US20030072326A1 - Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric - Google Patents

Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric Download PDF

Info

Publication number
US20030072326A1
US20030072326A1 US09928509 US92850901A US2003072326A1 US 20030072326 A1 US20030072326 A1 US 20030072326A1 US 09928509 US09928509 US 09928509 US 92850901 A US92850901 A US 92850901A US 2003072326 A1 US2003072326 A1 US 2003072326A1
Authority
US
Grant status
Application
Patent type
Prior art keywords
plurality
port
associated
link
input
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
Application number
US09928509
Inventor
Mehdi Alasti
Kamran Sayrafian-Pour
Vahid Tabatabaee
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
PTS Corp
Original Assignee
ZAGROS NETWORKS Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • H04L47/24Flow control or congestion control depending on the type of traffic, e.g. priority or quality of service [QoS]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/50Queue scheduling
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/50Queue scheduling
    • H04L47/62General aspects
    • H04L47/622Queue service order
    • H04L47/623Queue service order weighted service order
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/25Routing or path finding through a switch fabric
    • H04L49/253Connections establishment or release between ports
    • H04L49/254Centralized controller, i.e. arbitration or scheduling
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/25Routing or path finding through a switch fabric
    • H04L49/256Routing or path finding in ATM switching fabrics
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/30Peripheral units, e.g. input or output ports
    • H04L49/3045Virtual queuing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q3/00Selecting arrangements
    • H04Q3/42Circuit arrangements for indirect selecting controlled by common circuits, e.g. register controller, marker
    • H04Q3/52Circuit arrangements for indirect selecting controlled by common circuits, e.g. register controller, marker using static devices in switching stages, e.g. electronic switching arrangements

Abstract

Arbitration for a switch fabric (e.g., an input-buffered switch fabric) is performed. For a first port, a link subset from a set of links associated with the first port is determined. Each link from the link subset is associated with its own candidate packet and is associated with its own weight value. A link from the link subset for the first port is selected based on the weight value associated with each link from the link subset for the first port. For a second port, a link subset from a set of links associated with the second port is determined. Each link from the link subset associated with the second port is associated with its own candidate packet and is associated with its own weight value. The determining for the second port is performed in parallel with the determining for the first port. A link from the link subset for the second port is selected based on the weight value associated with each link from the link subset of associated with the second port. The selecting for the second port is performed in parallel with the selecting for the first port.

