|
|
• Содержание выпуска • • Artificial Intelligence, Intelligence Systems, Neural Networks • • Software and Hardware for Distributed Systems and Supercomputers • • Mathematical Foundations of Programming • • Information Systems in Culture and Education • • Healthcare Information Systems • • Methods for Optimal Control and Control Theory • • Mathematical Modelling •
Software and Hardware for Distributed Systems and Supercomputers
Responsible for the Section: Sergei Abramov, Dr. Phys.-Math.Sci.,
corresponding member of RAS
On the left: assigned number of the paper, submission date, the number
of A5 pages contained in the paper,
and the reference to the full-text PDF
.
Article #
14_2018
18 с.
PDF |
submitted on 29th
Dec 2017 displayed on
website on 01th
Nov
2018 I. A. Chernov, N. N. Nikitina
Effective scanning of parameter space in a Desktop Grid for
identification of a hydride decomposition model
We consider parameter identification of a hydride
decomposition model by scanning the parameter space in parallel.
Such problem is resource demanding, but suits best for Desktop Grid
computing. Considering task retrieval as a game, we show that the
search process can be improved to produce solutions faster in
comparison with random search, with no or minor additional cost. (In Russian).
Key words: Desktop Grid, task scheduling, BOINC, Enterprise
Desktop Grid, high-throughput computing, distributed computing,
modeling of hydride decomposition |
article citation |
http://psta.psiras.ru/read/psta2018_4_35-52.pdf |
DOI |
https://doi.org/10.25209/2079-3316-2018-9-4-35-52 |
Article #
15_2018
16 с.
PDF |
submitted on 29th
Dec 2017 displayed on
website on 01th
Nov
2018 I. A. Chernov, N. N. Nikitina
Effective scanning of parameter space in a Desktop Grid for
identification of a hydride decomposition model
We consider parameter identification of a hydride
decomposition model by scanning the parameter space in parallel.
Such problem is resource demanding, but suits best for Desktop Grid
computing. Considering task retrieval as a game, we show that the
search process can be improved to produce solutions faster in
comparison with random search, with no or minor additional cost.
Key words: Desktop Grid, task scheduling, BOINC, Enterprise
Desktop Grid, high-throughput computing, distributed computing,
modeling of hydride decomposition. |
article citation |
http://psta.psiras.ru/read/psta2018_4_53-68.pdf |
DOI |
https://doi.org/10.25209/2079-3316-2018-9-4-53-68 |
Article #
23_2018
14 с.
PDF |
submitted on 09th Nov 2018 displayed on
website on 30th
Nov
2018 Y. V. Shevchuk
Vbinary: variable length integer coding revisited
The article introduces Vbinary, a parametrized
variable-length prefix integer coding. The
coding is considered by means of examples in comparison
with existing codings, including Golomb/Rice and Elias
codings. A naming schema is proposed that
allows to specify the coding parameters concisely.
Vbinary conding uses unusual n-ary extension technique which
makes the coding versatile, usable for both
bit-based and byte-based data streams. By varying
parameters, Vbinary coding can be made efficient for small
numbers or large numbers, tailored to
specific data distribution, tuned for efficient encoding and
decoding. Potential uses for the coding are network
protocols, on-disk and in-memory data
representation, and final stages of data compression algorithms.
Key words: coding of integers, variable length coding, prefix
code, parametrized coding, data compression. |
article citation |
http://psta.psiras.ru/read/psta2018_4_239-252.pdf |
DOI |
https://doi.org/10.25209/2079-3316-2018-9-4-239-252 |
Article #
29_2018
21 с.
PDF |
submitted on 03th Jule 2018 displayed on
website on 17th Dec
2018 Vladimir Roganov, Valerii Osipov,
German Matveyev
Integration of Mantevo MiniAMR and OpenTS. Comparison of OpenTS
DMPI and Open MPI
Mantevo MiniAMR is a miniapp for adaptive mesh
refinement (AMR) that mimics commonly found
workloads and communication patterns in AMR
applications, compact proxy for octree-based AMR. Adaptive Mesh
Refinement is a frequently used technique for
efficiently solving partial differential equations
(PDEs): A finer mesh is used in regions that require a higher
resolution (e.g., because of turbulence), and
a coarse mesh is used to cover less sensitive regions.
The authors describe integration of MiniAMR and OpenTS
parallel programming system. There is
performed comparison between OpenTS DMPI and Open MPI.
(In Russian).
