permitted burst size axe on the order of two to four packets. According to one implementation, policy meter 212 maintains a count of packets received at the assured level of service. This count is reset to zero each time a packet of other than the assured service level is received, thereby allowing policy meter 212 to know when the permitted burst size has been exceeded.
If the packet is within the permitted burst size, then policy meter 212 marks the packet as "assured service", step 330. Analogous to the premium service marking discussed above with reference to step 308, the assured service marking is a predetermined bit pattern in the packet header that is expected by the components of network 110 to identify assured service packets. Policy meter 212 then places the packet into the secondary output queue 232, step 332. At this point, forwarding engine 230 is responsible for forwarding the packet to network 110.
Returning to steps 326 and 328, if policy meter 212 determines that the packet is nat to receive assured service (step 326), or that the packet is not within the permitted burst size (step 328), then policy meter 212 marks the packet as "best effort service", step 334. Analogous to the premium service maxking discussed above with reference to step 308, the best effort service marking is a predeterminerd bit pattern in the packet header that is expected by the components of network 110 to identify best effort service packets.
Policy meter 212 then places the packet into the secondary output queue 232, step 332.
At this point, forwarding engine 230 is responsible for forwarding the packet to network 110.
In the discussions above, reference is made to various specific units of measurement (e.g., numbers of bytes). Alternate embodiments of the present invention can be based on different units either smaller or larger than the specific units discussed above (e.g., bits or multiple-byte wards).
Also in the discussions above, reference is made to three levels of service (premium, assured, and best effort). It is to be appreciated that the present invention is not limited to the use of only three levels of service, and that greater or fewer levels could be implemented. By way of example, the assured level of service need not be included, or additional levels of "preferential" service may be added. It is also to be appreciated that additional holding queues and/or forwarding queues can be added to accommodate such additional levels of service.
Also in the discussions above, reference is mare to device 130 of Figure 2 including multiple shaper/meters 210, one for each flow through device 130. In alternate embodiments, a single shaper/meter 210 may be used to control multiple flows through device 130.
Also in the discussions above, reference is made to multiple forwarding queues 232 and 234. Alternatively, a single queue could be used with additional control logic in forwarding engine 230 to distinguish between premium and non-premium packets within the single queue. Similarly, rather than having individual forwarding queues 232 and 234 for each flow, a single queue structure could be used with multiple flows, with additional control logic in forwarding engine 230 to distinguish between the different flows.
Also in the discussions above, reference is made to packets being dropped under certain circumstances. For example, if a packet is to receive premium service and the holding queue is full (as discussed with reference to Figure 3). According to alternate embodiments, such packets may be re-classified (e.g", to best effort service) rather than being dropped. Similarly, reference is made to packets being re-classified in certain under certain circumstances (e.g., a packet to receive assured service is not within a permitted burst size}. According to alternate embodiments, such packets may be dropped rather than re-classified.
Therefore, a method and apparatus for providing differentiated services using a mufti-level queuing mechanism has been described. :Packets of data that are to receive a premium level of service can be temporarily stored in: a holding queue prior to being placed in a forwarding queue and subsequently transferred over the network.
The temporary storage of the packets allows the device to ensure that a particular transfer rate of premium packets is not exceeded. Furthermore, the placement of premium level packets into the holding queue, and thus the determination of when it is okay to send the packets, is done in a quick manner employing simple addition.
Thus, a method and apparatus for providing differentiated services using a multi-level queuing mechanism has been described. Whereas many alterations and modifications of the present invention will be comprehended by a person skilled in the art after having read the foregoing description, it is to be understood that the particular embodiments shown and described by way of illustration are in no way intended to be considered limiting. References to details of particular embodiments are not intended to limit the scope of the claims.
In the following detailed description numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances well known methods, procedures, components, and circuits have not been described in detail so as not to obscure the present invention.
In alternative embodiments, the present invention may be applicable to implementations of the invention in integrated circuits or chip sets, wireless implementations, switching systems products and trarnsmission systems products. For purposes of this application, the terms switching systems products shall be taken to mean private branch exchanges (PBXs}, central office switching systems that interconnect subscribers, toll/tandem switching systems for interconnecting trunks between switching centers, and broadband core switches found at the center of a service provider's network that may be fed by broadband edge switches or access multiplexors, and associated signaling, and support systems and services. The terra transmission systems products shall be taken to mean products used by service providers to provide interconnection between their subscribers and their networks such as Loop systems, and which provide multiplexing, aggregation and transport between a service provider's switching systems across the wide area, and associated signaling and support systems and services.
Some portions of the detailed descriptions which follow are presented in terms of algorithms and symbolic representations of operation;; on data bits within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self consistent sequence of steps leading to a desired result. The steps are those requiring physical manipulations of physical quantities. Usualhr, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like. It should be borne in mind, however, that aI1 of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the following discussions, it is appreciated that throughout the present invention, discussions utilizing terms such as "processing" or "computing" or "calculating" or "determining" or "displaying" or the like, refer to the action and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices.
The present invention makes use of a multi-level queuing structure to provide differentiated services. Packets are forwarded into a network supporting differentiated services by using the multi-level queue structure. A first queue level is a set of forwarding queues into which packets are placed prier to being forwarded into the network. The amount of time a packet remains in the first queue level is dependent on the total number ofpackets in the first level as well as the rate at which the forwarding device is able to provide packets to the network. Additionally, packets that are to receive a premium service level may also be temporarily placed into a second level queue, referred.
to as a hold queue, prior to being placed in the frst queue level. The present invention ensures that packets are transferred from the hold queue into a first level queue at not higher than a preconfigured rate, and thus that the rate of transfer of premium service level packets to the network does not exceed the preconfigured rate.
Figure 1 illustrates a network environment into which one embodiment of the present invention can be incorporated. A network 110 is illustrated interconnecting multiple devices 12I, 122, 123, 124, and 130. Netwa~rk 110 is intended to represent any of a wide variety of conventional networks, such as the Internet, an intranet, a local area network (LAN), a subset of a larger network (e.g., a portion of the Internet), etc. Devices 121-I24 and 130 are intended to represent any of a w de variety of conventional devices that can receive data from and/or transfer data to network 1 I 0. Examples of such devices include "end points", such as workstations and server~~, or connections to other networks, such as routers, bridges, gateways, etc.
Network 110 is configured to provide differentiated services to data packets.
According to one embodiment of the present invention, the differentiated services include three different levels of service: best effort service, assured service, and premium service.