Description

    CROSS-REFERENCE TO RELATED APPLICATIONS
  • The present invention is related to following applications: “Method and Apparatus for Weighted Arbitration Scheduling Separately at the Input Ports and the Output Ports of a Switch Fabric” [Attorney Docket-ZGRO 002/00US], and “Method and Apparatus for Arbitration Scheduling with a Penalty for a Switch Fabric” [Attorney Docket-ZGRO 003/00US], both of which are incorporated herein by reference.[0001]
  • BACKGROUND OF THE INVENTION
  • The present invention relates generally to telecommunication switches. More specifically, the present invention relates to parallel, weighted arbitration scheduling for a switch fabric (e.g., an input-buffered switch fabric). [0002]
  • Known switch fabrics with crossbar architectures exist where data cells received on the multiple input ports of the switch are sent to the various output ports of the switch. Scheduling techniques ensure that the data cells received from different input ports are not sent to the same output port at the same time. These techniques determine the temporary connections between input ports and output ports, via the switch fabric, for a given time slot. [0003]
  • Scheduling techniques can be evaluated based on a number of performance requirements to a broad range of applications. Such performance requirements can include, for example, operating at a high speed, providing a high throughput (i.e., scheduling the routing of as many data cells as possible for each time slot), guaranteeing quality of service (QoS) for specific users, and being easily implemented in hardware. Known scheduling techniques trade one or more performance areas for other performance areas. [0004]
  • For example, U.S. Pat. No. 5,500,858 to McKeown discloses one known scheduling technique for an input-queued switch. This known scheduling technique uses rotating priority iterative matching to schedule the routing of data across the crossbar of the switch fabric. When the data cells are received at the input ports in a uniform manner (i.e., in a uniform traffic pattern), this known scheduler can produce a high throughput of data cells across the switch fabric. When the data cells are received at the input ports, however, in a non-uniform manner more typical of actual data traffic, the throughput from this known scheduling technique substantially decreases. [0005]
  • Thus, a need exists to provide a scheduling technique that can perform effectively for multiple performance requirements, such as for example, operating at a high speed, providing a high throughput, guaranteeing QoS, and being easily implemented in hardware. [0006]
  • SUMMARY OF THE INVENTION
  • Arbitration for a switch fabric (e.g., an input-buffered switch fabric) is performed. For a first port, a link subset from a set of links associated with the first port is determined. Each link from the link subset is associated with its own candidate packet and is associated with its own weight value. A link from the link subset for the first port is selected based on the weight value associated with each link from the link subset for the first port. For a second port, a link subset from a set of links associated with the second port is determined. Each link from the link subset associated with the second port is associated with its own candidate packet and is associated with its own weight value. The determining for the second port is performed in parallel with the determining for the first port. A link from the link subset for the second port is selected based on the weight value associated with each link from the link subset of associated with the second port. The selecting for the second port is performed in parallel with the selecting for the first port. [0007]
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • FIG. 1 illustrates a system block diagram of a switch, according to an embodiment of the present invention. [0008]
  • FIG. 2 shows a system block diagram of the scheduler shown in FIG. 1 [0009]
  • FIG. 3 shows a flowchart of an arbitration process, according to an embodiment of the present invention. [0010]
  • FIG. 4 shows a system block diagram of a grant arbiter, according to an embodiment of the present invention. [0011]
  • FIG. 5 shows a system block diagram of an accept arbiter, according to an embodiment of the present invention. [0012]
  • FIG. 6 shows elements related to an example of a grant step of arbitration within a switch, according to an embodiment of the present invention. [0013]
  • FIG. 7 shows elements related to an example of an accept step of arbitration based on the example shown in FIG. 6. [0014]
  • FIG. 8 shows a system block diagram of a scheduler, according to another embodiment of the present invention. [0015]
  • FIG. 9 shows an example of a link map between input ports and output ports based on two different arbitration decisions for a given time slot.[0016]
  • DETAILED DESCRIPTION
  • Embodiments of the present invention relate to parallel, weighted arbitration scheduling for a switch fabric. The scheduling can be performed at a set of ports for a switch fabric, for example, at a set of input ports and/or a set of output ports. Each port from the set of ports has its own set of links. On a per port basis, a subset of links from the set of links associated with that port is determined. Each link from the determined subset of links for that port is associated with a candidate packet. Each link from the set of links for that port is associated with a weight value. On a per port basis, a link from the determined subset of links for that port is selected based on the weight value for determined subset of links for that port. [0017]
  • A term “link” can be, for example, a potential path across a crossbar switch within the switch fabric between an input port and an output port. In other words, a given input port can potentially connected to any of many output ports within the crossbar switch. For a given time slot, however, a given input port will typically be connected to at most only one output port via a link. For a different time slot, that given input port can be connected to at most one output port via a different link. Thus, the crossbar switch can have many links (i.e., potential paths) for any given input port and for any given output port, although for a given time slot, only certain of those links will be activated. [0018]
  • A link is associated with a candidate packet when a packet is buffered at the input port for that link (e.g., buffered within a virtual output queue associated with that input port and the destination output port). Note that although the term “candidate packet” is used in reference to data queued at the input port, the other types of data such as cells can be considered. [0019]
  • The term “weight value” can be, for example, a value associated with a link based on a bandwidth-reserved rate assigned for that link. In other words, a bandwidth can be allocated to different links within the switch fabric based on the reserved rates of those links. In such an example, the weight value for each link can be updated in every time slot according to the reserved rate, the last scheduling decision and a penalization for non-backlogged, high weight-value links. [0020]
  • The scheduling techniques described herein can be considered as to three aspects. First, the scheduling techniques (or arbitration techniques) can combine parallel arbitration (among the set of input ports and/or among the set of output ports) with weighted arbitration. In other words, scheduling can be performed among the output ports in parallel and/or among the input ports in parallel while also being based on weight values for the links being considered for scheduling. [0021]
  • Second, the scheduling techniques can consider weighted values of the links separately from the perspective of the input ports and from the perspective of the output ports. Thus, a given link between its associated input port and output port has two different weight values (one from the input port perspective and one from the output port perspective) that are maintained separately by the respective input port and output port. [0022]
  • Third, the scheduling techniques can assess a penalty for non-backlogged links having a relatively high weight value. Thus, for a given port, any associated links without a candidate packet and having a weight value greater than the weight value of the link selected during arbitration can have their respective weight value penalized. [0023]
  • FIG. 1 illustrates a system block diagram of a switch, according to an embodiment of the present invention. Switch fabric [0024] 100 includes crossbar switch 110, input ports 120, output ports 130 and scheduler 140. Crossbar 110 is connected to input ports 120 and output ports 130. Scheduler 140 is coupled to crossbar switch 110, input ports 120 and output ports 130.
  • As shown for the top-most input port [0025] 120 of FIG. 1, each input port 120 has a set of queues 121 into which packets received at the input port are buffered. More specifically, each queue 121 is a virtual output queue (VOQ) uniquely associated with a specific output port 130. Thus, received packets (each designating a particular destination output port) are buffered in the appropriate VOQ for its destination output port.
  • In general, as packets are received at the input ports [0026] 120, they are subsequently routed to the appropriate output port 130 by the crossbar switch 110. Of course, packets received at different input ports 120 and destined for the same output port 130 can experience contention within the crossbar switch 110. Scheduler 140 resolves such contention, as discussed below, based on an arbitration (or scheduling) process.
  • Scheduler [0027] 140 uses a parallel, matching scheme that supports rate provisioning. Using this rate-provisioning scheme, scheduler 140 is capable of supporting quality of service (QoS) in traffic engineering in the network (to which switch 100 is connected; not shown). In addition, scheduler 140 provides a high throughput in the switch fabric.
  • Note that input line cards (coupled to the switch fabric [0028] 100 but not shown in FIG. 1) can perform the scheduling and intra-port rate-provisioning among all flows that are destined to the same output port. The switch fabric 100 can operate on a coarser granularity and can perform inter-port rate provisioning, and can consider the flows that share the same input/output pair as a bundled aggregate flow. In this way, the number of micro flows is seamless to the rate-provisioning scheme used by the switch fabric 100 and its complexity is independent of the number of micro-flows.
  • Generally speaking, scheduler [0029] 140 performs three steps during the arbitration process: generating requests, generating grants and generating accepts. The grant and accept steps are carried out according to the reserve rates of the links associated with the specific input ports 120 and output ports 130. To keep track of the priorities of different links, scheduler 140 assigns a weight value (or credit value), for example, to every link at every port.
  • In other words, a given input port [0030] 120 can be associated with a set of links across crossbar switch 110, whereby the given input port 120 can be connected to a set of output ports 130 (e.g., every output port 130). Similarly, a given output port 130 is associated with a separate set of links across crossbar switch 110, whereby the given output port 130 can be connected to a set of input ports 120 (e.g., every input port 120). Scheduler 140 can be configured so that, for example, a link with a higher weight value has a higher priority. A weight vector can represent the weight values for the set of links associated with a given port. In other words, a given link can have an associated weight value; a set of links for a given port can have an associated weight vector, where the weight vector comprises a set of weight values.
  • The weight vectors can be represented mathematically. More specifically, a weight vector, i.e., [0031] CIi (n)=(CII i(n), . . . , CIN i(n)), can be assigned to input port i, and similarly, a weight vector, i.e., COj (n)=(COI j(n), . . . , CON j(n)), can be assigned to output port j, where n is the time index. The kth entry (i.e., the kth weight value), whereI≦k≦N, of every weight vector corresponds to the kth link of the associated port.
