OFAR-CM: Efficient Dragonfly Networks with Simple Congestion Management

García, María Jesús Márquez; Vallejo, Enrique; Beivide, Ramón; Valero, Mateo; Rodríguez, Germán

doi:10.1109/hoti.2013.16

Cited by 17 publications

(6 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although this mechanism can be implemented without virtual channels, the poor capacity of the escape subnetwork can become a system bottleneck leading to a significant performance drop in congested scenarios. The use of additional congestion management mechanisms to mitigate this problem has been studied in [19], but still very long paths are theoretically possible. In addition, this mechanism does not work with Wormhole flow-control, and the hops on the escape subnetwork can significantly increase the latency of some packets, which makes alternative mechanisms appealing.…”

Section: Previous Workmentioning

confidence: 99%

Efficient Routing Mechanisms for Dragonfly Networks

García

Vallejo

Beivide

et al. 2013

2013 42nd International Conference on Parallel Processing

Self Cite

View full text Add to dashboard Cite

Abstract-High-radix hierarchical networks are costeffective topologies for large scale computers. In such networks, routers are organized in supernodes, with local and global interconnections. These networks, known as Dragonflies, outperform traditional topologies such as multi-trees or tori, in cost and scalability. However, depending on the traffic pattern, network congestion can lead to degraded performance. Misrouting (non-minimal routing) can be employed to avoid saturated global or local links. Nevertheless, with the current deadlock avoidance mechanisms used for these networks, supporting misrouting implies routers with a larger number of virtual channels. This exacerbates the buffer memory requirements that constitute one of the main constraints in high-radix switches.In this paper we introduce two novel deadlock-free routing mechanisms for Dragonfly networks that support on-thefly adaptive routing. Using these schemes both global and local misrouting are allowed employing the same number of virtual channels as in previous proposals. Opportunistic Local Misrouting obtains the best performance by providing the highest routing freedom, and relying on a deadlock-free escape path to the destination for every packet. However, it requires Virtual Cut-Through flow-control. By contrast, Restricted Local Misrouting prevents the appearance of cycles thanks to a restriction of the possible routes within supernodes. This makes this mechanism suitable for both Virtual Cut-Through and Wormhole networks.Evaluations show that the proposed deadlock-free routing mechanisms prevent the most frequent pathological issues of Dragonfly networks. As a result, they provide higher performance than previous schemes, while requiring the same area devoted to router buffers.

show abstract

Section: Previous Workmentioning

confidence: 99%

Efficient Routing Mechanisms for Dragonfly Networks

García

Vallejo

Beivide

et al. 2013

2013 42nd International Conference on Parallel Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Hence, in each port, only one or two VCs are necessary (depending if the VC belongs to the escape subnetwork). However, this mechanism does not guarantee bounded paths per se, and it requires a congestion management mechanism to avoid saturation in the escape subnetwork, [García et al 2013]. Restricted Local Misrouting (RLM, [García et al 2013a]) allows for local misrouting within any group of a canonical dragonfly without increasing the number of required VCs.…”

Section: Related Workmentioning

confidence: 99%

Topological Characterization of Hamming and Dragonfly Networks and Its Implications on Routing

Camarero

Vallejo

Beivide

2014

ACM Trans. Archit. Code Optim.

Self Cite

View full text Add to dashboard Cite

Current High-Performance Computing (HPC) and data center networks rely on large-radix routers. Hamming graphs (Cartesian products of complete graphs) and dragonflies (two-level direct networks with nodes organized in groups) are some direct topologies proposed for such networks. The original definition of the dragonfly topology is very loose, with several degrees of freedom, such as the inter-and intragroup topology, the specific global connectivity, and the number of parallel links between groups (or trunking level). This work provides a comprehensive analysis of the topological properties of the dragonfly network, providing balancing conditions for network dimensioning, as well as introducing and classifying several alternatives for the global connectivity and trunking level. From a topological study of the network, it is noted that a Hamming graph can be seen as a canonical dragonfly topology with a high level of trunking. Based on this observation and by carefully selecting the global connectivity, the Dimension Order Routing (DOR) mechanism safely used in Hamming graphs is adapted to dragonfly networks with trunking. The resulting routing algorithms approximate the performance of minimal, nonminimal, and adaptive routings typically used in dragonflies but without requiring virtual channels to avoid packet deadlock, thus allowing for lower cost router implementations. This is obtained by properly selecting the link to route between groups based on a graph coloring of network routers. Evaluations show that the proposed mechanisms are competitive with traditional solutions when using the same number of virtual channels and enable for simpler implementations with lower cost. Finally, multilevel dragonflies are discussed, considering how the proposed mechanisms could be adapted to them. ACM Reference Format:Cristóbal Camarero, Enrique Vallejo, and Ramón Beivide. 2014. Topological characterization of Hamming and dragonfly networks and its implications on routing. ACM Trans.

show abstract

“…The case of F N requires special consideration as it produces a pathological scenario for this topology when combined with the standard routing functions (Min and Valiant), but performs relatively well with the other routings. The culprit for its low performance is the network becoming highly congested, which could be alleviated by implementing a congestion management algorithm, e.g., [13]. A detailed examination showed that with other Dragonfly link-arrangements, the performance of F N with Min and Valiant is much closer to that of F P and F N P , but still substantially lower.…”

Section: Dragonfly Topologymentioning

confidence: 99%

High-Performance, Low-Complexity Deadlock Avoidance for Arbitrary Topologies/Routings

Pascual

Navaridas

2018

Proceedings of the 2018 International Conference on Supercomputing

View full text Add to dashboard Cite

Recently, the use of graph-based network topologies has been proposed as an alternative to traditional networks such as tori or fattrees due to their very good topological characteristics. However they pose practical implementation challenges such as the lack of deadlock avoidance strategies. Previous proposals are either exceedingly complex, underutilise network resources or lack flexibility. We propose-and prove formally-three generic, low-complexity deadlock avoidance mechanisms that only require local information. The main strengths of our method are its topology-and routingindependence and that the virtual channel count is bounded by the length of the longest path. We evaluate our proposed mechanisms against previous proposals through an extensive simulation study to measure the impact on the performance using both synthetic and realistic traffic. First we compare against a well-known HPC mechanism for dragonfly and achieved similar performance level. Then we moved to Graph-based networks and show that our mechanisms can greatly outperform traditional, spanning-tree based mechanisms, even if these use a much larger number of virtual channels. Overall, we find that our proposal provides a simple, flexible and high performance deadlock-avoidance solution.

show abstract

OFAR-CM: Efficient Dragonfly Networks with Simple Congestion Management

Cited by 17 publications

References 19 publications

Efficient Routing Mechanisms for Dragonfly Networks

Efficient Routing Mechanisms for Dragonfly Networks

Topological Characterization of Hamming and Dragonfly Networks and Its Implications on Routing

High-Performance, Low-Complexity Deadlock Avoidance for Arbitrary Topologies/Routings

Contact Info

Product

Resources

About