Best effort service is the lowest level, guaranteeing only that the network will do its best to try and deliver packets of data to their destinations. Currently, many networks (such as the Internet) provide best effort service. Assured service is a higher level of service, ensuring that packets of data are unlikely to be dropped from the network so long as the amount of such data stays within a particular profile (e.g., not larger than a particular size or greater than a particular rate). Premium service i;c the highest level of service. Packets of data at the premium service level are given preferential treatment through the network, and are almost certain not to be dropped from the network so long as the amount of such data stays within a particular profile (e.g., premium packets are not sent at greater than a particular rate).
It is to be appreciated that network I 10 includes multiple components, including any combination of routers, switches, bridges, gateways, workstations, servers, etc. to transfer packets of data. In order to support differentiated services, each of the components within network i I O needs to be able to identify and properly treat packets of different service levels. This can include, for example, the components giving priority to higher service level packets by forwarding them prior to any lower service level packets, or using dedicated paths or channels for higher priority packets. According to one embodiment of the present invention, each of the packets of data being transferred through the network includes a packet header that identifies the service level for the packet. Alternate embodiments can identify service levels in other manners, such as by separate signaling or use of dedicated input and output ports.
According to one embodiment of the present invention, packets of data to be sent using premium and assured service levels are required to conform to particular profiles.
For the assured service level, the particular profile identifies a particular "expected capacity" (e.g., not greater than a particular rate, such as 100k bits per second; or a particular burst rate, such as four packets back-to-back). For the premium service level, the particular prof le identifies a particular packet rate that will not be exceeded (e.g., not greater than twenty packets per minute).
Edge devices, such as, shaping/metering device 130, operate to ensure that such profiles are not violated. By doing so at the edge devices at which data is being input to network 110, each component within network 110 is not required to verify that packets of data conform to the requirements for their identified Ieve1 of service.
Thus, in the illustrated embodiment of Figure l, data is received by shaping/metering device 130 from a source 135. It is to be appreciated that source 135 can be an end point (e.g., a server or workstation), or alternatively a connection to another WO 0013030'1 PCT/US99/2'1381 _g_ network (e.g., a bridge or gateway to a company's LAN). Upon receipt of a packet of data from source 135, device 130 analyzes the packet to ensure that the profiles are nofi being violated and forwards packets to network 110. If one of the profiles is being violated, then packets of data may be dropped or re-labeled to be forwarded to network 110 at a different service level. Packets that are dropped are simply ignored by device 130 and are not forwarded to network 110. The operation of device 130 in ensuring that profiles are not being violated is discussed in more detail below with reference to the following figures.
It should be noted that devices 12I-124 may also provide metering and shaping functionality for packets of data received by the devices from other sources (not shown) to be provided to network 110, as well as for packets of data being forwarded to other destinations by the devices.
Figure 2 is a diagram illustrating a metering and shaping device according to one embodiment of the present invention. In the discussion of Figure 2 and subsequent f gores, reference is made to the forwarding and/or transferring of packets by different logic units of device 130. According to one embodiment of the present invention, device 130 includes a data buffer (not shown) having multiple locations for the storage of data packets. When a data packet is received by device 130, the data is stored in a location of the buffer and remains in that location until the data i~ forwarded out of device 130 to network 110. The discussion of "movement" of such packets within device 130 in this embodiment refers to the transferring or forwarding of packet "pointers" which identify the locations of the buffer. However, according to ali:ernate embodiments of the present invention, the actual data itself may be transferred or forwarded between multiple buffers (not shown) within device 130.
As illustrated in Figure 2, data packets are received by device I30 from source 135. Device 130 is capable of handling multiple "flows". A flow refers to a particular input port to output port mapping of device I30. According to one embodiment, each flow is identified by a set of identifying information in the packet header, including the Internet Protocol (IP) source address, IP destination address, source port, destination port, and protocol identifier. These five pieces of information are also collectively referred to as a "5-tuple". A packet of data received by device I:SO is input to classifier 205.
Classifier 205 compares a set of preprogrammed 5-tuples (which can include "wildcards") to the received 5-tuple for the packet and outputs an identifier of a particular flow for the packet. The number of flows within a device, such as a muter, is typically on the order of 30 to 50. However, it is to be appreciated that fewer or greater flows could be included within a device. The use of classifiers and flows within devices is well known to those skilled in the art and thus will not be discussed further except as it pertains to the present invention.
The identifier output from classifier 205 identifies a particular shaper/policy meter to which the packet is forwarded, as well as ultimately a set of forwarding queues and a forwarding engine. For ease of explanation, only a single shaper/policy meter 2I0 and forwarding engine 230 are illustrated in Figure 2. It is to be appreciated, however, that a different shaper/policy meter and forwarding engine (as well as holding queue and forwarding queues) can be included for each of the different flows supported by device 130. Although, it is also to be appreciated that in embodiments where shaper/policy meter 210 are implemented in software, various procedures and sections of software code can be shared by different flows.
The packet is then provided to policy meter and shaper 2I0. Shaper/meter 210 accesses the service type information in the packet header (typically provided by the original source) to determine whether the packet is to receive a preferred level of service.
Based on the level of service the packet is to receive, as well as when previous packets of data have been forwarded by device 130, shaper/meter 210 either forwards the packet to shaper 214 for temporary storage in hold queue 218 or to forwarding engine 230.
Shaper/meter 210 is responsible for ensuring that the configured policies for different levels of service are maintained. By way of example, if the owner of a particular source has contracted with the owner of device 130 that the source will not attempt to send greater than x number of premium level packets per minute to device 130 fox forwarding to nerivork 110, shaper/meter 210 ensures that this "contracted for" rate is not exceeded. In fact, according to one embodiment of the present invention if the contracted for rate is exceeded then some of the premium level packets may be dropped.
Eventually, packets that are not dropped are forwarded to one of the output queues 232 or 234. The packets are then forwarded on to network 110 from queues 232 and 234 by forwarding engine 230. Packets of data at the premium service level are placed in premium output queue 234, while the packets at the a;~sured or best effort service levels are placed in secondary output queue 232. Forwarding engine 230 controls the w0 00/30307 PCT/US99/27381 forwarding of packets to network 1 I 0, with priority being given to packets in premium output queue 234.
According to one embodiment of the present invention, forwarding a packet of data to one of the output queues 232 or 234 is accomplished by sending the packet to forwarding engine 230. The placement of the packet into the proper one of queues 232 and 234, as well as the proper location within that queue, is performed by forwarding engine 230. Forwarding engine 230 may determine the proper one of queues 232 and 234 based on service level information contained in the header of the data packet, or alternatively from parameters received from shaper/meter 210 when the packet is transferred to forwarding engine 230. According to one embodiment of the present invention, each of queues 232 and 234 is a conventional first-in first-out (FIFO) queue.
