Development of benchmark automation suite and evaluation of various high-performance computing systems

Rho, Seungwoo; Park, Geunchul; Choi, Ji Eun; Park, Chan-Yeol

doi:10.1007/s10586-020-03167-2

Cited by 2 publications

(2 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unfortunately, benchmarking a cluster is a complex and expensive procedure. The availability of special tools to automate this process [6] only slightly reduces the testing cost. Therefore, the problem of obtaining simple formulae that allow calculating the performance of a cluster with an acceptable error (up to 3%) and studying the dependence of this performance on its main parameters without resorting to a real testing procedure is very relevant.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Estimating the Performance of Computing Clusters without Accelerators Based on TOP500 Results

Rybintsev

2022

Mathematics

View full text Add to dashboard Cite

Based on an analysis of TOP500 results, a functional dependence of the performance of clusters without accelerators according to the Linpack benchmark on their parameters was determined. The comparison of calculated and tested results showed that the estimation error does not exceed 2% for processors of different generations and manufacturers (Intel, AMD, Fujitsu) with different technologies of a system interconnect. The achieved accuracy of the calculation allows successful prediction of the performance of a cluster when its parameters (node performance, number of nodes, number of network interfaces, network technology, remote direct memory access, or remote direct memory access over converged Ethernet mode) are changed without resorting to a complex procedure of real testing.

show abstract

Section: Introductionmentioning

confidence: 99%

“…processor performance • node performance • number of nodes • system connectivity (Ethernet, InfiniBand, Omni-Path, etc.) • network topology (Fat Tree [6], Dragonfly [7], Jellyfish [8], 6D Torus [9], etc.) It is rather difficult to reveal the actual dependence of R max on the listed parameters, since some of them cannot be determined numerically.…”

mentioning

confidence: 99%

Estimating the Performance of Computing Clusters without Accelerators Based on TOP500 Results

Rybintsev

2022

Mathematics

View full text Add to dashboard Cite

show abstract

Development of automation tools for managing exchange rates based on nonlinear dynamics methods

Shevnina

Gagarina²,

Chirkov³

et al. 2022

Modeling of Systems and Processes

View full text Add to dashboard Cite

Within the framework of this work, the tasks of studying the subject area of exchange rate management software, comparative analysis of several software solutions were solved. To implement the server part of the PM KB, the Python programming language was chosen. The Django framework formed the basis of the server part of the PM UKB. To implement the client, tools such as the Jinja template engine for collecting HTML pages, the Bootstrap framework for working with a grid and styles were used, and the JS language was used to create interactivity. The paper also presents a general scheme of the algorithm in a graphical form. Further in the article, the program blocks of authentication, data unloading, switching on and off the exchange, collecting modified data, adding the control data block to the point exchange rate management page, updating data in the database, updating data of specific rates are considered.

show abstract

Development of benchmark automation suite and evaluation of various high-performance computing systems

Cited by 2 publications

References 38 publications

Estimating the Performance of Computing Clusters without Accelerators Based on TOP500 Results

Estimating the Performance of Computing Clusters without Accelerators Based on TOP500 Results

Development of automation tools for managing exchange rates based on nonlinear dynamics methods

Contact Info

Product

Resources

About