  • The weight values associated with the links are updated by scheduler [0032] 140 according to reserved rates of the links and last scheduling decision. In other words, for each time slot, the weight value associated with every link is increased by the link's reserved rate and decreased when the link is served (i.e., when that link is selected during the arbitration process so that a packet is scheduled for transit via that link). Thus, the weight value of a link indicates how much service is owed to that link. Said another way, the weight value indicates the extent to which a given link is given priority over other links where that priority increases over time until the link is serviced. The reserved rates of the links can be predefined and/or can be adjusted during the operation of the switch.
  • In addition, certain weight values are updated based on a penalty. More specifically, the weight values associated with non-backlogged, high-weight-value links are penalized during a given time slot. In other words, for a given port, any associated links without a candidate packet (buffered at the associated virtual output queue) and having a weight value greater than the weight value of the link selected during the arbitration process have their weight values penalized. The weight values of such links can be, for example, decreased an amount related to the link bandwidth. [0033]
  • The operation of scheduler [0034] 140 can also be represented mathematically. More specifically, consider input port i and output port j, and suppose that CImax j(n) and COmax j(n) are the maximum weights selected in the accept and grant steps, respectively. The reserved rate for link (i,k) is rik and Aik(n) is the serving indicator of that link, i.e., A ik ( n ) = { 1 if ( i , k ) is served 0 otherwise ( 1 )
    Figure US20030072326A1-20030417-M00001
  • For link (i, k) and at input port i, the penalty for a non-backlogged, high-weight-value link, DI[0035] k i(n), is DI k i ( n ) = { 1 if ( i , k ) is non - backlogged and CI k i ( n ) CI max i ( n ) 0 otherwise ( 2 )
    Figure US20030072326A1-20030417-M00002
  • CO[0036] max j(n) is defined for output port j in a similar way. For link (j,k) and at output port j, the penalty for a non-backlogged, high-weight-value link, DOk j(n), is DO k j ( n ) = { 1 if ( j , k ) is non - backlogged and CO k j ( n ) CO max j ( n ) 0 otherwise ( 3 )
    Figure US20030072326A1-20030417-M00003
  • Note that DIs and DO's specify the weight values that are decremented to penalize the corresponding links. Hence, the weight vector updating rule for the k-th element of input port i and output port j are, [0037]
  • CI k i(n+1)=CI k i(n)+r ik(n)−(DI k i(n)+A ik(n))
  • CO k j(n+1)=CO k j(n)+r kj(n)−(DO k j(n)+Akj(n))  (4)
  • Penalizing advantageously limits a non-backlogged link from increasing unboundedly. Without penalization, a weight value for a non-backlogged link could increase unboundedly. Then, when such a link receives a number of packets, the link would distract the service of the other links due to its very high weight value. Moreover, the output pattern of such a scheduler would become very bursty. An alternative approach of reducing the weight value to zero inappropriately introduces a delay on any low-rate links that are non-backlogged most of the time. Thus, the penalizing herein reduces the weight value of a non-backlogged link, for example, by the link's throughput. [0038]
  • In an alternative embodiment, the weight values of the links within a weight vector can be adjusted (either increased or decreased) (separate from the above-described weight vector adjustment). The weight vector can be so adjusted without affecting the overall performance of the scheduler because the rate-provisioning method described herein is based on the relative differences between link weight values, not on their absolute values. [0039]
  • FIG. 2 shows a system block diagram of the scheduler shown in FIG. 1. As shown in FIG. 2, scheduler [0040] 140 includes request generator 210, grant arbiters 220, accept arbiters 230 and decision generator 240. Request generator 210 receives input signals from the input ports 120. Request generator 210 is connected to grant arbiters 220 and accept arbiters 230. A given grant arbiter 220 is connected to each accept arbiter 230. The accept arbiters 230 are connected to decision generator 240. Decision generator 240 provides output signals to crossbar switch 110 and provides feedback signals to grant arbiters 220 and accept arbiters 230.
  • FIG. 3 shows a flowchart of an arbitration process, according to an embodiment of the present invention. At step [0041] 300, packets are received at input ports 120. Input signals are provided to request generator 210 based on the received packets. At step 310, request generator 210 can generate a request for each packet received at an input port 120 based on the received input signals. This request identifies, for example, the source input port 120 and the destination output port 130 for a given packet, and represents a request to transit the crossbar switch 110. Accordingly, the requests generated by request generator 210 are provided to the appropriate grant arbiters 220.
  • At step [0042] 320, grant arbiters 220 determine which links have an associated candidate packet based on the requests received from request generator 210. In other words, request generator 210 generates a request(s) for each link associated with a buffered candidate packet(s). Thus, grant arbiters 220 can determine which links have an associated candidate packet, for example, by identifying for which input port 120 a request has been generated.
  • At step [0043] 330, grant arbiters 220 generate grants based on the requests received from request generator 210. Grant arbiters 220 can be configured on a per output-port basis or on a per input-port basis. In other words, step 320 can be performed on a per output-port basis or on a per input-port basis. For example, where the grants are determined on a per input-port basis the request associated with a particular input port 120 is sent to the corresponding grant arbiter 220. In such a configuration, requests from the first input port 120 are sent to the first grant arbiter 220; requests from the second input port 120 are sent to the second grant arbiter 220; and requests from the nth input port 120 are sent to the nth grant arbiter 220.
  • Alternatively, where grants are determined on a per output-port basis, the request associated with a particular output port [0044] 130 is sent to the corresponding grant arbiter 220. In such a configuration, a request that designates the first destination output port 130 is sent to the first grant arbiter 220; a request that designates the second output port 130 is sent to the second grant arbiter 220; and a request that designates the nth output port 130 is sent to the nth grant arbiter 220.
  • Grant arbiters [0045] 220 send an arbitration signal indicative of a grant to the appropriate accept arbiters 230. More specifically, a given grant arbiter 220 can receive a set of requests (i.e., as few as no requests or as many requests as there are associated links). In the case of a grant arbiter 220 that receives one or more requests, that grant arbiter 220 sends an arbitration signal indicative of a grant to the accept arbiter associated with that grant.
  • At step [0046] 340, accept arbiters 230 generate accepts based on the grants generated by grant arbiters 220. Accept arbiters 230 be configured on either a per input-port basis or a per output-port basis depending on the configuration of the grant arbiters 220. In other words, step 340 can be performed on a per input-port basis or on a per output-port basis. More specifically, if step 330 is performed on a per input-port basis by the grant arbiters 220, then step 340 is performed on a per output-port basis by accept arbiters 230. Similarly, if step 330 is performed on a per output-port basis by grant arbiters 220, then step 340 is performed on a per input-port basis by accept arbiters 230. Once the accepts are generated by accept arbiters 230, arbitration signals indicating the accepts are provided to the decision generator 240.
  • At step [0047] 350, decision generator 240 generates an arbitration decision for a given time slot based on the accepts generated by the accept arbiters 230 and provides a signal indicative of the arbitration results for the given time slot to crossbar switch 110. In addition, the signal indicative of the arbitration results is also sent from decision generator 240 to the grant arbiters 220 and accept arbiters 230 so that the weight values can be updated. The weight values are updated based on which requests were winners in the arbitration process. In addition, certain weight values will be penalized based on this feedback information from decision generator 240. Weight values are penalized for links having a weight value higher than the link selected but not having a candidate packet buffered at their associated virtual output queues. Said another way, in the cases where a link with a higher weight value than the selected link but no buffered candidate packet (awaiting switching across the crossbar switch 110), then that link should be accordingly penalized and its weight value reduced.
  • Note that although the arbitration process has been described in connection with FIG. 2 for a given time slot, arbitration can be performed multiple times iteratively within a given time slot. In such an embodiment, for example, arbitration winners from prior iterations within a given time slot are removed from consideration and additional iterations of arbitration is performed for the arbitration losers to thereby provide more arbitration winners within a given time slot. [0048]
  • FIG. 4 shows a system block diagram of a grant arbiter, according to an embodiment of the present invention. A given grant arbiter [0049] 220 includes selection unit 221, weight-value registers 222, update unit 223 and logic “and” 224. Selection unit 221 receives requests Rlj through RNj from request generator 210 and provides an arbitration signal indicative of a grant, Glj through GNj to an accept arbiter 230. Although a selection unit 221 typically provides a single arbitration signal indicative of a grant, FIG. 4 shows the multiple connections from a selection unit 221 upon which a given arbitration signal, Glj through GNj, can be carried to an accept arbiter 230.
  • The arbitration signal indicative of a grant is also provided to logic “and” [0050] 224 from selection unit 221. Logic “and” 224 also receives a request, Rj, and is coupled to update unit 223. Update unit 223 is also coupled to weight-value registers 222. Weight-value registers are also coupled to selection unit 221 and provide a signal back to update unit 223. Update unit 223 also receives a feedback signal indicative of the arbitration results for which an accept, Aj, was generated.
  • FIG. 5 shows a system block diagram of an accept arbiter, according to an embodiment of the present invention. A given accept arbiter [0051] 230 includes selection unit 231, weight-value registers 232, update unit 233 and logic “and” 234. Selection unit 231 receives a set of arbitration signals each indicative of a grant (i.e., zero or more signals from Gil through GiN) from the corresponding grant arbiters 220 (shown in FIG. 2). Selection unit 231 produces at most one arbitration signal indicative of an accept, Ail through AiN. Selection unit 231 also provides the at most one arbitration signal indicative of an accept to logic “and” 234. Logic “and” 234 also receives a request Ri and produces a signal to update unit 233. Update unit 233 provides a signal to weight-value registers 232.
  • Weight-value registers [0052] 232 provide a signal to selection unit 231 and to update unit 233. In addition, update unit 233 also receives an arbitration signal indicative of an accept, Ai.
  • FIG. 6 shows elements related to an example of the arbitration process within a switch, according to an embodiment of the present invention. FIG. 6 represents the weight values for links across a crossbar switch that connects input ports to output ports. The example of FIG. 6 is based on the grant step of arbitration being performed on a per output-port basis. [0053]
  • As shown in FIG. 6, a given output port [0054] 1 can be connected across the crossbar switch by links 610, 620, 630 and 640 to the various input ports 1, 2, 3 and 4, respectively. As shown in FIG. 6, lines 610, 620, 630 and 640 have weight-values w11=2, w21=3, w31=1 and w41=4, respectively. For the virtual output queues of each input port, the virtual output queues are labeled in FIG. 6 with an index that indicates the combination of an input port and output port.