Forwarding engine 230 can use any of a wide: variety of conventional arbitration algorithms for selecting packets from queues 232 and 234 for output to network 110.
Examples of such algorithms include giving any pacl~et in premium output queue priority over any packet in secondary output queue 232, and a weighted round robin scheme that favors output queue 234. In the illustrated embodiment, forwarding engine 230 outputs data to network I 10 at its fastest possible: speed, which is dependent on the physical connection between device 130 and networlw; 110.
According to one embodiment of the present invention, classifier 205, shaper/meter 210, and forwarding engine 230 are implemented in software. in this embodiment, software instructions to carry out the processes of classifier 205, shaper/meter 210, and forwarding engine 230 are stored in a system memory (not shown) of device 130 of Figure 2 and executed by a processor (not shown) of device 130.
According to one alternate embodiment of the present invention, classifier 205, shaper/meter 210, and forwarding engine 230 are implemented in firmware, such as a ROM or Flash memory of the device 130. According to another alternate embodiment of the present invention, classifier 205, shaper/meter 210, and forwarding engine 230 are implemented in hardware, such as an application specific integrated circuit (ASIC) of device 130.
In the discussion to follow, reference is made to various times, such as "the current time". In the illustrated embodiment, time is referenced to a clock that "ticks" at a particular "tick time". The tick time is a value (e.g., in microseconds) that is calculated by dividing the minimum data packet size (64 bytes according to one implementation) by the maximum output line rate for device 130 (10 Megabits per second or l00 Megabits per second according to one implementation).
Figure 3 is a flowchart illustrating the steps followed by a policy meter and shaper according to one embodiment of the present invention. Shaper/meter 210 of Figure 2 first receives a packet of data (e.g., from classifier 205), step 302. Upon receipt of the packet, policy meter 212 checks whether the packet is to receive premium service, step 304. As discussed above, the packet header identifies the typo; of service the packet is to receive, so policy meter 212 can make this determination based on the packet header information.
If the packet is to receive premium service, then policy meter 212 checks whether it is permissible to send the packet at the present time, step 306: In the illustrated embodiment, policy meter 212 maintains a record of when it is permissible to send the next packet, referred to as the "okay to send" value or time. The okay to send time is updated each time policy meter 212 forwards a newly received premium service packet to either premium output queue 234 or holding queue 218. In step 306, policy meter 2i2 compares the current time to the okay to send time anal determines that it is permissible to send the packet at the current time if the current time is greater than or equal to the okay to send time.
If it is permissible to send the packet at the ciurrent time, then policy meter 212 marks the packet as "premium service", step 308. Thiis marking is, according to one embodiment of the present invention, a predetermined bit pattern in the packet header that is expected by the components of network 110 to identify premium service packets.
Aitennatively, the type of service information in the header that was originally provided to device 130 from the source 135 could be used by the components of network 110, and the marking step not performed.
Policy meter 212 then updates the okay to send time, step 310. The okay to send time is updated to be the current time plus the amount of time necessary to send the packet from the premium output queue 234 to network 110. It should be noted that this amount of time necessary to send the packet is the amount of time necessary to send the packet to network 110 at the "contracted for" rate, which may be longer than the actual time necessary to send the packet to network 110. By way of example, the connection to network 110 may be fast enough to support 50 packets per second, although the contracted for rate is 10 packets per second. Thus, the amount of time necessary to send the packet would be 0.1 seconds (1 ( 10) rather than 0.02 seconds (1 ( 50).
In the illustrated embodiment, the amount of time necessary to send the packet from the premium output queue 234 to network 110 is determined based on the actual size of the packet of data {e.g., number of bytes), including any header and/or tail information. According to one embodiment of the present invention, a lookup table is used that includes both the packet size and a corresponding amount of time.
This lookup table can be pre-computed, based on the known cloclk rate being used by policy meter 212 as well as the output transfer rate of device I30 and the contracted for rate.
In one implementation, the lookup table includes a different: entry for each possible packet size.
According to an alternate implementation, ranges of packet sizes are used (e.g.; less than I00 bytes, between 100 and 200 bytes, between 200 and 400 bytes, etc.).
After updating the okay to send value, policy meter 212 places the packet into the premium output queue 234, step 312. At this point, forwarding engine 230 is responsible for forwarding the packet to network 110.
Returning to step 306, if it is not permissible t:a send the packet at the current time, policy meter 212 checks whether there is sufficient space in hold queue 218 to store an additional packet of data, step 314. According to one implementation, the determination of whether there is sufficient space in hold queue 218 is made based on a maximum amount of time into the future that a packet will be held by shaper/meter 210 for placing into the forwarding queue. Policy meter 212 compares the current okay to send time minus the current time to the maximum amount of time into the future that a packet will be held for placing into the forwarding queue. If the current okay to send value minus the current time exceeds the maximum amount of time, then there is insufficient space in hold queue 218. Otherwise, there is sufficient space in hold queue 218.
If policy meter 212 determines that there is insufficient space in hold queue 218, then policy meter 2I2 drops the data packet, step 316. In other words, the data packet is ignored by device 130 and is not forwarded to network 110. According to an alternate embodiment, rather than dropping the packet, policy meter 212 re-marks the packet (e.g., as best effort, step 334 discussed below), and forward s the packet to network 110 with the new marking.
However, if policy meter 2I2 determines that there is sufficient space in hold queue 218, then policy meter 212 marks the packet as "premium service", step 318, analogous to step 308 discussed above. Policy meter :? 12 then updates the okay to send time, step 320. The okay to send time is updated to be the current okay to send time plus _13_ the amount of time necessary to send the packet from the premium output queue 234 to network 110. It should be noted that, analogous to the discussion above, this amount of time necessary to send the packet is the amount of time necessary to send the packet to network 110 at the "contracted for" rate, not necessarily the actual time necessary to send the packet to network 110.
The packet is then placed in the hold queue, step 322. Policy meter 212 forwards the packet as well as the current okay to send time to shaper 214, which in turn places the packet into the hold queue.
