US20130003755A1 - Priority-aware hierarchical communication traffic scheduling - Google Patents
Priority-aware hierarchical communication traffic scheduling Download PDFInfo
- Publication number
- US20130003755A1 US20130003755A1 US13/616,143 US201213616143A US2013003755A1 US 20130003755 A1 US20130003755 A1 US 20130003755A1 US 201213616143 A US201213616143 A US 201213616143A US 2013003755 A1 US2013003755 A1 US 2013003755A1
- Authority
- US
- United States
- Prior art keywords
- scheduler
- communication traffic
- priority
- schedulers
- scheduling
- 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
Images
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
- H04L47/22—Traffic shaping
-
- 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
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/245—Traffic characterised by specific attributes, e.g. priority or QoS using preemption
-
- 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
- H04L47/52—Queue scheduling by attributing bandwidth to queues
- H04L47/527—Quantum based scheduling, e.g. credit or deficit based scheduling or token bank
-
- 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
- H04L47/60—Queue scheduling implementing hierarchical scheduling
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Priority-aware hierarchical communication traffic scheduling methods and apparatus are disclosed. Communication traffic is scheduled by schedulers. Each scheduler supports a multiple priority scheduling function to schedule communication traffic having any of multiple priorities. The communication traffic scheduled by the schedulers is further scheduled based on the priorities of the communication traffic scheduled by each of the schedulers. Traffic shaping may also be integrated into a hierarchical scheduler. The priority-aware further scheduling and the optional shaping may be implemented, for example, using multiple Active lists, one per traffic priority, and a Pending list to which a scheduler can be added if it has communication traffic to schedule but is not allowed to schedule that traffic due to the shaping.
Description
- This application is a continuation of U.S. patent application Ser. No. 11/976,459 filed on Oct. 24, 2007. The entire contents of the related patent application are incorporated herein by reference.
- This invention relates generally to communications and, in particular, to priority-aware scheduling of communication traffic.
- Scheduling is done in multiple stages in a hierarchical scheduler. Leaf-level schedulers in a hierarchical scheduler might include some combination of Strict Priority (SP) and Weighted Round Robin/Deficit Round Robin (WRR/DRR) schedulers. Higher levels of a scheduler might include a combination of optional shapers followed by SP, Time Division Multiplexing (TDM) calendar, or DRR scheduling. None of these algorithms used in a second or higher level of a scheduler hierarchy factors in dynamically changing priority of the traffic scheduled by lower level members.
- Traditional DRR scheduling, for example, employs a single Active-state list that is not priority-aware, and also does not provide for shaping. Although Modified DRR or (MDRR) scheduling might use one or more SP queues and a DRR at the lowest priority level, the combination is scheduled in a fixed priority order. The SP queues and the DRR scheduler have fixed priorities relative to each other, and only that fixed priority order is applied in subsequent scheduling among the queues and the DRR scheduler.
- Thus, there remains a need for improved techniques for priority-aware hierarchical scheduling.
- According to an aspect of the invention, there is provided an apparatus comprising: a plurality of schedulers that schedule communication traffic, each scheduler supporting a multiple priority scheduling function to schedule communication traffic having any of a plurality of priorities; and a priority-aware scheduler, operatively coupled to the plurality of schedulers, that further schedules communication traffic from the plurality of schedulers based on priorities of communication traffic scheduled by each of the plurality of schedulers.
- The priority-aware scheduler may further schedule traffic based on priorities by maintaining a plurality of Active lists respectively associated with the plurality of priorities. Each scheduler of the plurality of schedulers may schedule communication traffic to the Active list associated with the priority of communication traffic it is currently scheduling.
- In some embodiments, the priority-aware scheduler services the plurality of Active lists in priority order according to a DRR scheduling algorithm.
- The apparatus may also include a shaper, operatively coupled between a scheduler of the plurality of schedulers and the priority-aware scheduler, that controls scheduling of communication traffic to the priority-aware scheduler by the scheduler to which the shaper is operatively coupled.
- The shaper may allow the scheduler to which it is operatively coupled to schedule communication traffic to the priority-aware scheduler at up to a maximum rate. The scheduler can be added to a Pending list after the shaper has reached its maximum rate.
- In some embodiments, the shaper manages the maximum rate by issuing tokens to the scheduler to which it is operatively coupled. The scheduler schedules communication traffic to an Active list when it has communication traffic to schedule and it has sufficient shaper tokens to schedule the communication traffic, and is added to the Pending list when it has communication traffic to schedule but it does not have sufficient shaper tokens to schedule the communication traffic.
