Level-Spread: A New Job Allocation Policy for Dragonfly Networks

Zhang, Yijia; Tuncer, Ozan; Kaplan, Fulya; Olcoz, Katzalin; Leung, Vitus J.; Coskun, Ayse K.

doi:10.1109/ipdps.2018.00121

Cited by 10 publications

(3 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Zhang et al 21 introduces an allocation policy for dragonfly networks where jobs are spread within the smallest network level such that a given job can fit in at the time of its allocation.…”

Section: Related Workmentioning

confidence: 99%

“…The effects of task placement on Dragonfly networks have been addressed in some studies but to the best of our knowledge none has discussed its impact on the all‐to‐all FFT algorithm specifically. Prisacari et al 18,19 compare the impact of different job placement policies and routing strategies. Yang et al 20 study the effects of task placement on Dragonfly network and further investigate the contiguous versus noncontiguous allocation effects on applications. Zhang et al 21 introduce an allocation policy for dragonfly networks where jobs are spread within the smallest network level such that a given job can fit in at the time of its allocation. On the topic of rank‐reordering, Esposito et al 22 analyze the combination of grid topologies and rank reordering for different matrix sizes and number of nodes of a Cray XC system. …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A scheduling policy to save 10% of communication time in parallel fast Fourier transform

Aseeri

Chatterjee

Verma

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

The fast Fourier transform (FFT) has applications in almost every frequency related studies, e.g. in image and signal processing, and radio astronomy. It is also used to solve partial differential equations used in fluid flows, density functional theory, many-body theory, and others. Three-dimensional 3 FFT has large time complexity ( 3 log 2 ). Hence, parallel algorithms are made to compute such FFTs. Popular libraries perform slab division or pencil decomposition of 3 data. None of the existing libraries have achieved perfect inverse scaling of time with ( −1 ≈ ) cores because FFT requires all-to-all communication and clusters hitherto do not have physical all-to-all connections. Dragonfly, one of the popular topologies for the interconnect, supports hierarchical connections among the components. Thus, we show that if we align the all-to-all communication of FFT with the physical connections of Dragonfly topology we will achieve a better scaling and reduce communication time.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A scheduling policy to save 10% of communication time in parallel fast Fourier transform

Aseeri

Chatterjee

Verma

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Figure 1 demonstrates runtime variability when running fast Fourier Transform programs [9,10] for solving the Navier-Stokes equations on Hazelhen and Shaheen II. Both these computers use adaptive routing and job placement to speed up job throughput [18], but this can result in some run time variability, a subject of current research [7,28,35,38,39].…”

Section: Lessons Learnedmentioning

confidence: 99%

Reproducibility in Benchmarking Parallel Fast Fourier Transform based Applications

Aseeri

Muite

Takahashi

2019

Companion of the 2019 ACM/SPEC International Conference on Performance Engineering

View full text Add to dashboard Cite

An overview of concerns observed in allowing for reproducibility in parallel applications that heavily depend on the three dimensional distributed memory fast Fourier transform are summarized. Suggestions for reproducibility categories for benchmark results are given. CCS CONCEPTS• Mathematics of computing → Computation of transforms;• Theory of computation → Massively parallel algorithms;• Software and its engineering → Software performance; • Hardware → Testing with distributed and parallel systems.

show abstract