Figure 4 is a flowchart illustrating the steps followed by shaper 214 in placing a packet into the hold queue according to one embodinnent of the present invention. Shaper 214 f rst checks whether the difference between the okay to send time and the current time is less than a threshold amount, step 402. If the difference is less than the threshold amount, then shaper 214 places the packet in premiu~:n output queue 234, step 312 of Figure 3. In one implementation, this threshold amount is on the order of 10-12 clock ticks. This comparison to a threshold amount is made to prevent shaper 214 from expending the time to place a packet into the hold queue if it is going to be forwarded out of the queue very quickly anyway. If the difference in step 402, however, is not less than the threshold amount, then shaper 2I4 masks the okay to send time by the size of the queue, step 404. This masking generates a pointer into the hold queue based on the okay to send time.
Figure 5 is a block diagram illustrating an example holding queue according to one embodiment of the present invention. As illustrated, holding queue 500 is a "circular" queue having storage capacity for y packets of data. The masking of the okay to send time by the size of the queue causes the pointer to wrap around queue 500.
Returning to Figure 4, shaper 214 f nds the first available location in the holding queue using the pointer (from step 404), step 406. According to one implementation, the first available location is found by first accessing the .queue location identified by the pointer generated in step 404. If that queue location is available (that is, no other packets are currently being stored there), then the finding process is complete.
Otherwise, the pointer is incremented by one and the newly pointed to queue location is checked for availability. If the newly pointed to queue location is available, then the finding process is complete, otherwise the incrementing and checking continues until an available location is found. It is to be appreciated that, because the determination was made in step 314 of Figure 3 that there is sufficient space in the hold queue, eventually an available location will be found in step 406.
Upon finding the first available location within the queue, shaper 214 places the packet into that queue location, step 408.
According to one alternate embodiment of the present invention, shaper 214 does not perform step 402. In this alternate embodiment, shaper 214 always places packets into holding queue 218 without regard for the difference between the okay to send time and the current time. According to another alternate embodiment of the present invention, different processes are used to find an available location in holding queue 218.
By way of example, different starting locations can be used for starting the search for the first available queue location, or different incrementa~tion values can be used.
Returning to Figure 3, once the packet is placed in the hold queue, shaperlrneter 210 waits until it is permissible to send the packet to network 110, step 324.
Once it is okay to send the packet, sender 216 of Figure 2 places the packet into premium output queue 234, step 312.
Figure 6 is a flowchart illustrating the steps followed in forwarding packets from holding queue 218 to premium output queue 234 according to one embodiment of the present invention. An interrupt is generated every tick time, step 602, by a component (not shown) of the system in which the present invention is embodied. In response to the interrupt, sender 2I 6 accesses the current location of :hold queue 218, step 604. Sender 216 maintains a pointer into hold queue 218 that is incremented by one in response to each interrupt, continually wrapping around the queue. Sender 216 checks whether there is a packet in the current location of hold queue 218, :>tep 606. If there is a packet, then sender 216 forwards the packet to premium output queue 234, step 608.
Otherwise, the process simply ends, and then repeats at the next interrupt.
Returning to step 304 of Figure 3, if the packet is not to receive premium service, then policy meter 2i2 checks whether the packet is to receive assured service, step 326.
As discussed above, the packet header identifies the t~rpe of service the packet is to receive, so policy meter 212 can make this determination based on the packet header information. If the packet is to receive assured service, then policy meter 2I2 checks whether the packet size is within a permitted burst size, step 328. The permitted burst size is a preconfigured maximum size or rate which packets at the assured service level are not to exceed. Typical values for the permitted burst size axe on the order of two to four packets. According to one implementation, policy meter 212 maintains a count of packets received at the assured level of service. This count is reset to zero each time a packet of other than the assured service level is received, thereby allowing policy meter 212 to know when the permitted burst size has been exceeded.
If the packet is within the permitted burst size, then policy meter 212 marks the packet as "assured service", step 330. Analogous to the premium service marking discussed above with reference to step 308, the assured service marking is a predetermined bit pattern in the packet header that is expected by the components of network 110 to identify assured service packets. Policy meter 212 then places the packet into the secondary output queue 232, step 332. At this point, forwarding engine 230 is responsible for forwarding the packet to network 110.
Returning to steps 326 and 328, if policy meter 212 determines that the packet is nat to receive assured service (step 326), or that the packet is not within the permitted burst size (step 328), then policy meter 212 marks the packet as "best effort service", step 334. Analogous to the premium service maxking discussed above with reference to step 308, the best effort service marking is a predeterminerd bit pattern in the packet header that is expected by the components of network 110 to identify best effort service packets.
Policy meter 212 then places the packet into the secondary output queue 232, step 332.
At this point, forwarding engine 230 is responsible for forwarding the packet to network 110.
In the discussions above, reference is made to various specific units of measurement (e.g., numbers of bytes). Alternate embodiments of the present invention can be based on different units either smaller or larger than the specific units discussed above (e.g., bits or multiple-byte wards).
Also in the discussions above, reference is made to three levels of service (premium, assured, and best effort). It is to be appreciated that the present invention is not limited to the use of only three levels of service, and that greater or fewer levels could be implemented. By way of example, the assured level of service need not be included, or additional levels of "preferential" service may be added. It is also to be appreciated that additional holding queues and/or forwarding queues can be added to accommodate such additional levels of service.
Also in the discussions above, reference is mare to device 130 of Figure 2 including multiple shaper/meters 210, one for each flow through device 130. In alternate embodiments, a single shaper/meter 210 may be used to control multiple flows through device 130.
Also in the discussions above, reference is made to multiple forwarding queues 232 and 234. Alternatively, a single queue could be used with additional control logic in forwarding engine 230 to distinguish between premium and non-premium packets within the single queue. Similarly, rather than having individual forwarding queues 232 and 234 for each flow, a single queue structure could be used with multiple flows, with additional control logic in forwarding engine 230 to distinguish between the different flows.
Also in the discussions above, reference is made to packets being dropped under certain circumstances. For example, if a packet is to receive premium service and the holding queue is full (as discussed with reference to Figure 3). According to alternate embodiments, such packets may be re-classified (e.g", to best effort service) rather than being dropped. Similarly, reference is made to packets being re-classified in certain under certain circumstances (e.g., a packet to receive assured service is not within a permitted burst size}. According to alternate embodiments, such packets may be dropped rather than re-classified.
Therefore, a method and apparatus for providing differentiated services using a mufti-level queuing mechanism has been described. :Packets of data that are to receive a premium level of service can be temporarily stored in: a holding queue prior to being placed in a forwarding queue and subsequently transferred over the network.