- The shaper may be one of a plurality of shapers operatively coupled between respective schedulers of the plurality of schedulers and the priority-aware scheduler, with each shaper controlling scheduling of communication traffic to the priority-aware scheduler by the scheduler to which the shaper is operatively coupled. Each shaper may allow the scheduler to which it is operatively coupled to schedule communication traffic to the priority-aware scheduler at up to a respective maximum rate. A scheduler that is operatively coupled to a shaper can be added to a Pending list after its shaper has reached its maximum rate.
- A scheduler that is operatively coupled to a shaper may move from the Pending list to an Active list when it has communication traffic to schedule and it is issued sufficient tokens to schedule the communication traffic by its shaper.
- The plurality of shapers may include shapers having different respective maximum rates.
- A method is also provided, and includes: scheduling communication traffic in a plurality of schedulers, each scheduler supporting a multiple priority scheduling function to schedule communication traffic having any of a plurality of priorities; and further scheduling the communication traffic scheduled in the plurality of schedulers based on priorities of communication traffic scheduled by each of the plurality of schedulers.
- The further scheduling may involve maintaining a plurality of Active lists respectively associated with the plurality of priorities, in which case the scheduling in the plurality of schedulers involves each scheduler scheduling communication traffic to the Active list associated with the priority of communication traffic it is currently scheduling.
- In some embodiments, the further scheduling involves servicing the plurality of Active lists in priority order according to a DRR scheduling algorithm.
- The method may also include shaping the communication traffic that is scheduled by a scheduler of the plurality of schedulers prior to the further scheduling. Shaping may involve allowing communication traffic to be scheduled by the scheduler at up to a maximum rate, and maintaining a Pending list to which the scheduler is added after the maximum rate is reached.
- The allowing may involve managing the maximum rate by issuing tokens to the scheduler, in which case the scheduling by the scheduler may involve scheduling communication traffic to an Active list when it has communication traffic to schedule and it has sufficient shaper tokens to schedule the communication traffic. The scheduler is added to the Pending list when it has communication traffic to schedule but it does not have sufficient shaper tokens to schedule the communication traffic.
- In some embodiments, scheduling involves moving a scheduler from the Pending list to an Active list when it has communication traffic to schedule and it is issued sufficient shaper tokens to schedule the communication traffic.
- Different respective maximum rates may be used for the shaping for different schedulers.
- According to another aspect of the invention, there is provided an apparatus comprising: an interface that enables the apparatus to receive communication traffic from a plurality of schedulers that support a multiple priority scheduling function to schedule communication traffic having any of a plurality of priorities; and a priority-aware scheduler, operatively coupled to the interface, that further schedules communication traffic scheduled by the plurality of schedulers based on priorities of communication traffic scheduled by each of the plurality of schedulers.
- The priority-aware scheduler may further schedule communication traffic by maintaining a plurality of Active lists, respectively associated with a plurality of communication traffic priorities, to which the plurality of schedulers can schedule communication traffic.
- In some embodiments, the priority-aware scheduler services the plurality of Active lists in priority order according to a DRR scheduling algorithm.
- Other aspects and features of embodiments of the present invention will become apparent to those ordinarily skilled in the art upon review of the following description.
- Examples of embodiments of the invention will now be described in greater detail with reference to the accompanying drawings.
-
FIG. 1 is a block diagram of a DRR architecture. -
FIG. 2 is a block diagram of an MDRR architecture. -
FIG. 3 is a block diagram of a hierarchical scheduler according to an embodiment of the invention. -
FIG. 4 is a flow diagram illustrating a method according to an embodiment of the invention. -
FIG. 1 is a block diagram of a DRR architecture. TheDRR architecture 10 shown inFIG. 1 includestraffic queues DRR scheduler 18. Thequeues DRR scheduler 18 scheduling up to a quantum of traffic from each queue in each round. Any remainder of the allowed quantum for a queue after the queue is serviced in a round, also referred to as a “deficit”, is carried forward to the next round. Further details of DRR scheduling may be found, for example, in: M. Shreedhar and George Varghese, “Efficient Fair Queuing using Deficit Round Robin”, Proc. of ACM SIGCOMM '95, August 1995. -
FIG. 2 is a block diagram of an MDRR architecture. Thearchitecture 20 includes atraffic queue 22 from which traffic is directly scheduled by theSP scheduler 36, and threetraffic queues DRR scheduler 34 and then by theSP scheduler 36. This represents a traditional form of MDRR, which involves SP scheduling among a strict priority queue and the output of a DRR scheduler. When a scheduling request is received by an MDRR scheduler, theSP queue 22 is scheduled. If theSP queue 22 is empty, then thequeue DRR scheduler 34 is scheduled. - An overview of other example implementations of MDRR can be found, for example, in: Luciano Lenzini, Enzo Mingozzi, and Giovanni Stea, “Bandwidth and Latency Analysis of Modified Deficit Round Robin Scheduling Algorithms”, Valuetools '06, Pisa, Italy, 11-13 Oct. 2006.