  • For example, input port [0055] 1 has a virtual output queue labeled Q11 associated with the output port 1. This queue has no buffered candidate packets received at input port 1 and destined for output port 1. Input port 1 also has a series of other virtual output queues associated with the remaining destination output ports, such as for example, Q12 through to QIN. The remaining input ports have similar virtual output queues. For purposes of the illustration in FIG. 6, input ports 2 and 3 both have buffered candidate packets in the associated virtual output queues related to output port 1, i.e., Q21 of input port 2 and Q31 of input port 3. The output ports 1 and 4, however, do not have candidate packets buffered for the destination output port 1; in other words, Q11 and Q41, do not have any buffered candidate packets.
  • Following the example of FIG. 6, the grant step of arbitration is performed by selecting a subset of links for which each has a candidate packet buffered at the associated virtual output queue. As mentioned above, in this example of FIG. 6, only link [0056] 620 and link 630 have an associated candidate packet.
  • Next, a grant is determined for the link having the highest weight value from the selected subset of links. In this example, the link [0057] 620 has the highest weight-value (i.e., w21 equal to 3) which is greater than the weight-value for the link 630 (i.e., w31 equal to 1). Thus, a grant is generated for link 620.
  • Note that although FIG. 6 shows an example of the grant step for output port [0058] 1, the other output ports also perform the grant step in parallel. Thus, just as output port 1 produces a grant for input port 2, the remaining output ports also produce at most one grant for an associated input port (which possibly can also be input port 2, or some other input port).
  • FIG. 7 shows elements related to an example of the accept step of arbitration based on the example shown in FIG. 6. As shown in FIG. 7, the accept step is performed on a per input-port basis; this corresponds to the grant step being performed on a per output-port basis. For purposes of clarity, FIG. 7 shows specific details for only input port [0059] 2 while omitting the similar details for the remaining input ports.
  • In the example shown in FIG. 7, input port [0060] 2 has received a grant for links 710, 720 and 730. The received grant for link 710 corresponds to the grant sent from output port 1 to input port 2 shown in FIG. 6. The received grants for links 720 and 730 (received from output ports 2 and 4, respectively) were generated in parallel with the grant for link 710, although not shown in FIG. 6.
  • During the accept step shown by FIG. 7, input port [0061] 2 will select the link having the highest weight value, which in this case is the link 730. In other words, an accept is generated for the link 730 because its weight value (i.e., w′24 equal to 7) is greater than the weight value of the remaining links 710 and 720 (i.e., w′21 equal to 4 and w′22 equal to 3).
  • Note that the weight values for the links from the perspective of the input ports are different than the weight values for the links from the perspective of the output ports. More particularly, each output port and each input port will maintain its own distinct weight vector for its respective links. Thus, the weight-value for a particular link from the output port may have a different weight-value for that same link from the perspective of the input port. For example, note that link [0062] 620 (shown in FIG. 6) from the perspective of input port 2 has a different weight value (w21 equal to 3) than for the weight value for link 710 (shown in FIG. 7) from the perspective of output port 1 (w′21 equal to 4). In sum, the weight values for a link from the output port perspective can be separate and independent from the weight values for the link from the input port perspective. Following the examples shown in FIGS. 6 and 7, certain weight values are updated based on a penalty. For example, the link between input port 4 and output 1 is penalized. As shown in FIG. 6, the link 620 is selected during the grant step because it has the highest weight value (w21 equal to 3) among the links associated a candidate packet (e.g., links 620 and 630). Of the remaining links for output port 1, links 610 and 640 are not associated with a candidate packet. Of these two links, only link 640 has a weight value (w41 equal to 4) greater than the weight value of the selected link (i.e., w21 equal to 3 for link 620). Thus, the weight value for the link between output port 1 and input port 4 is penalized. The weight value for this link should be penalized from both the perspective of the output port and the input port. Thus, from the perspective of output port 1, the weight value w21, for link 640 is penalized, for example, by reducing it from a value of 4 to 3. In addition, the weight value, w′41, for the link between input port 4 and output 1 from the perspective of input port 4 (not shown in FIGS. 6 and 7) is also reduced, for example, by a penalty of 1.
  • FIG. 8 shows a system block diagram of a scheduler, according to another embodiment of the present invention. As shown in FIG. 8, scheduler [0063] 440 includes request generator 441, first-stage arbiters 442, second-stage arbiters 443, decision generators 444 and 445, and matching combiner 446. Note that FIG. 8 shows the first-stage arbiters and second-stage arbiters at a first time, t1, and at a second time, t2. At the first time, t1, the first-stage arbiters and second-stage arbiters are labeled as 422 and 443, respectively; at the second time, t2, the first-stage arbiters and second-stage arbiters are labeled as 422′ and 443′, respectively. First-stage arbiters 442 and 442′ are physically the same devices; second-stage arbiters 443 and 443′ are physically the same devices. FIG. 8 shows the transmission of arbitration signals from first-stage arbiters 442 and second-stage arbiters 443 (determined during the first time, t1) to second-stage arbiters 443′ and first-stage arbiters 442′, respectively (determined during the second time t2).
  • Scheduler [0064] 440 operates in a manner similar to the scheduler discussed in reference to FIGS. 1 through 7, except that scheduler 440 performs two parallel sets of arbitration. Thus, rather than allowing the arbiters to remain idle during one half of the arbitration process, the arbiters of scheduler 440 operate for a second time during its otherwise idle time within a given time slot (or within a given iteration within the time slot). Consequently, scheduler 440 allows a second arbitration process to be performed in parallel without any additional hardware in the form of additional arbiters; matching combiner 446 is the only additional hardware for this embodiment of a scheduler over the scheduler discussed in reference to FIGS. 1 through 7.
  • In other words, the first-stage arbiters [0065] 442 and second-stage arbiters 443 perform the grant step of arbitration on a per input-port basis and on a per output-port basis, respectively. This grant step of arbitration can be performed during the first time, t1, independently by the first-stage arbiters 442 and second-stage arbiters 443. Then, the first-stage arbiters 442′ and second-stage arbiters 443′ perform the accept step of arbitration on a per output-port basis and on a per input-port basis, respectively, based on the grants generated by the second-stage arbiters 443 and the first-stage arbiters 442, respectively. The accept step can be performed by the first-stage arbiters 442′ and second-stage arbiters 443′ during the second time, t2. Again, note that the first-stage arbiters 442 and 442′ are physically the same devices; second-stage arbiters 443 and 443′ are physically the same devices.
  • The arbitration signals indicative of accepts are provided to decision generators [0066] 444 and 445, which independently generate separate arbitration decisions. These arbitration decisions are then provided to matching combiner 446, which provides an integrated arbitration decision for the associated switch fabric.
  • The matching combiner [0067] 446 can provide an integrated arbitration decision in a number of ways. For example, matching combiner 446 can determine the matching efficiency for each received arbitration decision (from decision generator 444 and from decision generator 445), and then output the arbitration decision having a higher matching efficiency for that time slot. For example, for a given a time slot, the matching combiner 446 might determine that the arbitration decision from decision generator 444 has the higher matching efficiency and select that arbitration decision. Then, for a subsequent time slot, the matching combiner 446 might select the arbitration decision from decision generator 445 if it has the higher matching efficiency. The matching efficiency can be, for example, the percentage of links that are scheduled for a given time slot.
  • Alternatively, matching combiner [0068] 445 can alternate each time slot between the two received arbitration decisions. In such an embodiment, the matching combiner 445 can select the arbitration decision from decision generator 444 at one time slot, then select the arbitration decision from decision generator 445 at the next time slot, and so on.
  • In yet another alternative, matching combiner [0069] 445 can select different portions of the switch fabric and the corresponding optimal portions of the arbitration decisions. In other words, matching combiner 445 can consider different portions of the switch fabric, and then, for each portion, matching combiner 445 can select the arbitration decision from either the decision generator 444 or decision generator 445 that is optimal (or at least not less optimal) for that portion of the switch fabric.
  • FIG. 9 shows an example of a link map between input ports and output ports based on two different arbitration decisions for a given time slot. The example shown in FIG. 9 illustrates different links within the switch fabric and the corresponding arbitration decisions. In FIG. 9, the solid lines between the input ports and the output ports can represent the arbitration decision from decision generator [0070] 444; the dotted lines between input ports and output ports can represent the arbitration decision from decision generator 445.
  • In the example shown in FIG. 9, the switch fabric can be considered in three sets of ports: input ports [0071] 1 through 3 and output ports 1 through 3; input ports 4 through 6 and output ports 4 through 7; and input ports 7 through 8 and output port 8. For the first set of ports, the number of arbitration decisions from decision generator 444 (i.e., the solid lines) exceeds the number of arbitration decisions from decision generator 445 (i.e., the dotted lines). Thus, for the first set of ports, the arbitration decisions from decision generator 444 is optimal. For the second set of ports, the number of arbitration decisions from decision generator 445 (i.e., the dotted lines) exceeds the number of arbitration decisions from decision generator 444 (i.e., the solid lines). Thus, for the second set of ports, the arbitration decisions from decision generator 445 are optimal. For the third set of ports, the number of arbitration decisions from decision generator 444 (i.e., the solid lines) equals the number of arbitration decisions from decision generator 445 (i.e., the dotted lines). Thus, for the third set of ports, the arbitration decisions from either decision generator 444 or 445 are sufficient.
  • Although the present invention has been discussed above in reference to examples of embodiments and processes, other embodiments and/or processes are possible. For example, although various embodiments have been described herein in reference to a switch fabric having an equal number of input ports and output ports, other embodiments are possible where the switch fabric has a number of input ports different from the number output ports. [0072]
  • Note that although examples of embodiments of switch fabric discussed above use the rate-provisioning method on both a per input-port basis and a per output-port basis, other embodiments can use the rate-provisioning method on a per input-port basis only or on a per output-port basis only. In such an embodiment, for example, the rate-provisioning method discussed herein can be used for the output ports while another method (e.g., the iSLIP method disclosed in U.S. Pat. No. 5,500,858, which is incorporated herein for background purposes) can be used for the input ports. Such an embodiment can have, for example, a greater number of input ports (e.g., each having a relatively low throughput) than the number of output ports (e.g., each having a relatively high throughput). [0073]