The temporary storage of the packets allows the device to ensure that a particular transfer rate of premium packets is not exceeded. Furthermore, the placement of premium level packets into the holding queue, and thus the determination of when it is okay to send the packets, is done in a quick manner employing simple addition.
Thus, a method and apparatus for providing differentiated services using a multi-level queuing mechanism has been described. Whereas many alterations and modifications of the present invention will be comprehended by a person skilled in the art after having read the foregoing description, it is to be understood that the particular embodiments shown and described by way of illustration are in no way intended to be considered limiting. References to details of particular embodiments are not intended to limit the scope of the claims.
Claims (21)
1. A method of providing differentiated services in a network, the method comprising:
checking whether a packet of data is to receive a preferred level of service;
and if the packet is not to receive the preferred level of service, then placing the packet in a first forwarding queue, otherwise, checking whether it is permissible to forward the packet to a device in the network at the current time, and if it is permissible to forward the packet to the device at the current time, then placing the packet in a second forwarding queue, otherwise temporarily placing the packet in a holding queue before placing the packet in the second forwarding queue.
checking whether a packet of data is to receive a preferred level of service;
and if the packet is not to receive the preferred level of service, then placing the packet in a first forwarding queue, otherwise, checking whether it is permissible to forward the packet to a device in the network at the current time, and if it is permissible to forward the packet to the device at the current time, then placing the packet in a second forwarding queue, otherwise temporarily placing the packet in a holding queue before placing the packet in the second forwarding queue.
2. The method of claim 1, further comprising:
forwarding packets of data to the device in the network from the first forwarding queue and the second forwarding queue, giving priority to packets of data in the second forwarding queue.
forwarding packets of data to the device in the network from the first forwarding queue and the second forwarding queue, giving priority to packets of data in the second forwarding queue.
3. The method of claim 1, further comprising:
checking, prior to placing the packet in the holding queue, whether there is sufficient space in the holding queue for the packet; and placing the packet in the holding queue if there is sufficient space in the holding queue, otherwise dropping the packet.
checking, prior to placing the packet in the holding queue, whether there is sufficient space in the holding queue for the packet; and placing the packet in the holding queue if there is sufficient space in the holding queue, otherwise dropping the packet.
4. The method of claim 1, further comprising:
indexing into the holding queue based on the time at which the indexing occurs;
and placing a packet at the indexed location of the holding queue into the second forwarding queue.
indexing into the holding queue based on the time at which the indexing occurs;
and placing a packet at the indexed location of the holding queue into the second forwarding queue.
5. The method of claim 1, further comprising:
updating an indicator of when it is okay to forward a next packet of data to the device; and using, upon receipt of the next packet of data, the indicator to determine whether it is permissible to forward the next packet of data to the device at the current time.
updating an indicator of when it is okay to forward a next packet of data to the device; and using, upon receipt of the next packet of data, the indicator to determine whether it is permissible to forward the next packet of data to the device at the current time.
6. The method of claim 5, wherein the updating the indicator comprises:
adding together the current time and an amount of time required to send the packet of data to the device.
adding together the current time and an amount of time required to send the packet of data to the device.
7. An apparatus for providing differentiated services in a network, the apparatus comprising:
a first forwarding queue for forwarding packets of data to another device in the network;
a second forwarding queue for forwarding packets of data to another device in the network;
a holding queue to temporarily store packets of data prior to being placed in the second forwarding queue; and a policy meter, coupled to the first and second forwarding queues and to the holding queue, to check whether a packet of data is to receive a preferred level of service, and, if the packet is not to receive the preferred level of service, then to place the packet in the first forwarding queue, otherwise, to check whether it is permissible to forward the packet to the device at the current time, and if it is permissible to forward the packet to the device at the current time, then to place the packet in the second forwarding queue, otherwise to place the packet in the holding queue.
a first forwarding queue for forwarding packets of data to another device in the network;
a second forwarding queue for forwarding packets of data to another device in the network;
a holding queue to temporarily store packets of data prior to being placed in the second forwarding queue; and a policy meter, coupled to the first and second forwarding queues and to the holding queue, to check whether a packet of data is to receive a preferred level of service, and, if the packet is not to receive the preferred level of service, then to place the packet in the first forwarding queue, otherwise, to check whether it is permissible to forward the packet to the device at the current time, and if it is permissible to forward the packet to the device at the current time, then to place the packet in the second forwarding queue, otherwise to place the packet in the holding queue.
8. The apparatus of claim 7, further comprising:
a forwarding engine, coupled to the first and second forwarding queues, to forward packets of data to the device in the network from the first forwarding queue and the second forwarding queue, giving priority to packets of data in the second forwarding queue.
a forwarding engine, coupled to the first and second forwarding queues, to forward packets of data to the device in the network from the first forwarding queue and the second forwarding queue, giving priority to packets of data in the second forwarding queue.
9. The apparatus of claim 7, wherein the policy meter is further to:
check, prior to placing the packet in the holding queue, whether there is sufficient space in the holding queue for the packet; and place the packet in the holding queue if there is sufficient space in the holding queue, and otherwise drop the packet.
check, prior to placing the packet in the holding queue, whether there is sufficient space in the holding queue for the packet; and place the packet in the holding queue if there is sufficient space in the holding queue, and otherwise drop the packet.
10. The apparatus of claim 7, further comprising:
send control logic to index into the holding queue based on the time at which the indexing occurs, and to place a packet at the indexed location of the holding queue into the second forwarding queue.
send control logic to index into the holding queue based on the time at which the indexing occurs, and to place a packet at the indexed location of the holding queue into the second forwarding queue.
11. The apparatus of claim 7, wherein the policy meter is further to:
update an indicator of when it is okay to forward a next packet of data to the device; and use, upon receipt of the next packet of data, the indicator to determine whether it is permissible to forward the next packet of data to the device at the current time.
update an indicator of when it is okay to forward a next packet of data to the device; and use, upon receipt of the next packet of data, the indicator to determine whether it is permissible to forward the next packet of data to the device at the current time.
12. The apparatus of claim 11, wherein the policy meter is to update the indicator by adding together the current time and an amount of time required to send the packet of data to the device.