- According to an aspect of the present invention, a modified version of a DRR algorithm is used in hierarchical scheduling. Jitter of high priority traffic in a multi-level DRR instance can thereby be reduced.
- As described in further detail herein, embodiments of the present invention may provide for priority-aware scheduling of communication traffic among lower-level schedulers, taking into account the priorities of traffic that is currently scheduled by those lower-level schedulers. Where each scheduler supports a multiple priority scheduling function to schedule traffic having any of a number of different priorities, the priority of traffic being scheduled by a lower-level scheduler at any time may dynamically change. A higher-level scheduler implementing an embodiment of the invention schedules traffic from its lower-level schedulers in accordance with such dynamic priorities.
- Consider a case where a leaf-level scheduler is an MDRR and the next higher-level scheduler schedules across multiple member MDRRs. In this example, one of the member MDRRs (MDRR1) can schedule
priority 1 traffic and at the same time another member MDRR (MDRR3) can schedule priority 3 traffic. However, if a standard scheduler is employed at the next higher level to schedule among multiple of those leaf-level MDRRs, it will not pay attention to the priority levels of the traffic scheduled by individual members. - For instance, an SP scheduler will schedule the member MDRRs in a fixed and static priority order that is configured and not changed, a TDM calendar will not pay attention to priority, and a DRR will schedule from a single Active list based only on deficits and packet lengths, without differentiating the two member MDRRs based on priority. Because such standard higher-level scheduler algorithms are not aware of dynamic changes in priority, any of those may introduce jitter, especially when the number of members at the leaf level is high (>32).
- An algorithm according to an aspect of the present invention, however, will pay attention to dynamic changes in priority of the traffic scheduled by its members, and puts them in an appropriate one of a number of lists in some embodiments. As a result, jitter may be reduced at all priority levels. In the above example, MDRR1 might be in
Active list # 1, with MDRR3 being in Active list #3. Assuming thatActive list # 1 is associated with higher priority traffic than Active list #3,Active list # 1 will be serviced first, followed by Active list #3. Since the decision dynamically depends on the priority level of traffic selected by a leaf member, MDRR1 may move to Active list #2 if the next packet it schedules has priority 2, for example. - Some embodiments also integrate one or more shapers into a hierarchical scheduler.
-
FIG. 3 is a block diagram of a hierarchical scheduler according to an embodiment of the invention. Thehierarchical scheduler 40 includesmember schedulers queues 42/44/46, 52/54/56,optional shapers schedulers aware scheduler 62, and amemory 68 and acontroller 69 that are operatively coupled to each other and to all of the other scheduler components. The numeral 60 inFIG. 3 is intended to generally denote an interface that enables the priority-aware scheduler 62 to further schedule traffic that has already been scheduled by theschedulers - It should be appreciated that the
scheduler 40 ofFIG. 3 is intended solely for illustrative purposes, and that the present invention is in no way limited to the particular example embodiment explicitly shown in the drawing and described herein. For example, a hierarchical scheduler could include more than the twomember schedulers FIG. 3 . Also, as noted above, theshapers member schedulers - The memory-based implementation shown in
FIG. 3 is also intended to be illustrative. Each of theschedulers - More generally, other embodiments may include further, fewer, and/or different components interconnected in a similar or different manner than shown.
- Those skilled in the art will appreciate that the
queues - The
memory 68 may also include one or more memory devices of any of various types. As discussed in further detail below, thememory 68 may be used to store Active lists and a Pending list that are used in some embodiments. - Software, hardware, firmware, or combinations thereof may be used to implement the
schedulers shapers aware scheduler 62, and thecontroller 69. Any or all of microprocessors for executing software stored on a computer-readable medium, such as in thememory 68, Application Specific Integrated Circuits (ASICs), Programmable Logic Devices (PLDs), and Field Programmable Gate Arrays (FPGAs) are examples of devices that may be suitable for implementing these components. Given such a broad range of possible implementations, these components are described below primarily in terms of their functions. Based on these functional descriptions, a person skilled in the art would be enabled to implement embodiments of the invention in any of various ways. - Considering the actual interconnections between the components shown in
FIG. 3 , these may, at least to some extent, be implementation-dependent. Physical connections such as traces on an electronic circuit card, other types of wired connections, and wireless connections, as well as logical connections between software components through shared memory areas or registers, represent examples of operative couplings that may be provided in embodiments of the invention. - In operation, the
schedulers queues 42/44/46, 52/54/56, respectively. Eachscheduler member schedulers - The priority-
aware scheduler 62 is operatively coupled to the plurality of schedulers, through theshapers schedulers - In one embodiment, the priority-
aware scheduler 62 schedules traffic by maintaining multiple Active lists, illustratively as lists in thememory 68. The Active lists are respectively associated with different traffic priorities. Eachmember scheduler - The effect of an Active list itself could potentially be implemented in a number of ways. For example, an Active list might include an identifier of each
member scheduler particular queue 42/44/46, 52/54/56 from which traffic has been scheduled by themember schedulers - The present invention is in no way limited to any particular scheme for managing Active lists, and references herein to scheduling traffic to an Active list is intended to convey all of the above options, as well as others that may be or become apparent to those skilled in the art.