Claims (47)

    What is claimed is:
  1. 1. A method for arbitrating for a switch fabric having a plurality of ports, each port from the plurality of ports having its own plurality of links, comprising:
    determining, on a per port basis, a subset of links from the plurality of links associated with that port, each link from the determined subset of links for that port being associated with a candidate packet, each link from the plurality of links for that port being associated with a weight value; and
    selecting, on a per port basis, a link from the determined subset of links for that port based on the weight value for determined subset of links for that port.
  2. 2. The method of claim 1, wherein the weight value for the selected link for that port is greater than the weight value associated with each remaining link from the determined subset of links for that port.
  3. 3. The method of claim 1, wherein the weight value associated with each link from the plurality of links associated with that port corresponds to a priority for that port and a second port associated with that link.
  4. 4. The method of claim 1, wherein the candidate packet associated with a link from the plurality of links is buffered at an input port associated with that link.
  5. 5. The method of claim 1, wherein the determining and the selecting are performed on a per output port basis.
  6. 6. The method of claim 5, further comprising:
    receiving, on a per output port basis, a plurality of requests associated with that port, each request from the plurality of requests associated with that port being associated with a link from the plurality of links associated with that port and being associated with a candidate packet; and
    sending, on a per output port basis, a grant associated with the selected link for that output port.
  7. 7. The method of claim 6, further comprising:
    determining, on a per input port basis, a subset of links from a plurality of links associated with that input port based on any grants received at that input port; and
    selecting, on a per input port basis, at most one link from the determined subset of links associated with that input port and having a weight value greater than a weight value associated with each remaining link from the determined subset of links for that input port.
  8. 8. The method of claim 1, wherein the determining and the selecting are performed on a per input port basis.
  9. 9. The method of claim 8, further comprising:
    receiving, on a per input port basis, a plurality of requests associated with that port, each request from the plurality of requests associated with that port being associated with a link from the plurality of links associated with that port and being associated with a candidate packet; and
    sending, on a per input port basis, a grant associated with the selected link for that input port.
  10. 10. The method of claim 9, further comprising:
    determining, on a per output port basis, a subset of links from a plurality of links associated with that output port based on any grants received at that output port; and
    selecting, on a per output port basis, at most one link from the determined subset of links associated with that output port and having a weight value greater than a weight value associated with each remaining link from the determined subset of links for that output port.
  11. 11. A method for arbitrating for a switch fabric, comprising:
    determining, for a first port, a link subset from a plurality of links associated with the first port, each link from the link subset being associated with its own candidate packet and being associated with its own weight value;
    selecting a link from the link subset for the first port based on the weight value associated with each link from the link subset for the first port;
    determining, for a second port, a link subset from a plurality of links associated with the second port, each link from the link subset associated with the second port being associated with its own candidate packet and being associated with its own weight value, the determining for the second port being performed in parallel with the determining for the first port; and
    selecting a link from the link subset for the second port based on the weight value associated with each link from the link subset of associated with the second port, the selecting for the second port being performed in parallel with the selecting for the first port.
  12. 12. The method of claim 11, wherein the weight value associated with each link from the plurality of links associated with the first port corresponds to a priority between the first port and a remaining port associated with that link.
  13. 13. The method of claim 11, further comprising:
    sending, from the first port, a grant associated with link selected for the first port;
    sending, from the second port, a grant associated with the link selected for the second port;
    determining, for a third port, a subset of links from a plurality of links associated with the third port based on any grants received at the third port; and
    selecting, for the third port, at most one link from the determined subset of links associated with the third port based on the weight value associated with each link from the determined subset of links for the third port.
  14. 14. The method of claim 11, wherein the first port is an output port and the second port is an output port.
  15. 15. The method of claim 14, further comprising:
    receiving, at the first port, a plurality of requests associated with the first port, each request from the plurality of requests associated with the first port being associated with a candidate packet;
    sending, from the first port, a grant associated with the selected link for first port;
    receiving, at the second port, a plurality of requests associated with the second port, each request from the plurality of requests associated with the second port being associated with a candidate packet; and
    sending, from the second port, a grant associated with the selected link for second port.
  16. 16. The method of claim 15, further comprising:
    determining, for a first input port, a subset of links from a plurality of links associated with the first input port based on any grants received at the first input port; and
    selecting, for a first input port, at most one link from the determined subset of links associated with the first input port and having a weight value greater than a weight value associated with each remaining link from the determined subset of links for the first input port.
  17. 17. The method of claim 11, wherein the first port is an input port and the second port is an input port.
  18. 18. The method of claim 17, further comprising:
    receiving, at the first port, a plurality of requests associated with the first port, each request from the plurality of requests associated with the first port being associated with a candidate packet;
    sending, from the first port, a grant associated with the selected link for first port;
    receiving, at the second port, a plurality of requests associated with the second port, each request from the plurality of requests associated with the second port being associated with a candidate packet; and
    sending, from the second port, a grant associated with the selected link for second port.
  19. 19. The method of claim 18, further comprising:
    determining, for a first output port, a subset of links from a plurality of links associated with the first output port based on any grants received at the first output port; and
    selecting, for a first output port, at most one link from the determined subset of links associated with the first output port and having a weight value greater than a weight value associated with each remaining link from the determined subset of links for the first output port.
  20. 20. An apparatus, comprising:
    a plurality of input ports;
    a plurality of output ports each being coupled to each input port from the plurality of input ports by a link; and
    a plurality of arbiters each being uniquely associated with its own output port from the plurality of output ports, the plurality of arbiters being configured to operate in parallel with each other, a first arbiter from the plurality of arbiters being associated with a first output port from the plurality of output ports, the first arbiter having a selection unit configured to transmit an arbitration signal based on a weight value associated with each link associated with the first output port and having a candidate packet.
  21. 21. The apparatus of claim 20, further comprising:
    an arbiter associated with a first input port from the plurality of input ports and having a selection unit configured to receive at least one arbitration signal from a respective output port from the plurality of output ports, the selection unit of the arbiter associated with the first input port being configured to transmit an arbitration signal based on a weight value associated with each link associated with the received at least one arbitration signal.
  22. 22. The apparatus of claim 21, further comprising:
    a second plurality of arbiters each being uniquely associated with its own input port from the plurality of input ports, the second plurality of arbiters being configured to operate in parallel with each other.
  23. 23. The apparatus of claim 22, wherein:
    each arbiter from the second plurality of arbiters is configured to receive the arbitration signal from any arbiter from the first plurality of arbiters, an arbiter associated with a first input port from the plurality of input ports being configured to transmit an arbitration signal based on: (1) a weight value associated with each link associated with the first input port and (2) any arbitration signals received from the plurality of arbiters.
  24. 24. An apparatus, comprising:
    a plurality of input ports;
    a plurality of output ports each being coupled to each input port from the plurality of input ports by a link; and
    a plurality of arbiters each being uniquely associated with its own input port from the plurality of input ports, the plurality of arbiters being configured to operate in parallel with each other, a first arbiter from the plurality of arbiters being associated with a first input port from the plurality of input ports, the first arbiter having a selection unit configured to transmit an arbitration signal based on a weight value associated with each link associated with the first input port and having a candidate packet.
  25. 25. The apparatus of claim 24, further comprising:
    an arbiter associated with a first output port from the plurality of output ports and having a selection unit configured to receive at least one arbitration signal from a respective input port from the plurality of input ports, the selection unit of the arbiter associated with the first output port being configured to transmit an arbitration signal based on a weight value associated with each link associated with the received at least one arbitration signal.
  26. 26. The apparatus of claim 24, further comprising:
    a second plurality of arbiters each being uniquely associated with its own output port from the plurality of output ports, the second plurality of arbiters being configured to operate in parallel with each other.
  27. 27. The apparatus of claim 26, wherein:
    each arbiter from the second plurality of arbiters is configured to receive the arbitration signal from any arbiter from the first plurality of arbiters, an arbiter associated with a first output port from the plurality of output ports being configured to transmit an arbitration signal based on: (1) a weight value associated with each link associated with the first output port and (2) any arbitration signals received from the plurality of arbiters.
  28. 28. An apparatus, comprising:
    a first selection unit associated with its own plurality of links, the first selection unit being configured to receive an activate-link signal and being configured to transmit an arbitration signal based on a weight value associated with each link from the plurality of links for the first selection unit and based on the activation-link signal received at the first selection unit;
