Technology-Driven, Highly-Scalable Dragonfly Topology

Kim, John; Dally,; Scott,; Abts,

doi:10.1109/isca.2008.19

Cited by 486 publications

(325 citation statements)

References 27 publications

Supporting

Mentioning

319

Contrasting

Unclassified

Order By: Relevance

“…Multi-level direct networks have been proposed recently by several researchers as a scalable topology for connecting a large number of nodes together [1], [2], [3], [4]. The basic idea behind these networks is to have a topology that resembles an all-to-all at each level of the hierarchy which gives the impression of a highly connected network.…”

Section: The Dragonfly Interconnectmentioning

confidence: 99%

“…We focus on the Cascade system which is based on the dragonfly topology designed by Kim et al [1]. The Cascade (Cray XC30) system uses the Aries router as its building block and has been used in supercomputers such as Edison at NERSC, Lawrence Berkeley National Laboratory and Piz Daint at the Swiss National Supercomputing Centre.…”

Section: The Dragonfly Interconnectmentioning

confidence: 99%

“…The implementations for different routing schemes build upon the generic model and customize it to improve the prediction capability and computation time. In the following description of the routing schemes that are based on schemes proposed by Kim et al [1], it is assumed that a message is sent from the source router s to the destination router d.…”

Section: B Parallel Network Routing Predictionmentioning

confidence: 99%

“…This has pushed the HPC community towards designing low-diameter fast networks with large bisection bandwidth. The dragonfly topology [1] and its variants [2], [3], [4] are actively being explored as the interconnects that satisfy all these requirements.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Maximizing Throughput on a Dragonfly Network

Jain

Bhatelé

et al. 2014

SC14: International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

Abstract-Interconnection networks are a critical resource for large supercomputers. The dragonfly topology, which provides a low network diameter and large bisection bandwidth, is being explored as a promising option for building multi-Petaflop/s and Exaflop/s systems. Unlike the extensively studied torus networks, the best choices of message routing and job placement strategies for the dragonfly topology are not well understood. This paper aims at analyzing the behavior of an interconnect based on the dragonfly topology for various routing strategies, job placement policies, and application communication patterns. Our study is based on a novel model that predicts traffic on individual links for direct, indirect, and adaptive routing strategies. We analyze results for individual communication patterns and some common parallel job workloads. The predictions presented in this paper are for a 100+ Petaflop/s prototype machine with 92,160 highradix routers and 8.8 million cores.

show abstract

Section: The Dragonfly Interconnectmentioning

confidence: 99%

Section: The Dragonfly Interconnectmentioning

confidence: 99%

Section: B Parallel Network Routing Predictionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Maximizing Throughput on a Dragonfly Network

Jain

Bhatelé

et al. 2014

SC14: International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

show abstract

“…There have been many recent proposals for scale-out multipath data center topologies such as Clos networks [2,16,23], direct networks like HyperX [1], Flattened Butterfly [20], DragonFly [21], and even randomly connected topologies proposed in Jellyfish [27]. Many current proposals use ECMP-based techniques which are inadequate to utilize all paths, or to dynamically load balance traffic.…”

Section: Related Workmentioning

confidence: 99%

Dahu: Commodity switches for direct connect data center networks

Radhakrishnan

Tewari

Kapoor

et al. 2013

Architectures for Networking and Communications Systems

View full text Add to dashboard Cite

Solving "Big Data" problems requires bridging massive quantities of compute, memory, and storage, which requires a very high bandwidth network. Recently proposed direct connect networks like HyperX [1] and Flattened Butterfly [20] offer large capacity through paths of varying lengths between servers, and are highly cost effective for common data center workloads. However data center deployments are constrained to multi-rooted tree topologies like Fat-tree [2] and VL2 [16] due to shortest path routing and the limitations of commodity data center switch silicon.In this work we present Dahu 1 , simple enhancements to commodity Ethernet switches to support direct connect networks in data centers. Dahu avoids congestion hot-spots by dynamically spreading traffic uniformly across links, and forwarding traffic over non-minimal paths where possible. By performing load balancing primarily using local information, Dahu can act more quickly than centralized approaches, and responds to failure gracefully. Our evaluation shows that Dahu delivers up to 500% improvement in throughput over ECMP in large scale HyperX networks with over 130,000 servers, and up to 50% higher throughput in an 8,192 server Fat-tree network.

show abstract

PDNOC: Partially diagonal network‐on‐chip for high efficiency multicore systems

Leppänen

Liljeberg

et al. 2014

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYWith the constantly increasing of number of cores in multicore processors, more emphasis should be paid to the on-chip interconnect. Performance and power consumption of an on-chip interconnect are directly affected by the network topology. Researchers have proposed various topologies to optimize these metrics. The efficiency can also be optimized by proper mapping of applications. Therefore in this paper, we propose a novel partially diagonal network-on-chip (PDNOC) design that takes advantage of both heterogeneous network topology and congestion-aware application mapping. We analyse the partially diagonal network in terms of interconnect structure, area usage, power consumption, routing algorithm and implementation complexity. The key insight that enables the PDNOC is that most communication patterns in real-world applications are hot-spot and bursty. We implement a full system simulation environment using SPLASH-2 benchmarks. Performance metrics of standard mesh, concentrated mesh, full diagonal mesh and four types of the proposed PDNOC are measured in terms of network latency, application execution time and energy delay product. Evaluation results show that on average, the proposed PDNOC designs provide up to 36% improvement in execution time over concentrated mesh, and 3.6 better energy delay product over fully connected diagonal network. PDNOC design with two adjacent PD networks is a better candidate for higher efficiency, while four PD networks provide better performance.

show abstract

Technology-Driven, Highly-Scalable Dragonfly Topology

Cited by 486 publications

References 27 publications

Maximizing Throughput on a Dragonfly Network

Maximizing Throughput on a Dragonfly Network

Dahu: Commodity switches for direct connect data center networks

PDNOC: Partially diagonal network‐on‐chip for high efficiency multicore systems

Contact Info

Product

Resources

About