Ensuring Deadlock-Freedom in Low-Diameter InfiniBand Networks

Schneider, Timo; Bibartiu, Otto; Hoefler, Torsten

doi:10.1109/hoti.2016.015

Cited by 9 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it is possible for the packet that traverses the network to modify the VL it is stored. This functionality is available in some strategies [30] that differ from us in the way the SL-to-VL tables are filled in. It is important to mention that IBA defines a maximum of 16 different SLs and VLs, although some manufacturers, such as Mellanox, build IBA devices supporting 16 SLs but only 9 VLs.…”

Section: Infiniband Implementationmentioning

confidence: 99%

Combining Source-adaptive and Oblivious Routing with Congestion Control in High-performance Interconnects using Hybrid and Direct Topologies

Yébenes

Rocher-González

Escudero-Sahuquillo

et al. 2019

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Hybrid and direct topologies are cost-efficient and scalable options to interconnect thousands of end nodes in high-performance computing (HPC) systems. They offer a rich path diversity, high bisection bandwidth, and a reduced diameter guaranteeing low latency. In these topologies, efficient deterministic routing algorithms can be used to balance smartly the traffic flows among the available routes. Unfortunately, congestion leads these networks to saturation, where the HoL blocking effect degrades their performance dramatically. Among the proposed solutions to deal with HoL blocking, the routing algorithms selecting alternative routes, such as adaptive and oblivious, can mitigate the congestion effects. Other techniques use queues to separate congested flows from non-congested ones, thus reducing the HoL blocking. In this article, we propose a new approach that reduces HoL blocking in hybrid and direct topologies using source-adaptive and oblivious routing. This approach also guarantees deadlock-freedom as it uses virtual networks to break potential cycles generated by the routing policy in the topology. Specifically, we propose two techniques, called Source-Adaptive Solution for Head-of-Line Blocking Avoidance (SASHA) and Oblivious Solution for Head-of-Line Blocking Avoidance (OSHA). Experiment results, carried out through simulations under different traffic scenarios, show that SASHA and OSHA can significantly reduce the HoL blocking.

show abstract

Section: Infiniband Implementationmentioning

confidence: 99%

Combining Source-adaptive and Oblivious Routing with Congestion Control in High-performance Interconnects using Hybrid and Direct Topologies

Yébenes

Rocher-González

Escudero-Sahuquillo

et al. 2019

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

show abstract

“…Distance-based deadlock avoidance: Seminal works on distance-based deadlock avoidance in store-and-forward networks were introduced by Günther [1] and Gopal [29]. Several current systems employ such mechanisms (or a variation of them), such as IBM PERCS [30] or Cray Cascade [3], and have been extended to commodity InfiniBand [31]. In these proposals, the amount of required VCs increases with the maximum path length.…”

Section: Related Workmentioning

confidence: 99%

“…NUE routing [36] provides better results by combining path computing and VL assignment in a single calculation, but still assigns VLs to complete paths. Schneider et al extend distance-based deadlock avoidance to Infiniband in [31], but still determine a single fixed output VC per packet.…”

Section: Related Workmentioning

confidence: 99%

FlexVC: Flexible Virtual Channel Management in Low-Diameter Networks

Fuentes

Vallejo

Beivide

et al. 2017

2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS)

View full text Add to dashboard Cite

Deadlock avoidance mechanisms for lossless lowdistance networks typically increase the order of virtual channel (VC) index with each hop. This restricts the number of buffer resources depending on the routing mechanism and limits performance due to an inefficient use. Dynamic buffer organizations increase implementation complexity and only provide small gains in this context because a significant amount of buffering needs to be allocated statically to avoid congestion. We introduce FlexVC, a simple buffer management mechanism which permits a more flexible use of VCs. It combines statically partitioned buffers, opportunistic routing and a relaxed distancebased deadlock avoidance policy. FlexVC mitigates Head-of-Line blocking and reduces up to 50% the memory requirements. Simulation results in a Dragonfly network show congestion reduction and up to 37.8% throughput improvement, outperforming more complex dynamic approaches. FlexVC merges different flows of traffic in the same buffers, which in some cases makes more difficult to identify the traffic pattern in order to support nonminimal adaptive routing. An alternative denoted FlexVC-minCred improves congestion sensing for adaptive routing by tracking separately packets routed minimally and nonminimally, rising throughput up to 20.4% with 25% savings in buffer area.

show abstract