Parallel maximum weight bipartite matching algorithms for scheduling in input-queued switches

Fayyazi, Morteza; Kaeli, David; Meleis, Waleed

doi:10.1109/ipdps.2004.1302904

Cited by 14 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using MWM (Maximum Weighted Matching) as crossbar schedules is known to result in 100% switch throughput and near-optimal queueing delays under various traffic patterns [15], but each MWM takes O(N 2.5 log W ) time to compute using the state-of-theart algorithm [4], where W is the maximum possible length of a VOQ. Motivated by this, various parallel exact or approximate MWM algorithms (e.g., [2,5]) have been proposed to reduce its time complexity. However, the time complexities of all these algorithms above are still too high to be used in high-line-rate high-radix switches.…”

Section: Related Workmentioning

confidence: 99%

Sliding-Window QPS (SW-QPS): A Perfect Parallel Iterative Switching Algorithm for Input-Queued Switches

Meng,

Gong,

Jun

2020

Preprint

View full text Add to dashboard Cite

In this work, we first propose a parallel batch switching algorithm called Small-Batch Queue-Proportional Sampling (SB-QPS). Compared to other batch switching algorithms, SB-QPS significantly reduces the batch size without sacrificing the throughput performance and hence has much lower delay when traffic load is light to moderate. It also achieves the lowest possible time complexity of O(1) per matching computation per port, via parallelization. We then propose another algorithm called Sliding-Window QPS (SW-QPS). SW-QPS retains and enhances all benefits of SB-QPS, and reduces the batching delay to zero via a novel switching framework called sliding-window switching. In addition, SW-QPS computes matchings of much higher qualities, as measured by the resulting throughput and delay performances, than QPS-1, the state-of-the-art regular switching algorithm that builds upon the same underlying bipartite matching algorithm.

show abstract

Section: Related Workmentioning

confidence: 99%

Sliding-Window QPS (SW-QPS): A Perfect Parallel Iterative Switching Algorithm for Input-Queued Switches

Meng,

Gong,

Jun

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The most representative among them are [19,[27][28][29]. A parallel algorithm with a sub-linear per-node computational complexity of O( √ N log 2 N ) was proposed in [27] for computing MWM exactly in a bipartite graph. However, this algorithm requires the use of O(N 3 ) processors.…”

Section: Parallel/distributed Mwm Algorithmsmentioning

confidence: 99%

SERENADE: A Parallel Randomized Algorithm Suite for Crossbar Scheduling in Input-Queued Switches

Gong,

Liu,

Yang

et al. 2017

Preprint

View full text Add to dashboard Cite

Most of today's high-speed switches and routers adopt an input-queued crossbar switch architecture. Such a switch needs to compute a matching (crossbar schedule) between the input ports and output ports during each switching cycle (time slot). A key research challenge in designing large (in number of input/output ports N ) input-queued crossbar switches is to develop crossbar scheduling algorithms that can compute "high quality" matchings -i.e., those that result in high switch throughput (ideally 100%) and low queueing delays for packets -at line rates. SERENA is one such algorithm: it outputs excellent matching decisions that result in 100% switch throughput and reasonably good queueing delays. However, since SERENA is a centralized algorithm with O(N ) computational complexity, it cannot support switches that both are large and have a very high line rate per port. In this work, we propose SERENADE (SERENA, the Distributed Edition), a parallel iterative algorithm that emulates SERENA in only O(log N ) iterations between input ports and output ports, and hence has a time complexity of only O(log N ) per port. We prove that SERENADE can exactly emulate SERENA. We also propose an early-stop version of SERENADE, called O-SERENADE, to only approximately emulate SERENA. Through extensive simulations, we show that O-SERENADE can achieve 100% throughput and that it has similar as or slightly better delay performance than SERENA under various load conditions and traffic patterns.

show abstract

“…The time complexity of this algorithm can be reduced to O(log n) using the O(1) parallel selection algorithm described in [6] with O(n 4 ) processing elements using a CRCW PRAM model. This O(log n) parallel algorithm is also described in [7].…”

Section: Providing a Basis Feasible Sets And Nc-search Algorithmmentioning

confidence: 99%

An adjustable linear time parallel algorithm for maximum weight bipartite matching

Fayyazi

Kaeli

Meleis

2006

Information Processing Letters

View full text Add to dashboard Cite

Parallel maximum weight bipartite matching algorithms for scheduling in input-queued switches

Cited by 14 publications

References 20 publications

Sliding-Window QPS (SW-QPS): A Perfect Parallel Iterative Switching Algorithm for Input-Queued Switches

Sliding-Window QPS (SW-QPS): A Perfect Parallel Iterative Switching Algorithm for Input-Queued Switches

SERENADE: A Parallel Randomized Algorithm Suite for Crossbar Scheduling in Input-Queued Switches

An adjustable linear time parallel algorithm for maximum weight bipartite matching

Contact Info

Product

Resources

About