13. An apparatus for providing differentiated services in a network, the apparatus comprising:
first means for forwarding packets of data to another device in the network;
second means for forwarding packets of data to another device in the network;
means for temporarily storing packets of data prior to being provided to the second means for forwarding; and metering means, coupled to the first and second means and the means for temporarily storing packets, for checking whether a packet of data is to receive a preferred level of service, and, if the packet is not to receive the preferred level of service;
then for providing the packet to the first means for forwarding packets of data, otherwise, for checking whether it is permissible to forward the packet to the device at the current time, and if it is permissible to forward the packet to the device at the current time, then for providing the packet to the second means for forwarding packets of data, otherwise for providing the packet to the means for temporarily storing packets.
first means for forwarding packets of data to another device in the network;
second means for forwarding packets of data to another device in the network;
means for temporarily storing packets of data prior to being provided to the second means for forwarding; and metering means, coupled to the first and second means and the means for temporarily storing packets, for checking whether a packet of data is to receive a preferred level of service, and, if the packet is not to receive the preferred level of service;
then for providing the packet to the first means for forwarding packets of data, otherwise, for checking whether it is permissible to forward the packet to the device at the current time, and if it is permissible to forward the packet to the device at the current time, then for providing the packet to the second means for forwarding packets of data, otherwise for providing the packet to the means for temporarily storing packets.
14. The apparatus of claim 13, further comprising:
means, coupled to the first and second means, for forwarding packets of data to the device in the network from the first and second means, giving priority to packets of data from the second means.
means, coupled to the first and second means, for forwarding packets of data to the device in the network from the first and second means, giving priority to packets of data from the second means.
15. The apparatus of claim 13, further comprising:
sending means for indexing into the means for temporarily storing packets based on the time at which the indexing occurs, and for providing a packet at the indexed location of the means for temporarily storing packets to the second means.
sending means for indexing into the means for temporarily storing packets based on the time at which the indexing occurs, and for providing a packet at the indexed location of the means for temporarily storing packets to the second means.
16. A machine-readable medium having stored thereon a plurality of instructions, designed to be executed by a processor, for providing differentiated services in a network by implementing a function to:
check whether a packet of data is to receive a preferred level of service; and if the packet is not to receive the preferred level of service, then place the packet in a first forwarding queue, otherwise, check whether it is permissible to forward the packet to a device in the network at the current time, and if it is permissible to forward the packet to the device at the current time, then place the packet in a second forwarding queue, otherwise temporarily place the packet in a holding queue before placing the packet in the second forwarding queue.
check whether a packet of data is to receive a preferred level of service; and if the packet is not to receive the preferred level of service, then place the packet in a first forwarding queue, otherwise, check whether it is permissible to forward the packet to a device in the network at the current time, and if it is permissible to forward the packet to the device at the current time, then place the packet in a second forwarding queue, otherwise temporarily place the packet in a holding queue before placing the packet in the second forwarding queue.
17. The machine-readable medium of claim 16, wherein the plurality of instructions are further for implementing a function to forward packets of data to the device in the network from the first forwarding queue and the second forwarding queue, giving priority to packets of data in the second forwarding queue.
18. The machine-readable medium of claim 16, wherein the plurality of instructions are further for implementing a function to:
check, prior to placing the packet in the holding queue, whether there is sufficient space in the holding queue for the packet; and place the packet in the holding queue if there is sufficient space in the holding queue, and otherwise drop the packet.
check, prior to placing the packet in the holding queue, whether there is sufficient space in the holding queue for the packet; and place the packet in the holding queue if there is sufficient space in the holding queue, and otherwise drop the packet.
19. The machine-readable medium of claim 16, wherein the plurality of instructions are further for implementing a function to:
index into the holding queue based on the time at which the indexing occurs;
and place a packet at the indexed location of the holding queue into the second forwarding queue.
index into the holding queue based on the time at which the indexing occurs;
and place a packet at the indexed location of the holding queue into the second forwarding queue.
20. The machine-readable medium of claim 16, wherein the plurality of instructions are further for implementing a function to:
update an indicator of when it is okay to forward a next packet of data to the device; and use, upon receipt of the next packet of data, the indicator to determine whether it is permissible to forward the next packet of data to the device at the current time.
update an indicator of when it is okay to forward a next packet of data to the device; and use, upon receipt of the next packet of data, the indicator to determine whether it is permissible to forward the next packet of data to the device at the current time.
21. The machine-readable medium of claim 20, wherein the plurality of instructions to update the indicator include instructions to add together the current time and an amount of time required to send the packet of data to the device.
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/195,573 | 1998-11-18 | ||
US09/195,573 US6608816B1 (en) | 1998-11-18 | 1998-11-18 | Method and apparatus for providing differentiated services using a multi-level queuing mechanism |
PCT/US1999/027381 WO2000030307A1 (en) | 1998-11-18 | 1999-11-17 | Method and apparatus for providing differentiated services |
Publications (1)