- It should also be noted that a
member scheduler aware scheduler 62 as well, need not necessarily output any traffic by themselves; they need not even hold any actual traffic or packet data. A scheduler might be considered a sort of manager that decides which queue's traffic is to either proceed to next level scheduling or be output. A scheduler itself does not need to take temporary “ownership” of the traffic it schedules, even though it may have integrated intelligence and memory to handle packet data. - According to one embodiment, the priority-
aware scheduler 62 services the Active lists in priority order according to a DRR scheduling algorithm, although other algorithms could potentially be used. - In the example
hierarchical scheduler 40, theshapers schedulers aware scheduler 62. - The
shapers member scheduler aware scheduler 62 pass or denying it. It is not outputting any traffic. Eachshaper scheduler aware scheduler 62. Theshapers aware scheduler 62 by themember schedulers shapers - More generally, a set of shapers may include shapers that have the same rate and/or shapers that have different rates. To be clear, the same hierarchical scheduler might include some shapers that have the same rate and others that have different rates, or all of the shapers may have different rates. Any combination of common and different rates may be provided, for example, by configuring the shapers with desired shaping rates or patterns.
- As noted above, the priority-
aware scheduler 62 may manage priority-based scheduling using Active lists, with one Active list per traffic priority level. A Pending list may also be maintained by the priority-aware scheduler 62 to enable theshapers - A
scheduler shaper - The Pending list could instead be a Pending state, with 1 bit per
scheduler - As noted above, Active lists may be implemented as link lists with queues/schedulers as members. In one embodiment, the members are serviced in order, head-to-tail. A member might stay at the head of the list as long as it has a deficit as defined in a DRR algorithm used by the priority-
aware scheduler 62. Pending list functionality could effectively be incorporated into the Active lists and a scheduler state. A member in an Active list might have a Pending state bit=0, whereas the same member will have Pending state bit=1 when it runs out of shaper tokens. Hence, in an implementation where there is no Pending list in the form of an actual list but there is Pending list functionality supported by a Pending state bit, that bit could be part of every scheduler state, with the value of the bit determining whether the scheduler is in an active or a pending state. - Another option for the Pending list would be to store entries in the form of {q,p}48, {q,p}58, where {q,p}m is a tuple identifying the winner queue (q) of scheduler m with priority p. The entry for scheduler m can be updated every time a queue q with priority p higher than {q,p}m comes in, even when the scheduler m is waiting in the Pending list. This mechanism could be used to ensure that higher priority traffic is scheduled before lower priority traffic when shaper tokens are replenished, for example.
- A similar mechanism could be used to allow lower-level schedulers to move between Active lists before currently scheduled communication traffic has been scheduled from an Active list by a higher-level priority-aware scheduler. For example, if the priority p of communication traffic changes for a scheduler that is in Active list m which is not for p, the scheduler can be removed from the Active list m and added to the tail of a different Active list n assigned for priority p.
- Shapers such as 64, 66 may use tokens to manage their maximum rates. Shaper tokens representing a total amount of traffic that a shaper can permit to be scheduled within a certain period of time might be refreshed periodically. A
shaper scheduler member schedulers shapers scheduler 48 is able to schedule a packet of traffic to an Active list when the length of the packet is less than or equal to the total number of bytes corresponding to the number of tokens it has been issued by itsshaper 64. Ascheduler - When tokens for a
shaper scheduler aware scheduler 62. In this case, ascheduler shaper member scheduler - The mechanism by which a
scheduler shaper scheduler scheduler shaper scheduler - In some embodiments, the
controller 69 controls shaper alerts, updates, movement of schedulers in and out of the Active and Pending lists, and possibly other functions. Thecontroller 69, although shown inFIG. 3 as a single component, can be broken down into a group of smaller and distributed controllers to control any or all of these functions in a distributed but co-ordinated fashion. - As shown near the
interface 60, the priority-aware scheduler 62 may receive from theschedulers shapers aware scheduler 62 can then further schedule the traffic based on priorities, which as noted above may dynamically change, in that the lower-level schedulers - An output of the priority-
aware scheduler 62 is similarly shown as an identifier of a queue. In other embodiments, actual scheduled traffic, in the form of packets or other traffic blocks, could be transferred between theschedulers aware scheduler 62, and/or output by the priority-aware scheduler. - In schedulers with more levels of hierarchy, multiple instances of the priority-
aware scheduler 62, or even an entirehierarchical scheduler 40, may feed to the next level, which can similarly include a priority-aware scheduler. In that case, the priority-aware scheduler 62, like itsmember schedulers aware scheduler 62. Traffic priority p may also be provided as an output if there is a next level of scheduling that has a priority-aware scheduler. Otherwise, if the level after the priority-aware scheduler 62 is not priority aware, the priority-aware scheduler 62 may simply output q, as shown. - Thus, as will be apparent from the foregoing, each
member scheduler aware scheduler 62 effectively aggregates the decisions of the member schedulers. The outputs of the member schedulers are optionally shaped. - If the higher-level scheduler were a standard DRR instead of the priority-
aware scheduler 62, it would select traffic from one of themember schedulers - On the other hand, a priority-
aware scheduler 62 might use multiple priority-based Active lists instead of just one Active list as in standard DRR. Amember scheduler level scheduler 62 can move to another Active list if traffic having a different priority is available in any its queues. Amember scheduler - Embodiments of the invention have been described above primarily in the context of an apparatus.
