An analysis of the topological properties of the interlaced bypass torus (iBT) networks

Zhang, Peng; Deng, Yuefan

doi:10.1016/j.aml.2012.05.013

Cited by 6 publications

(3 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The idea of optimizing MPL can be dated back to Cerf et al, who first calculated the lower bound of MPL given arbitrary graph size and degree [17]. Zhang et al modified torus by adding bypass links to construct interlaced bypass torus (iBT) [49,51], designed efficient broadcast algorithms on iBT [50] and evaluated its performance by building a re-configurable cluster and system [52]. Feng et al further explored 6D mesh/iBT on custom routing algorithms [25] and performance evaluation by simulation [24].…”

Section: Related Workmentioning

confidence: 99%

Optimal circulant graphs as low-latency network topologies

Huang¹,

Ramos²,

Deng³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Communication latency has become one of the determining factors for the performance of parallel clusters. To design low-latency network topologies for high-performance computing clusters, we optimize the diameters, mean path lengths, and bisection widths of circulant topologies. We obtain a series of optimal circulant topologies of size 2 5 through 2 10 and compare them with torus and hypercube of the same sizes and degrees. We further benchmark on a broad variety of applications including effective bandwidth, FFTE, Graph 500 and NAS parallel benchmarks to compare the optimal circulant topologies and Cartesian products of optimal circulant topologies and fully connected topologies with corresponding torus and hypercube. Simulation results demonstrate superior potentials of the optimal circulant topologies for communication-intensive applications. We also find the

show abstract

Section: Related Workmentioning

confidence: 99%

Optimal circulant graphs as low-latency network topologies

Huang¹,

Ramos²,

Deng³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore we will compare the fault tolerance ability of this deterministic network with the stochastic ones. The detailed description and the analysis of these networks can be found in [7].…”

Section: Algorithms For Shortcuts Constructing and Message Routingmentioning

confidence: 99%

Fault Tolerance of Small-World Regular and Stochastic Interconnection Networks

Demichev,

Ilyin,

Kryukov

et al. 2013

Preprint

View full text Add to dashboard Cite

show abstract

“…HPC clusters achieve their desired high performance through the use of an integrated hardware, software and network [10,11,12,13,14]. A typical cluster contains 500 to 1000 physical nodes, but larger ones are possible [15,16,17,18].…”

mentioning

confidence: 99%

Generalized Matrix Multiplication and its Object Oriented Model

Sedighi¹,

Deng²,

Zhang³

2014

SCPE

Self Cite

View full text Add to dashboard Cite

We claim that the current scheduling systems for high performance computing environments are unable to fairly distribute resources among the users, and as such, are unable to maximize the overall user satisfaction. We demonstrate that a user can game the system to cause a temporal starvation to the other users of the system, even though all users will eventually finish their job in the shared-computing environment. Undesired and unfair delays in the current fair-shared schedulers impede these schedulers from wide deployments to users with diverse usage profiles.

show abstract

An analysis of the topological properties of the interlaced bypass torus (iBT) networks

Cited by 6 publications

References 6 publications

Optimal circulant graphs as low-latency network topologies

Optimal circulant graphs as low-latency network topologies

Fault Tolerance of Small-World Regular and Stochastic Interconnection Networks

Generalized Matrix Multiplication and its Object Oriented Model

Contact Info

Product

Resources

About