    a second selection unit associated with its own plurality of links, the second selection unit being configured to receive an activate-link signal and being configured to transmit an arbitration signal based on a weight value associated with each link from the plurality of links for the second selection unit and based on the activation-link signal received at the second selection unit; and
    a third selection unit associated with its own plurality of links, the third selection unit being configured to receive any arbitration signals associated with the plurality of links for the third selection unit, the third selection unit being configured to transmit an arbitration signal based on a weight value associated with each link from the plurality of links for the third selection unit and based on any arbitration signals received at the third selection unit.
  29. 29. The apparatus of claim 28, wherein:
    the first selection unit is associated with a first output port, each link from the plurality of links for the first selection unit couples the first output port to an input port from a plurality of input ports, the arbitration signal for the first selection unit indicates a grant associated with the first output port and with an input port from the plurality of input ports; and
    the second selection unit is associated with a second output port, each link from the plurality of links for the second selection unit couples the second output port to an input port from a plurality of input ports, the arbitration signal for the second selection unit indicates a grant associated with the second output port and with an input port from the plurality of input ports.
  30. 30. The apparatus of claim 29, wherein:
    the third selection unit is associated with a first input port from the plurality of input ports, the arbitration signal to be transmitted from the third selection unit indicates an accept associated with the first input port and any output ports associated with any respective arbitration signal received at the third selection unit.
  31. 31. The apparatus of claim 28, wherein:
    the first selection unit is associated with a first input port, each link from the plurality of links for the first selection unit couples the first input port to an output port from a plurality of output ports, the arbitration signal for the first selection unit indicates a grant associated with the first input port and with an output port from the plurality of output ports; and
    the second selection unit is associated with a second input port, each link from the plurality of links for the second selection unit couples the second port to an output port from a plurality of output ports, the arbitration signal for the second selection unit indicates a grant associated with the second input port and with an output port from the plurality of output ports.
  32. 32. The apparatus of claim 31, further comprising:
    the third selection unit is associated with a first output port from the plurality of output ports, the arbitration signal to be transmitted from the third selection unit indicates an accept associated with the first output port and any input ports associated with any respective arbitration signal received at the third selection unit.
  33. 33. The apparatus of claim 28, wherein:
    the first selection unit is configured to operate in parallel with the second selection unit; and
    the third selection unit is configured to operate in parallel with a fourth selection unit.
  34. 34. A method for arbitrating for a switch, comprising:
    determining, in a first time slot, a plurality of grants associated with a plurality of output arbiters; and
    determining, in the first time slot, a plurality of grants associated with a plurality of input arbiters.
  35. 35. The method of claim 34, further comprising:
    determining, in a second time slot, a plurality of accepts associated with the plurality of input arbiters based on the plurality of grants determined for the plurality of output arbiters in the first time slot; and
    determining, in the second time slot, a plurality of accepts associated with the plurality of output arbiters based on the plurality of grants determined for the plurality of input arbiters in the first time slot.
  36. 36. The method of claim 35, further comprising:
    generating a plurality of arbitration selections based on the plurality of accepts associated with the plurality of input arbiters and based on the plurality of accepts associated with the plurality of output arbiters.
  37. 37. The method of claim 35, further comprising:
    selecting one from the group of (1) the plurality of accepts associated with the plurality of input arbiters, and (2) the plurality of accepts associated with the plurality of output arbiters, based on an optimal throughput associated with that selected group.
  38. 38. The method of claim 35, further comprising:
    selecting a plurality of arbitration selections based on the plurality of accepts associated with the plurality of input arbiters and based on the plurality of accepts associated with the plurality of output arbiters, the plurality of arbitration selections having a first portion and a second portion, the first portion of the plurality of arbitration selections being associated with a portion from the plurality of accepts associated with the plurality of input arbiters, the second portion of the plurality of arbitration selections being associated with a portion from the plurality of accepts associated with the plurality of output arbiters.
  39. 39. A method for arbitrating for a switch, comprising:
    determining a grant for a first output arbiter from a plurality of output arbiters;
    determining a grant for a first input arbiter from a plurality of input arbiters, the determining the grant for the first input arbiter being performed in parallel with the determining the grant for the first output arbiter;
    determining an accept for a second input arbiter from the plurality of input arbiters based on the determined grant for the first output arbiter; and
    determining an accept for a second output arbiter from the plurality of output arbiters based on the determined grant for the first input arbiter, the determining the accept for second output arbiter being performed in parallel with the determining the accept for the second input arbiter.
  40. 40. The method of claim 39, further comprising:
    selecting one from the group of: (1) the accept for the second input arbiter and (2) the accept for the second output arbiter.
  41. 41. The method of claim 39, further comprising:
    selecting, for a first time slot, one from the group of: (1) the accept for the second input arbiter and (2) the accept for the second output arbiter, the determining the accept for the second input arbiter and the accept for the second output arbiter being performed for the first time slot; and
    selecting, for a second time slot, one from the group of: (1) an accept for the second input arbiter for the second time slot and (2) an accept for the second output arbiter for the second time slot.
  42. 42. A method for arbitrating for a switch fabric, comprising:
    determining, for a first output port from a plurality of output ports, a link from a plurality of links associated with the first output port based on a weight value associated with each link from a subset of links from the plurality of links, each link from the subset of links being associated with its own candidate packet; and
    determining, for a first input port from a plurality of input ports, a link from a plurality of links associated with the first input port based on a weight value associated with each link from a subset of links from the plurality of links, each link from the subset of links being associated with its own candidate packet, the determining for the first input port being performed in parallel with the determining for the first output port.
  43. 43. The method of claim 42, further comprising:
    selecting, for a second output port from the plurality of output ports, a link from a plurality of links associated with the second output port including the link determined for the first input port, the selecting for the second output being based on a weight value associated with each link from the plurality of links associated with the second output port; and
    selecting, for a second input port from the plurality of input ports, a link from a plurality of links associated with the second input port including the link determined for the first output port, the selecting for the second input being based on a weight value associated with each link from the plurality of links associated with the second input port, the selecting for the second input port being performed in parallel with the selecting for the second output port.
  44. 44. The method of claim 42, wherein:
    the link selected for the second output port is selected from a subset of links from plurality of links for the second output port, each link from the subset of links associated with the second output port is associated with a determination of a corresponding input port; and
    the link selected for the second input port is selected from a subset of links from plurality of links for the second input port, each link from the subset of links associated with the second input port is associated with a determination of a corresponding output port.
  45. 45. An apparatus, comprising:
    a plurality of first-stage arbiters; and
    a plurality of second-stage arbiters each associated with the plurality of first-stage arbiters, each first-stage arbiter from the plurality of first-stage arbiters being configured to send its own grant signal to a second-stage arbiter from the plurality of second-stage arbiters during a first time slot, each second-stage arbiter from the plurality of second-stage arbiters being configured to send its own grant signal to a first-stage arbiter from the plurality of first-stage arbiters during the first time slot.
  46. 46. The apparatus of claim 46, wherein:
    a first first-stage arbiter from the plurality of first-stage arbiters being configured to receive at least one grant from a second-stage arbiter from the plurality of second-stage arbiters during a second time slot, the first first-stage arbiter being configured to send an accept signal based on any grant received from a second-stage arbiter from the plurality of second-stage arbiters; and
    a first second-stage arbiter from the plurality of second-stage arbiters being configured to receive at least one grant from a first-stage arbiter from the plurality of first-stage arbiters during the second time slot, the first second-stage arbiter being configured to send an accept signal based on any grant received from a first-stage arbiter from the plurality of first-stage arbiters.
  47. 47. The apparatus of claim 46, further comprising:
    a matching combiner coupled to the plurality of first-stage arbiters and the plurality of second-stage arbiters, the matching combiner configured to receive a plurality of accept signals from the plurality first-stage arbiters and to receive a plurality of accept signals from the plurality of second-stage arbiters, the matching combiner configured to send a plurality of arbitration selections based on the plurality of accept signals from the plurality first-stage arbiters and the plurality of accept signals from the plurality of second-stage arbiters.
US09928509 2001-08-14 2001-08-14 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric Abandoned US20030072326A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US09928509 US20030072326A1 (en) 2001-08-14 2001-08-14 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US09928509 US20030072326A1 (en) 2001-08-14 2001-08-14 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric
PCT/US2002/025508 WO2003017593A1 (en) 2001-08-14 2002-08-13 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric
US10747658 US7170903B2 (en) 2001-08-14 2003-12-30 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US10747658 Continuation US7170903B2 (en) 2001-08-14 2003-12-30 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric

Publications (1)

Publication Number Publication Date
US20030072326A1 true true US20030072326A1 (en) 2003-04-17

Family

ID=25456332

Family Applications (2)

Application Number Title Priority Date Filing Date
US09928509 Abandoned US20030072326A1 (en) 2001-08-14 2001-08-14 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric
US10747658 Active US7170903B2 (en) 2001-08-14 2003-12-30 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric

Family Applications After (1)

Application Number Title Priority Date Filing Date
US10747658 Active US7170903B2 (en) 2001-08-14 2003-12-30 Method and apparatus for parallel, weighted arbitration scheduling for a switch fabric

Country Status (2)

Country Link
US (2) US20030072326A1 (en)
WO (1) WO2003017593A1 (en)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030193941A1 (en) * 2002-03-30 2003-10-16 Bonomi Flavio Giovanni Packet scheduling particularly applicable to systems including a non-blocking switching fabric and homogeneous or heterogeneous line card interfaces
US20050226263A1 (en) * 2004-04-12 2005-10-13 Cisco Technology, Inc., A California Corporation Weighted random scheduling particularly applicable to packet switching systems
US7020131B1 (en) * 2001-12-24 2006-03-28 Applied Micro Circuits Corp. System and method for hierarchical switching
US20060165080A1 (en) * 2005-01-24 2006-07-27 International Business Machines Corporation Replicated distributed responseless crossbar switch scheduling
US7158512B1 (en) * 2002-04-01 2007-01-02 P-Cube Ltd. System and method for scheduling a cross-bar
US20080186961A1 (en) * 2001-12-20 2008-08-07 Kenneth Yi Yun System and Method for Reevaluating Granted Arbitrated Bids
US20080253294A1 (en) * 2001-12-14 2008-10-16 Alberto Alessandro Della Ripa Data link fault tolerance
US7453898B1 (en) 2002-03-30 2008-11-18 Cisco Technology, Inc. Methods and apparatus for simultaneously scheduling multiple priorities of packets
US20080317024A1 (en) * 2007-06-22 2008-12-25 Sun Microsystems, Inc. Switch arbitration
US7643493B1 (en) 2004-09-29 2010-01-05 Altera Corporation Method and apparatus for priority-provisioned arbitration scheduling for a switch fabric
US20110310729A1 (en) * 2010-06-17 2011-12-22 Cisco Technology, Inc. Maintaining balance of active links across network devices in a double-sided virtual port-channel environment
US8418129B1 (en) 2001-12-14 2013-04-09 Qualcomm Incorporated Method for automatically generating code to define a system of hardware elements
US8964771B2 (en) 2005-04-15 2015-02-24 Altera Corporation Method and apparatus for scheduling in a packet buffering network
US20150295858A1 (en) * 2014-04-11 2015-10-15 International Business Machines Corporation Simultaneous transfers from a single input link to multiple output links with a timesliced crossbar

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7167478B2 (en) * 2002-09-16 2007-01-23 Ericsson Inc. Versatile system for message scheduling within a packet operating system
US7522527B2 (en) * 2002-12-24 2009-04-21 Electronics And Telecommunications Research Institute Scalable crossbar matrix switch and arbitration method thereof
JP3819368B2 (en) * 2003-02-24 2006-09-06 株式会社東芝 Communication control apparatus, communication control method, a communication control with the server apparatus, communication control by the communication control with the server device method and a communication control program
US7600023B2 (en) * 2004-11-05 2009-10-06 Hewlett-Packard Development Company, L.P. Systems and methods of balancing crossbar bandwidth
US20070201497A1 (en) * 2006-02-27 2007-08-30 International Business Machines Corporation Method and system for high-concurrency and reduced latency queue processing in networks
US20070268825A1 (en) * 2006-05-19 2007-11-22 Michael Corwin Fine-grain fairness in a hierarchical switched system
US7778170B2 (en) * 2007-11-14 2010-08-17 Microsoft Corporation Spectrum and medium access allocation for fairness
KR101495811B1 (en) * 2008-06-09 2015-02-25 삼성전자주식회사 Apparatus and method for high speed packet routing system
US8683100B1 (en) 2011-06-21 2014-03-25 Netlogic Microsystems, Inc. Method and apparatus for handling data flow in a multi-chip environment using an interchip interface
US9021156B2 (en) 2011-08-31 2015-04-28 Prashanth Nimmala Integrating intellectual property (IP) blocks into a processor
US8874976B2 (en) 2011-09-29 2014-10-28 Intel Corporation Providing error handling support to legacy devices
US8711875B2 (en) 2011-09-29 2014-04-29 Intel Corporation Aggregating completion messages in a sideband interface
US8713240B2 (en) 2011-09-29 2014-04-29 Intel Corporation Providing multiple decode options for a system-on-chip (SoC) fabric
US8775700B2 (en) 2011-09-29 2014-07-08 Intel Corporation Issuing requests to a fabric
US8713234B2 (en) 2011-09-29 2014-04-29 Intel Corporation Supporting multiple channels of a single interface
US8805926B2 (en) 2011-09-29 2014-08-12 Intel Corporation Common idle state, active state and credit management for an interface
US9053251B2 (en) 2011-11-29 2015-06-09 Intel Corporation Providing a sideband message interface for system on a chip (SoC)
US9678906B2 (en) 2014-03-26 2017-06-13 International Business Machines Corporation Oldest link first arbitration between links grouped as single arbitration elements

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5493566A (en) * 1992-12-15 1996-02-20 Telefonaktiebolaget L M. Ericsson Flow control system for packet switches
US5689508A (en) * 1995-12-21 1997-11-18 Xerox Corporation Reservation ring mechanism for providing fair queued access in a fast packet switch networks
US5850399A (en) * 1997-04-04 1998-12-15 Ascend Communications, Inc. Hierarchical packet scheduling method and apparatus
US5923656A (en) * 1996-10-22 1999-07-13 Board Of Trustees Of The University Of Illinois Scalable broad band input-queued ATM switch including weight driven cell scheduler
US6069893A (en) * 1996-10-28 2000-05-30 Coreel Microsystems Asynchronous transfer mode switching architectures having connection buffers