FIG. 4 is a flow diagram illustrating a method according to an embodiment of the invention. Themethod 70 includes lower-level scheduling of communication traffic at 72. Each lower-level scheduler supports a multiple priority scheduling function to schedule communication traffic having any of a number of priorities. Traffic scheduled by the lower-level schedulers is optionally shaped at 74, and further scheduling is performed at 76 based on the priorities of the traffic scheduled by the lower-level schedulers. - The
method 70 represents an illustrative embodiment of the invention. Other embodiments may include further, fewer, or different operations performed in a similar or different order. The operations shown inFIG. 4 may also be performed in various ways. For example, a Pending list and/or shaper tokens may be used in shaping traffic at 74, and Active lists may be used for higher-level scheduling and shaping in some embodiments, as described above with reference toFIG. 3 . - The
method 70 may be recursive in some embodiments, with the operation at 76 effectively being replaced by the operations a 72, 74, and 76. A recursive method is used where there are more levels of priority-aware scheduling. For example, with reference again toFIG. 3 for clarity, Voice, Video, Data queues of two DSL subscribers might be at the leaf level similar to 42/44/46, 52/54/56. Those queues will be managed by the subscriberlevel MDRR schedulers subscriber MDRR schedulers aware scheduler 62. Several of these neighborhood concentrators (aggregators) may be aggregated again at the Central Office in a single piece of equipment, and that level corresponds to another priority-aware scheduler (not shown inFIG. 3 ), after the priority-aware scheduler 62. The Central Office equipment may also perform other functions such as Traffic Management for all levels, centrally, before sending the traffic downstream through the neighborhood concentrator. - Further variations may be or become apparent to those skilled in the art.
- From the foregoing, it will be appreciated that embodiments of the invention provide a scheduler algorithm suitable for second or higher level schedulers. Jitter of high priority traffic can be reduced in some embodiments by using a modified version of standard DRR. Modifications may include any or all of: multiple Active lists, including one list per priority level; a new concept of a Pending list to integrate an optional shaper into the algorithm for shaping the output of a lower-level member scheduler; and a new DRR-like algorithm to move a lower-level member scheduler from the Pending list to an Active list, and from one of the Active lists to another.
- The higher-level scheduling algorithm includes the following behaviours in some embodiments: a scheduler that has no traffic to schedule is in none of the lists; a scheduler that has traffic scheduled at a certain priority level (and also has sufficient shaper tokens in some embodiments) is a member of the Active list at that level; a scheduler that has traffic scheduled and its shaper enabled but does not have sufficient tokens, is in the Pending list; a scheduler moves from the Pending list to the Active list when sufficient shaper tokens become available; and a scheduler moves from one Active list to another Active list when it has traffic scheduled and sufficient shaper tokens but its scheduled traffic has changed priority.
- What has been described is merely illustrative of the application of principles of embodiments of the invention. Other arrangements and methods can be implemented by those skilled in the art without departing from the scope of the present invention.
- For example, embodiments of the invention may be applied to any of various technologies, including but in no way limited to Gigabit-capable Passive Optical Network (GPON) products, WiMax Base Stations, WiFi Local Area Network (LAN) controllers, Ethernet switches and routers, etc.
- In addition, although described primarily in the context of methods and systems, other implementations of the invention are also contemplated, as instructions stored on a computer-readable medium for instance.
Claims (23)
1. An apparatus comprising:
a plurality of schedulers that schedule communication traffic, each scheduler supporting a multiple priority scheduling function to schedule communication traffic having any of a plurality of priorities; and
a priority-aware scheduler, operatively coupled to the plurality of schedulers, that further schedules communication traffic from the plurality of schedulers based on priorities of communication traffic scheduled by each of the plurality of schedulers.