Publication Number | Publication Date |
---|---|
CA2351456A1 true CA2351456A1 (en) | 2000-05-25 |
Family
ID=22721927
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA002351456A Abandoned CA2351456A1 (en) | 1998-11-18 | 1999-11-17 | Method and apparatus for providing differentiated services |
Country Status (5)
Country | Link |
---|---|
US (2) | US6608816B1 (en) |
EP (1) | EP1135897A1 (en) |
AU (1) | AU1525600A (en) |
CA (1) | CA2351456A1 (en) |
WO (1) | WO2000030307A1 (en) |
Families Citing this family (68)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6542508B1 (en) * | 1998-12-17 | 2003-04-01 | Watchguard Technologies, Inc. | Policy engine using stream classifier and policy binding database to associate data packet with appropriate action processor for processing without involvement of a host processor |
US6633540B1 (en) | 1999-07-02 | 2003-10-14 | Nokia Internet Communications, Inc. | Real-time traffic shaper with keep-alive property for best-effort traffic |
US7149222B2 (en) * | 1999-12-21 | 2006-12-12 | Converged Access, Inc. | Integrated access point network device |
DE60003518T2 (en) * | 2000-02-08 | 2004-04-22 | Lucent Technologies Inc. | Guaranteed service type in a package-based system |
SE519229C2 (en) * | 2000-05-25 | 2003-02-04 | Telia Ab | Device and method of service differentiation in a communication network |
US7031267B2 (en) | 2000-12-21 | 2006-04-18 | 802 Systems Llc | PLD-based packet filtering methods with PLD configuration data update of filtering rules |
US7013482B1 (en) * | 2000-07-07 | 2006-03-14 | 802 Systems Llc | Methods for packet filtering including packet invalidation if packet validity determination not timely made |
AU2008203424B2 (en) * | 2000-07-27 | 2010-05-27 | Extreme Networks, Inc. | Voice and data wireless communications network and method |
US6404772B1 (en) * | 2000-07-27 | 2002-06-11 | Symbol Technologies, Inc. | Voice and data wireless communications network and method |
US8250357B2 (en) | 2000-09-13 | 2012-08-21 | Fortinet, Inc. | Tunnel interface for securing traffic over a network |
US7272643B1 (en) | 2000-09-13 | 2007-09-18 | Fortinet, Inc. | System and method for managing and provisioning virtual routers |
US7487232B1 (en) | 2000-09-13 | 2009-02-03 | Fortinet, Inc. | Switch management system and method |
US7574495B1 (en) | 2000-09-13 | 2009-08-11 | Fortinet, Inc. | System and method for managing interworking communications protocols |
US7111072B1 (en) | 2000-09-13 | 2006-09-19 | Cosine Communications, Inc. | Packet routing system and method |
US7457313B2 (en) * | 2000-09-25 | 2008-11-25 | General Instrument Corporation | Hierarchical prioritized round robin (HPRR) scheduling |
US6950393B1 (en) * | 2000-11-22 | 2005-09-27 | P-Cube, Ltd. | Method and apparatus for process flow random early discard in service aware networking systems |
US20020080784A1 (en) * | 2000-12-21 | 2002-06-27 | 802 Systems, Inc. | Methods and systems using PLD-based network communication protocols |
US6914883B2 (en) * | 2000-12-28 | 2005-07-05 | Alcatel | QoS monitoring system and method for a high-speed DiffServ-capable network element |
US7984147B2 (en) | 2000-12-29 | 2011-07-19 | Hewlett-Packard Development Company, L.P. | Apparatus and method for identifying a requested level of service for a transaction |
EP1239635A1 (en) * | 2001-03-07 | 2002-09-11 | BRITISH TELECOMMUNICATIONS public limited company | Method of providing service classes in a packet network |
US6940854B1 (en) * | 2001-03-23 | 2005-09-06 | Advanced Micro Devices, Inc. | Systems and methods for determining priority based on previous priority determinations |
US7065051B2 (en) * | 2001-03-27 | 2006-06-20 | Intel Corporation | Management and scheduling of data that is wirelessly transmitted between a base transceiver station and subscriber units |
EP1251669A1 (en) * | 2001-04-19 | 2002-10-23 | BRITISH TELECOMMUNICATIONS public limited company | Communications network |
US6904057B2 (en) * | 2001-05-04 | 2005-06-07 | Slt Logic Llc | Method and apparatus for providing multi-protocol, multi-stage, real-time frame classification |
US6901052B2 (en) | 2001-05-04 | 2005-05-31 | Slt Logic Llc | System and method for policing multiple data flows and multi-protocol data flows |
US7042848B2 (en) * | 2001-05-04 | 2006-05-09 | Slt Logic Llc | System and method for hierarchical policing of flows and subflows of a data stream |
US6944168B2 (en) * | 2001-05-04 | 2005-09-13 | Slt Logic Llc | System and method for providing transformation of multi-protocol packets in a data stream |
US7020143B2 (en) * | 2001-06-18 | 2006-03-28 | Ericsson Inc. | System for and method of differentiated queuing in a routing system |
US7181547B1 (en) * | 2001-06-28 | 2007-02-20 | Fortinet, Inc. | Identifying nodes in a ring network |
US7299297B2 (en) * | 2001-08-16 | 2007-11-20 | Lucent Technologies Inc. | Method and apparatus for protecting electronic commerce from distributed denial-of-service attacks |
DE50105319D1 (en) | 2001-08-29 | 2005-03-17 | Alcatel Sa | router |
ATE301902T1 (en) | 2001-08-29 | 2005-08-15 | Cit Alcatel | ROUTERS |
EP1335535A1 (en) * | 2002-01-31 | 2003-08-13 | BRITISH TELECOMMUNICATIONS public limited company | Network service selection |
US7254632B2 (en) * | 2002-04-26 | 2007-08-07 | P-Cube Ltd. | Apparatus and method for pattern matching in text based protocol |
US7376125B1 (en) | 2002-06-04 | 2008-05-20 | Fortinet, Inc. | Service processing switch |
US7203192B2 (en) | 2002-06-04 | 2007-04-10 | Fortinet, Inc. | Network packet steering |
US7177311B1 (en) * | 2002-06-04 | 2007-02-13 | Fortinet, Inc. | System and method for routing traffic through a virtual router-based network switch |
US7340535B1 (en) | 2002-06-04 | 2008-03-04 | Fortinet, Inc. | System and method for controlling routing in a virtual router system |
US7161904B2 (en) | 2002-06-04 | 2007-01-09 | Fortinet, Inc. | System and method for hierarchical metering in a virtual router based network switch |
US7283536B2 (en) * | 2002-06-11 | 2007-10-16 | Nokia Corporation | Multimode queuing system for DiffServ routers |
US6968374B2 (en) | 2002-07-03 | 2005-11-22 | Telefonaktiebolaget Lm Ericsson (Publ) | Quality of service (QOS) mechanism in an internet protocol (IP) network |
US7096383B2 (en) | 2002-08-29 | 2006-08-22 | Cosine Communications, Inc. | System and method for virtual router failover in a network routing system |
US7266120B2 (en) | 2002-11-18 | 2007-09-04 | Fortinet, Inc. | System and method for hardware accelerated packet multicast in a virtual routing system |
US7286482B2 (en) * | 2002-11-29 | 2007-10-23 | Alcatel Lucent | Decentralized SLS monitoring in a differentiated service environment |
US6940813B2 (en) * | 2003-02-05 | 2005-09-06 | Nokia Corporation | System and method for facilitating end-to-end quality of service in message transmissions employing message queues |
US7493623B2 (en) * | 2003-02-05 | 2009-02-17 | Nokia Corporation | System and method for identifying applications targeted for message receipt in devices utilizing message queues |
