On the Impact of Routing Algorithms in the Effectiveness of Queuing Schemes in High-Performance Interconnection Networks

Rocher-González, José; Escudero-Sahuquillo, Jesús; García, Pedro J.; Quiles, Francisco J.

doi:10.1109/hoti.2017.16

Cited by 11 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SASHA-4VC-DBBQ also achieves good results, close to those of BBQ-4VC and improving OSHA-4VC-DBBQ. As we mentioned before, adaptive and oblivious routing algorithm can be counterproductive when strong congestion scenarios appear [29], since it is dramatic for the network performance to spread congestion trees by the effect of routing algorithms. Figure 11 shows the execution-time results (in milliseconds) for 2D-Tori and KNS topologies (i.e., network Configurations #1 and #3 of Table 1) when the PTRANS trace-based traffic has been generated.…”

Section: Hot-spot Traffic Resultsmentioning

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.

Self Cite

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: Hot-spot Traffic Resultsmentioning

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.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Congestion can, however, originate from end-nodes themselves (it can then be referred to as end-node congestion), in that case spreading congested traffic will not solve the situation and will instead further increase the problems arising from that traffic to the rest of the topology. In particular, more traffic flows will share paths with congested traffic, hence increasing the probability of the latter causing head-of-line blocking to the former [6]. Furthermore, traffic that was routed adaptively loses the property of being transmitted in-order, potentially causing supplementary cost to the application or communication layer; for that reason the communication layer can mark packets to forbid them from being routed adaptively.…”

Section: Side-note On Adaptive Routingmentioning

confidence: 99%

Node-Type-Based Load-Balancing Routing for Parallel Generalized Fat-Trees

Gliksberg¹,

Quintin²,

García³

2018

2018 IEEE 4th International Workshop on High-Performance Interconnection Networks in the Exascale and Big-Data Era (HiPINEB)

View full text Add to dashboard Cite

High-Performance Computing (HPC) clusters are made up of a variety of node types (usually compute, I/O, service, and GPGPU nodes) and applications don't use nodes of a different type the same way. Resulting communication patterns reflect organization of groups of nodes, and current optimal routing algorithms for all-to-all patterns will not always maximize performance for group-specific communications. Since application communication patterns are rarely available beforehand, we choose to rely on node types as a good guess for node usage. We provide a description of node type heterogeneity and analyse performance degradation caused by unlucky repartition of nodes of the same type. We provide an extension to routing algorithms for Parallel Generalized Fat-Tree topologies (PGFTs) which balances load amongst groups of nodes of the same type. We show how it removes these performance issues by comparing results in a variety of situations against corresponding classical algorithms.

show abstract

“…W ITH the growth of computer networks in recent years, various problems have received significant attention. On the one hand, the design of interconnection networks is pivotal in several domains, such as high-performance computing (HPC) systems [1], [2]. The overall network performance is determined by the topology used in the interconnection network as well as the routing scheme [3].…”

Section: Introductionmentioning

confidence: 99%

“…The overall network performance is determined by the topology used in the interconnection network as well as the routing scheme [3]. On the other hand, studying the algorithms that move data around a network plays a significant role in the scalability of HPC applications [1], [2], and the network's quality of service (QoS). In particular, the time it takes to transmit a message between two communication sites, or the message delay, is one of several factors that affect a network's performance [4].…”

Section: Introductionmentioning

confidence: 99%