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5495474A (en) * 1991-03-29 1996-02-27 International Business Machines Corp. Switch-based microchannel planar apparatus
US5267235A (en) * 1992-05-21 1993-11-30 Digital Equipment Corporation Method and apparatus for resource arbitration
JP3082554B2 (en) * 1994-01-11 2000-08-28 株式会社日立製作所 Self-healing ring switch
US5500858A (en) * 1994-12-20 1996-03-19 The Regents Of The University Of California Method and apparatus for scheduling cells in an input-queued switch

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5493566A (en) * 1992-12-15 1996-02-20 Telefonaktiebolaget L M. Ericsson Flow control system for packet switches
US5689508A (en) * 1995-12-21 1997-11-18 Xerox Corporation Reservation ring mechanism for providing fair queued access in a fast packet switch networks
US5923656A (en) * 1996-10-22 1999-07-13 Board Of Trustees Of The University Of Illinois Scalable broad band input-queued ATM switch including weight driven cell scheduler
US6069893A (en) * 1996-10-28 2000-05-30 Coreel Microsystems Asynchronous transfer mode switching architectures having connection buffers
US5850399A (en) * 1997-04-04 1998-12-15 Ascend Communications, Inc. Hierarchical packet scheduling method and apparatus

Cited By (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8418129B1 (en) 2001-12-14 2013-04-09 Qualcomm Incorporated Method for automatically generating code to define a system of hardware elements
US20080256455A1 (en) * 2001-12-14 2008-10-16 Alberto Alessandro Della Ripa Method for Defining the Physical Configuration of a Communication System
US20080253294A1 (en) * 2001-12-14 2008-10-16 Alberto Alessandro Della Ripa Data link fault tolerance
US7965624B2 (en) 2001-12-14 2011-06-21 Qualcomm Incorporated Data link fault tolerance
US7889729B2 (en) 2001-12-20 2011-02-15 Qualcomm Incorporated System and method for reevaluating granted arbitrated bids
US20080186961A1 (en) * 2001-12-20 2008-08-07 Kenneth Yi Yun System and Method for Reevaluating Granted Arbitrated Bids
US7020131B1 (en) * 2001-12-24 2006-03-28 Applied Micro Circuits Corp. System and method for hierarchical switching
US7453898B1 (en) 2002-03-30 2008-11-18 Cisco Technology, Inc. Methods and apparatus for simultaneously scheduling multiple priorities of packets
US7184443B2 (en) 2002-03-30 2007-02-27 Cisco Technology, Inc. Packet scheduling particularly applicable to systems including a non-blocking switching fabric and homogeneous or heterogeneous line card interfaces
US7289443B1 (en) 2002-03-30 2007-10-30 Cisco Technology, Inc. Slow-start packet scheduling particularly applicable to systems including a non-blocking switching fabric and homogeneous or heterogeneous line card interfaces
US20030193941A1 (en) * 2002-03-30 2003-10-16 Bonomi Flavio Giovanni Packet scheduling particularly applicable to systems including a non-blocking switching fabric and homogeneous or heterogeneous line card interfaces
US7158512B1 (en) * 2002-04-01 2007-01-02 P-Cube Ltd. System and method for scheduling a cross-bar
US20050226263A1 (en) * 2004-04-12 2005-10-13 Cisco Technology, Inc., A California Corporation Weighted random scheduling particularly applicable to packet switching systems
US7061927B2 (en) * 2004-04-12 2006-06-13 Cisco Technology, Inc. Weighted random scheduling particularly applicable to packet switching systems
US8902883B1 (en) 2004-09-29 2014-12-02 Altera Corporation Method and apparatus for priority-provisioned arbitration scheduling for a switch fabric
US8451854B1 (en) * 2004-09-29 2013-05-28 Altera Corporation Method and apparatus for priority-provisioned arbitration scheduling for a switch fabric
US7643493B1 (en) 2004-09-29 2010-01-05 Altera Corporation Method and apparatus for priority-provisioned arbitration scheduling for a switch fabric
US20060165080A1 (en) * 2005-01-24 2006-07-27 International Business Machines Corporation Replicated distributed responseless crossbar switch scheduling
US8964771B2 (en) 2005-04-15 2015-02-24 Altera Corporation Method and apparatus for scheduling in a packet buffering network
US8116332B2 (en) * 2007-06-22 2012-02-14 Oracle America, Inc. Switch arbitration
US20080317024A1 (en) * 2007-06-22 2008-12-25 Sun Microsystems, Inc. Switch arbitration
US20110310729A1 (en) * 2010-06-17 2011-12-22 Cisco Technology, Inc. Maintaining balance of active links across network devices in a double-sided virtual port-channel environment
US8514743B2 (en) * 2010-06-17 2013-08-20 Cisco Technology, Inc. Maintaining balance of active links across network devices in a double-sided virtual port-channel environment
US20150295858A1 (en) * 2014-04-11 2015-10-15 International Business Machines Corporation Simultaneous transfers from a single input link to multiple output links with a timesliced crossbar
US20150295857A1 (en) * 2014-04-11 2015-10-15 International Business Machines Corporation Simultaneous transfers from a single input link to multiple output links with a timesliced crossbar
US9467396B2 (en) * 2014-04-11 2016-10-11 International Business Machines Corporation Simultaneous transfers from a single input link to multiple output links with a timesliced crossbar
US9479455B2 (en) * 2014-04-11 2016-10-25 International Business Machines Corporation Simultaneous transfers from a single input link to multiple output links with a timesliced crossbar

Also Published As

Publication number Publication date Type
US20040218600A1 (en) 2004-11-04 application
US7170903B2 (en) 2007-01-30 grant
WO2003017593A1 (en) 2003-02-27 application

Similar Documents

Publication Publication Date Title
Leonardi et al. Bounds on average delays and queue size averages and variances in input-queued cell-based switches
Ahmadi et al. A survey of modern high-performance switching techniques
US6324165B1 (en) Large capacity, multiclass core ATM switch architecture
US5561663A (en) Method and apparatus for performing communication rate control using geometric weighted groups
US6747971B1 (en) Crosspoint switch with independent schedulers
US6629147B1 (en) Segmentation and reassembly of data frames
Stephens et al. Implementing Distributed Packet Fair Queueing in a scalable switch architecture
US6064650A (en) Rate shaping in per-flow output queued routing mechanisms having output links servicing multiple physical layers
US7133399B1 (en) System and method for router central arbitration
US6038217A (en) Rate shaping in per-flow output queued routing mechanisms for available bit rate (ABR) service in networks having segmented ABR control loops
US5862126A (en) Connection admission control for ATM networks
US6810038B1 (en) Switch, scheduler thereof, and switch scheduling method
US5440546A (en) Packet switch
US6856622B1 (en) Multicast cell scheduling protocol
US6816492B1 (en) Resequencing packets at output ports without errors using packet timestamps and timestamp floors
US6134217A (en) Traffic scheduling system and method for packet-switched networks with fairness and low latency
US6477174B1 (en) Polling response selection using request monitoring in a network switch apparatus
US6975638B1 (en) Interleaved weighted fair queuing mechanism and system
US20020131413A1 (en) Method and apparatus for scheduling for packet-switched networks
US6064651A (en) Rate shaping in per-flow output queued routing mechanisms for statistical bit rate service
US6064677A (en) Multiple rate sensitive priority queues for reducing relative data transport unit delay variations in time multiplexed outputs from output queued routing mechanisms
US7158528B2 (en) Scheduler for a packet routing and switching system
US20050243853A1 (en) Unified scheduling and queueing architecture for a multiservice switch
US7099275B2 (en) Programmable multi-service queue scheduler
US6515991B1 (en) Combined unicast and multicast scheduling

Legal Events

Date Code Title Description
AS Assignment

Owner name: ZAGROS NETWORKS, INC., MARYLAND

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ALASTI, MEHDI;SAYRAFIAN-POUR, KAMRAN;TABATABAEE, VAHID;REEL/FRAME:012080/0272

Effective date: 20010809

AS Assignment

Owner name: PTS CORPORATION, CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:ZAGROS NETWORKS, INC.;REEL/FRAME:014441/0306

Effective date: 20030813