US6970426B1 (en) | 2003-05-14 | 2005-11-29 | Extreme Networks | Rate color marker |
US7720095B2 (en) | 2003-08-27 | 2010-05-18 | Fortinet, Inc. | Heterogeneous media packet bridging |
US9065741B1 (en) * | 2003-09-25 | 2015-06-23 | Cisco Technology, Inc. | Methods and apparatuses for identifying and alleviating internal bottlenecks prior to processing packets in internal feature modules |
US7621162B2 (en) * | 2003-12-30 | 2009-11-24 | Alcatel Lucent | Hierarchical flow-characterizing multiplexor |
US7818236B2 (en) * | 2004-09-13 | 2010-10-19 | Nyfix, Inc. | System for aggregating executions in a communication network for securities transactions and the like |
US7499419B2 (en) | 2004-09-24 | 2009-03-03 | Fortinet, Inc. | Scalable IP-services enabled multicast forwarding with efficient resource utilization |
US7808904B2 (en) | 2004-11-18 | 2010-10-05 | Fortinet, Inc. | Method and apparatus for managing subscriber profiles |
US8036123B1 (en) | 2005-01-07 | 2011-10-11 | Marvell International Ltd. | Integrated circuit for network stress testing |
US7580974B2 (en) * | 2006-02-16 | 2009-08-25 | Fortinet, Inc. | Systems and methods for content type classification |
US7668920B2 (en) * | 2006-03-01 | 2010-02-23 | Fortinet, Inc. | Electronic message and data tracking system |
JP4659657B2 (en) * | 2006-03-28 | 2011-03-30 | 富士通株式会社 | Frame multiplexer |
US8223642B2 (en) * | 2006-04-28 | 2012-07-17 | Tellabs San Jose, Inc. | Differentiated services using weighted quality of service (QoS) |
EP1953959A1 (en) * | 2007-02-01 | 2008-08-06 | British Telecommunications Public Limited Company | Data communication |
US20080318630A1 (en) * | 2007-06-25 | 2008-12-25 | Qualcomm Incorporated | Graceful coexistence for multiple communication protocols |
US7996520B2 (en) | 2007-09-19 | 2011-08-09 | Cisco Technology, Inc. | Behavioral classification of communication sessions using active session initiation |
US8385205B2 (en) * | 2007-09-20 | 2013-02-26 | Tellabs Operations, Inc. | Modeling packet traffic using an inverse leaky bucket |
US8284665B1 (en) * | 2008-01-28 | 2012-10-09 | Juniper Networks, Inc. | Flow-based rate limiting |
JP5630070B2 (en) * | 2010-05-14 | 2014-11-26 | 富士通株式会社 | Relay device, program and method |
US8064354B1 (en) * | 2010-07-21 | 2011-11-22 | Intelepeer, Inc. | Optimized path call routing with device identifier |
US8780896B2 (en) | 2010-12-29 | 2014-07-15 | Juniper Networks, Inc. | Methods and apparatus for validation of equal cost multi path (ECMP) paths in a switch fabric system |
US10169948B2 (en) | 2014-01-31 | 2019-01-01 | International Business Machines Corporation | Prioritizing storage operation requests utilizing data attributes |
EP3107266A1 (en) * | 2015-06-16 | 2016-12-21 | Saguna Networks Ltd. | Methods circuits devices systems and associated machine executable instructions for transporting packetized data across a cellular communications network |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE4434724C1 (en) | 1994-09-28 | 1995-11-23 | Siemens Ag | Information cell handling for asynchronous transfer mode communications device |
US5996019A (en) * | 1995-07-19 | 1999-11-30 | Fujitsu Network Communications, Inc. | Network link access scheduling using a plurality of prioritized lists containing queue identifiers |
US5946297A (en) | 1996-05-31 | 1999-08-31 | International Business Machines Corporation | Scheduling method and apparatus for supporting ATM connections having a guaranteed minimun bandwidth |
US6229812B1 (en) * | 1996-10-28 | 2001-05-08 | Paxonet Communications, Inc. | Scheduling techniques for data cells in a data switch |
US6493347B2 (en) * | 1996-12-16 | 2002-12-10 | Juniper Networks, Inc. | Memory organization in a switching device |
US6137807A (en) * | 1997-12-05 | 2000-10-24 | Whittaker Corporation | Dual bank queue memory and queue control system |
US6205150B1 (en) * | 1998-05-28 | 2001-03-20 | 3Com Corporation | Method of scheduling higher and lower priority data packets |
-
1998
- 1998-11-18 US US09/195,573 patent/US6608816B1/en not_active Expired - Lifetime
-
1999
- 1999-11-17 CA CA002351456A patent/CA2351456A1/en not_active Abandoned
- 1999-11-17 EP EP99957576A patent/EP1135897A1/en not_active Withdrawn
- 1999-11-17 WO PCT/US1999/027381 patent/WO2000030307A1/en not_active Application Discontinuation
- 1999-11-17 AU AU15256/00A patent/AU1525600A/en not_active Abandoned
-
2002
- 2002-09-10 US US10/241,088 patent/US6944128B2/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
US6608816B1 (en) | 2003-08-19 |
EP1135897A1 (en) | 2001-09-26 |
AU1525600A (en) | 2000-06-05 |
WO2000030307A1 (en) | 2000-05-25 |
US6944128B2 (en) | 2005-09-13 |
US20030012144A1 (en) | 2003-01-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6608816B1 (en) | Method and apparatus for providing differentiated services using a multi-level queuing mechanism | |
US7415533B1 (en) | Packet prioritization systems and methods using address aliases | |
US6738378B2 (en) | Method and apparatus for intelligent sorting and process determination of data packets destined to a central processing unit of a router or server on a data packet network | |
US7120149B2 (en) | Methods and system for resequencing out of order data packets | |
US6765905B2 (en) | Method for reducing packet data delay variation in an internet protocol network | |
US7787472B2 (en) | Hierarchical policers for enforcing differentiated traffic behavior | |
US7385993B2 (en) | Queue scheduling mechanism in a data packet transmission system | |
US8576863B2 (en) | Coordinated queuing between upstream and downstream queues in a network device | |
CN111800351B (en) | Congestion notification packet generation by a switch | |
EP1552632A1 (en) | System and method for wireless network channel management | |
US20030169690A1 (en) | System and method for separating communication traffic | |
US7110404B1 (en) | System and method for sending a packet to multiple destinations using a pipeline network processor | |
EP3334101B1 (en) | Load balancing eligible packets in response to a policing drop decision | |
US7016302B1 (en) | Apparatus and method for controlling queuing of data at a node on a network | |
US7382792B2 (en) | Queue scheduling mechanism in a data packet transmission system | |
EP1476994B1 (en) | Multiplexing of managed and unmanaged traffic flows over a multi-star network | |
US7072352B2 (en) | Inverse multiplexing of unmanaged traffic flows over a multi-star network | |
US10715455B2 (en) | Packet switching device modifying paths of flows of packets taken within while outputting packets in received intra-flow order but not necessarily inter-flow order | |
US7864943B2 (en) | Methods and systems for load sharing and preserving sequencing of signaling connection control part (SCCP) messages | |
KR20090060108A (en) | Packet scheduling apparatus and method thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FZDE | Discontinued |