2. The apparatus of claim 1 , wherein the priority-aware scheduler further schedules traffic based on priorities by maintaining a plurality of Active lists respectively associated with the plurality of priorities, and wherein each scheduler of the plurality of schedulers schedules communication traffic to the Active list associated with the priority of communication traffic it is currently scheduling.
3. The apparatus of claim 1 , wherein the priority-aware scheduler services the plurality of Active lists in priority order according to a Deficit Round Robin (DRR) scheduling algorithm.
4. The apparatus of claim 1 , further comprising:
a shaper, operatively coupled between a scheduler of the plurality of schedulers and the priority-aware scheduler, that controls scheduling of communication traffic to the priority-aware scheduler by the scheduler to which the shaper is operatively coupled.
5. The apparatus of claim 4 , wherein the shaper allows the scheduler to which it is operatively coupled to schedule communication traffic to the priority-aware scheduler at up to a maximum rate, and wherein the scheduler is added to a Pending list after the shaper has reached its maximum rate.
6. The apparatus of claim 2 , further comprising:
a shaper, operatively coupled between a scheduler of the plurality of schedulers and the priority-aware scheduler, that controls scheduling of communication traffic to the priority-aware scheduler by the scheduler to which the shaper is operatively coupled,
wherein the shaper allows the scheduler to which it is operatively coupled to schedule communication traffic to the priority-aware scheduler at up to a maximum rate, and wherein the scheduler is added to a Pending list after the shaper has reached its maximum rate.
7. The apparatus of claim 6 , wherein the shaper manages the maximum rate by issuing tokens to the scheduler to which it is operatively coupled, and wherein the scheduler schedules communication traffic to an Active list when it has communication traffic to schedule and it has sufficient shaper tokens to schedule the communication traffic, and wherein the scheduler is added to the Pending list when it has communication traffic to schedule but it does not have sufficient shaper tokens to schedule the communication traffic.
8. The apparatus of claim 6 ,
wherein the shaper comprises one of a plurality of shapers operatively coupled between respective schedulers of the plurality of schedulers and the priority-aware scheduler, each shaper controlling scheduling of communication traffic to the priority-aware scheduler by the scheduler to which the shaper is operatively coupled,
wherein each shaper allows the scheduler to which it is operatively coupled to schedule communication traffic to the priority-aware scheduler at up to a respective maximum rate, and
wherein each scheduler that is operatively coupled to a shaper is added to a Pending list after its shaper has reached its maximum rate.
9. The apparatus of claim 8 , wherein a scheduler that is operatively coupled to a shaper moves from the Pending list to an Active list when it has communication traffic to schedule and it is issued sufficient tokens to schedule the communication traffic by its shaper.
10. The apparatus of claim 8 , wherein the plurality of shapers comprises shapers having different respective maximum rates.
11. A method comprising:
scheduling communication traffic in a plurality of schedulers, each scheduler supporting a multiple priority scheduling function to schedule communication traffic having any of a plurality of priorities; and
further scheduling the communication traffic scheduled in the plurality of schedulers based on priorities of communication traffic scheduled by each of the plurality of schedulers.
12. The method of claim 11 , wherein the further scheduling comprises maintaining a plurality of Active lists respectively associated with the plurality of priorities, and wherein the scheduling in the plurality of schedulers comprises each scheduler scheduling communication traffic to the Active list associated with the priority of communication traffic it is currently scheduling.
13. The method of claim 11 , wherein the further scheduling comprises servicing the plurality of Active lists in priority order according to a Deficit Round Robin (DRR) scheduling algorithm.
14. The method of claim 11 , further comprising:
shaping the communication traffic that is scheduled by a scheduler of the plurality of schedulers prior to the further scheduling.
15. The method of claim 14 , wherein the shaping comprises:
allowing communication traffic to be scheduled by the scheduler at up to a maximum rate; and
maintaining a Pending list to which the scheduler is added after the maximum rate is reached.
16. The method of claim 12 , further comprising:
shaping the communication traffic that is scheduled by a scheduler of the plurality of schedulers prior to the further scheduling by allowing the scheduler to schedule communication traffic at up to a maximum rate, and maintaining a Pending list to which the scheduler is added after the maximum rate is reached.
17. The method of claim 16 ,
wherein the allowing comprises managing the maximum rate by issuing tokens to the scheduler,
wherein the scheduling by the scheduler comprises scheduling communication traffic to an Active list when it has communication traffic to schedule and it has sufficient shaper tokens to schedule the communication traffic, and
wherein the scheduler is added to the Pending list when it has communication traffic to schedule but it does not have sufficient shaper tokens to schedule the communication traffic.