Key words: dynamic parallelization, T-system with an open
architecture, OpenTS, T++ programming language, adaptive algorithm,
adaptive mesh refinement. |
article citation |
http://psta.psiras.ru/read/psta2018_4_361-381.pdf |
DOI |
https://doi.org/10.25209/2079-3316-2018-9-4-361-381 |
Article #
30_2018
15 с.
PDF |
submitted on 13th June 2018 displayed on
website on 11th Dec
2018 Andrey Kamenshchikov, Aleksandr
Oleynikov, Tamara Shirobokova
Standards of the interoperability in a high-performance
environment
The analysis of the interoperability problem in a
high-performance secure environment abroad
and in our country was made. This analysis was based
on a comparison of GRID, supercomputers, clusters and clouds.
The increase in the heterogeneity level and
the activity of the use of high-performance secure
environment is shown. The authors experience in solving the
of interoperability problem is briefly
described, and it is proposed to apply the methodology for
ensuring interoperability for wide class systems, fixed in
GOST R 55062-2012, for a high-performance
environment. (In Russian).
Key words: GRID, supercomputers, clusters, clouds,
interoperability, standards, information security. |
article citation |
http://psta.psiras.ru/read/psta2018_4_383-397.pdf |
DOI |
https://doi.org/10.25209/2079-3316-2018-9-4-383-397 |
Article #
31_2018
17 с.
PDF |
submitted on 12th Nov 2018 displayed on
website on 30th Dec
2018 Nikolay Dikarev, Boris Shabanov,
Aleksandr Shmelev
Insufficient memory bandwidth on Stencil code: the advantage of a
vector dataflow processor
The main factor limiting performance for the most
part of highperformance computing
applications is insufficient memory bandwidth, not
computational power. Software methods for overcoming this
drawback are block methods localizing memory
accesses within fast on-chip memory, and
“software pipelining” for organizing calculations in the form of
chains of arithmetic commands between memory
accesses.
Software pipelining was applied to the 2D and 3D Stencil programs in
vector dataflow processor. Achieved
performance was significantly higher than it is
possible to get for the best traditional processors. (In
Russian).
Key words: vector processor, dataflow architecture,
shared-memory multiprocessor, performance
evaluation. |
article citation |
http://psta.psiras.ru/read/psta2018_4_399-415.pdf |
DOI |
https://doi.org/10.25209/2079-3316-2018-9-4-399-415 |
Article #
35_2018
15 с.
PDF |
submitted on 09th Nov 2018 displayed on
website on 30th Dec
2018 Y. V. Shevchuk
Vbinary: variable length integer coding revisited
The article introduces Vbinary, a parametrized
variable-length prefix integer coding. The
coding is considered by means of examples in comparison
with existing codings, including Golomb/Rice and Elias
codings. A naming schema is proposed that
allows to specify the coding parameters concisely.
Vbinary conding uses unusual n-ary extension technique which
makes the coding versatile, usable for both
bit-based and byte-based data streams. By varying
parameters, Vbinary coding can be made efficient for small
numbers or large numbers, tailored to
specific data distribution, tuned for efficient encoding and
decoding. Potential uses for the coding are network
protocols, on-disk and in-memory data
representation, and final stages of data compression algorithms. (In Russian).
Key words: coding of integers, variable length coding, prefix
code, parametrized coding, data compression. |
article citation |
http://psta.psiras.ru/read/psta2018_4_477-491.pdf |
DOI |
https://doi.org/10.25209/2079-3316-2018-9-4-477-491 |
Article #
40_2018
31 с.
PDF |
submitted on 03th Nov 2018 displayed on
website on 30th Dec
2018 Nadezhda Zhivchikova, Yury
Shevchuk
Experiments with sensor data storage
performance
We consider the task of storing sensor data that
arrive in small portions from a large number
of sources. We suggest a general sensor data storing
model with RAM buffering and experimentally evaluate its
implementation based on Linux filesystems. We
use blktrace tool to study the slowdown that occurs with
larger number of sources, and discuss the ways to sustain
performance. We give experimental data for
write performance and write amplification in ext4 and f2fs
filesystems. (In Russian).
Key words: sensor data storage, Linux, filesystem, write
performance, write amplification, time series
data, TSDB. |
article citation |
http://psta.psiras.ru/read/psta2018_4_597-627.pdf |
DOI |
https://doi.org/10.25209/2079-3316-2018-9-4-597-627 |
• Содержание выпуска • • Artificial Intelligence, Intelligence Systems, Neural Networks • • Software and Hardware for Distributed Systems and Supercomputers • • Mathematical Foundations of Programming • • Information Systems in Culture and Education • • Healthcare Information Systems • • Methods for Optimal Control and Control Theory • • Mathematical Modelling •
|