18. The method of claim 16 , wherein the shaping comprises shaping the communication traffic that is scheduled by respective schedulers of the plurality of schedulers prior to the further scheduling by:
allowing communication traffic to be scheduled by the respective schedulers at up to a respective maximum rate; and
maintaining a Pending list to which each scheduler of the respective schedulers is added after the respective maximum rate of the shaping is reached.
19. The method of claim 18 , wherein the scheduling comprises moving a scheduler from the Pending list to an Active list when it has communication traffic to schedule and it is issued sufficient shaper tokens to schedule the communication traffic.
20. The method of claim 18 , wherein the allowing comprises allowing communication traffic to be scheduled by the respective schedulers at up to different respective maximum rates.
21. An apparatus comprising:
an interface that enables the apparatus to receive communication traffic from a plurality of schedulers that support a multiple priority scheduling function to schedule communication traffic having any of a plurality of priorities; and
a priority-aware scheduler, operatively coupled to the interface, that further schedules communication traffic scheduled by the plurality of schedulers based on priorities of communication traffic scheduled by each of the plurality of schedulers.
22. The apparatus of claim 21 , wherein the priority-aware scheduler further schedules communication traffic by maintaining a plurality of Active lists, respectively associated with a plurality of communication traffic priorities, to which the plurality of schedulers can schedule communication traffic.
23. The apparatus of claim 21 , wherein the priority-aware scheduler services the plurality of Active lists in priority order according to a Deficit Round Robin (DRR) scheduling algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/616,143 US20130003755A1 (en) | 2007-10-24 | 2012-09-14 | Priority-aware hierarchical communication traffic scheduling |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US11/976,459 US8339949B2 (en) | 2007-10-24 | 2007-10-24 | Priority-aware hierarchical communication traffic scheduling |
US13/616,143 US20130003755A1 (en) | 2007-10-24 | 2012-09-14 | Priority-aware hierarchical communication traffic scheduling |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/976,459 Continuation US8339949B2 (en) | 2007-10-24 | 2007-10-24 | Priority-aware hierarchical communication traffic scheduling |
Publications (1)
Publication Number | Publication Date |
---|---|
US20130003755A1 true US20130003755A1 (en) | 2013-01-03 |
Family
ID=40579846
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/976,459 Active 2028-06-01 US8339949B2 (en) | 2007-10-24 | 2007-10-24 | Priority-aware hierarchical communication traffic scheduling |
US13/616,143 Abandoned US20130003755A1 (en) | 2007-10-24 | 2012-09-14 | Priority-aware hierarchical communication traffic scheduling |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/976,459 Active 2028-06-01 US8339949B2 (en) | 2007-10-24 | 2007-10-24 | Priority-aware hierarchical communication traffic scheduling |
Country Status (2)
Country | Link |
---|---|
US (2) | US8339949B2 (en) |
WO (1) | WO2009054978A1 (en) |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8339949B2 (en) * | 2007-10-24 | 2012-12-25 | Cortina Systems Inc. | Priority-aware hierarchical communication traffic scheduling |
US20100232400A1 (en) * | 2009-03-11 | 2010-09-16 | Sony Corporation | Virtualizing single radio for multiple wireless interfaces in home mesh network |
US8761174B2 (en) * | 2009-03-11 | 2014-06-24 | Sony Corporation | Quality of service traffic recognition and packet classification home mesh network |
US8861445B2 (en) * | 2009-03-11 | 2014-10-14 | Sony Cororation | Multi-channel single radio communication in home mesh network |
US8194593B2 (en) * | 2009-03-11 | 2012-06-05 | Sony Corporation | Quality of service architecture for home mesh network |
US8780762B2 (en) * | 2009-03-11 | 2014-07-15 | Sony Corporation | Node query in ad hoc home mesh network |
US8437355B1 (en) | 2010-11-04 | 2013-05-07 | Adtran, Inc. | Systems and methods for scheduling business and residential services in optical networks |
US8943236B1 (en) * | 2011-06-21 | 2015-01-27 | Netlogic Microsystems, Inc. | Packet scheduling using a programmable weighted fair queuing scheduler that employs deficit round robin |
CN102594663A (en) * | 2012-02-01 | 2012-07-18 | 中兴通讯股份有限公司 | Queue scheduling method and device |
US10608940B2 (en) | 2014-05-07 | 2020-03-31 | Adtran, Inc. | Systems and methods for allocating network bandwidth across access modules |
US10225195B2 (en) | 2014-05-07 | 2019-03-05 | Adtran, Inc. | Telecommunication systems and methods using dynamic shaping for allocating network bandwidth |
EP3141032B1 (en) * | 2014-05-07 | 2019-03-13 | Adtran, Inc. | Telecommunication systems and methods using dynamic shaping for allocating network bandwidth |
US9647950B2 (en) | 2015-05-11 | 2017-05-09 | Ebay Inc. | System and method of site traffic control |
CN107104907B (en) * | 2016-02-19 | 2020-06-12 | 富士通株式会社 | Transmission control method, device and controller for network service |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050111462A1 (en) * | 2003-11-26 | 2005-05-26 | J. Rodney Walton | Quality of service scheduler for a wireless network |
US20060176883A1 (en) * | 2002-03-21 | 2006-08-10 | Cisco Technology, Inc. | Scheduling items using mini-quantum values |
US20070070895A1 (en) * | 2005-09-26 | 2007-03-29 | Paolo Narvaez | Scaleable channel scheduler system and method |
US8339949B2 (en) * | 2007-10-24 | 2012-12-25 | Cortina Systems Inc. | Priority-aware hierarchical communication traffic scheduling |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7046676B2 (en) * | 2001-11-01 | 2006-05-16 | International Business Machines Corporation | QoS scheduler and method for implementing quality of service with cached status array |
US20060187949A1 (en) | 2005-02-09 | 2006-08-24 | Ganesh Seshan | Queuing and scheduling architecture for a unified access device supporting wired and wireless clients |
-
2007
- 2007-10-24 US US11/976,459 patent/US8339949B2/en active Active
-
2008
- 2008-10-23 WO PCT/US2008/012028 patent/WO2009054978A1/en active Application Filing
-
2012
- 2012-09-14 US US13/616,143 patent/US20130003755A1/en not_active Abandoned
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060176883A1 (en) * | 2002-03-21 | 2006-08-10 | Cisco Technology, Inc. | Scheduling items using mini-quantum values |
US20050111462A1 (en) * | 2003-11-26 | 2005-05-26 | J. Rodney Walton | Quality of service scheduler for a wireless network |
US20070070895A1 (en) * | 2005-09-26 | 2007-03-29 | Paolo Narvaez | Scaleable channel scheduler system and method |
US8339949B2 (en) * | 2007-10-24 | 2012-12-25 | Cortina Systems Inc. | Priority-aware hierarchical communication traffic scheduling |
Also Published As
Publication number | Publication date |
---|---|
US8339949B2 (en) | 2012-12-25 |
WO2009054978A1 (en) | 2009-04-30 |
US20090109846A1 (en) | 2009-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8339949B2 (en) | Priority-aware hierarchical communication traffic scheduling | |
US6914882B2 (en) | Method and apparatus for improved queuing | |
US6122673A (en) | Port scheduler and method for scheduling service providing guarantees, hierarchical rate limiting with/without overbooking capability | |
US8730982B2 (en) | Scheduling of data transmission with minimum and maximum shaping of flows in a network device | |
US7742411B2 (en) | Highly-scalable hardware-based traffic management within a network processor integrated circuit | |
US7474668B2 (en) | Flexible multilevel output traffic control | |
USRE44119E1 (en) | Method and apparatus for packet transmission with configurable adaptive output scheduling | |
US6959002B2 (en) | Traffic manager for network switch port | |
US7986706B2 (en) | Hierarchical pipelined distributed scheduling traffic manager | |
JP3872716B2 (en) | Packet output control device | |
WO2007038431A2 (en) | Scaleable channel scheduler system and method | |
CN101436987A (en) | Method and apparatus for scheduling and configuring QoS service | |
JP2014187421A (en) | Communication device and packet scheduling method | |
CA2462793C (en) | Distributed transmission of traffic streams in communication networks | |
US20050111461A1 (en) | Processor with scheduler architecture supporting multiple distinct scheduling algorithms | |
CN100466593C (en) | Method of implementing integrated queue scheduling for supporting multi service | |
US8879578B2 (en) | Reducing store and forward delay in distributed systems | |
CN104022965B (en) | The method and apparatus that a kind of message goes out team's scheduling | |
US7385987B1 (en) | Scheduling system and method for multi-level class hierarchy | |
US7623456B1 (en) | Apparatus and method for implementing comprehensive QoS independent of the fabric system | |
US20020141416A1 (en) | Method and apparatus for improved scheduling technique | |
US7231425B1 (en) | Rate-based scheduling method and system | |
EP3021540B1 (en) | Scheduler and method for layer-based scheduling queues of data packets | |
CN102769566A (en) | Method and device for configuring multilevel scheduling system, and method and device for changing configuration of multilevel scheduling system | |
US6920111B1 (en) | Multiple update frequencies for counters in a multi-level shaping system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: CORTINA SYSTEMS, INC., CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:SINHA, SANTANU;REEL/FRAME:028962/0473 Effective date: 20071019 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |