|
|
 |
• Содержание выпуска •
• Supercomputing Software and Hardware •
• Artificial Intelligence, Intelligence Systems, Neural Networks •
• Information Systems in Culture and Education •
• Methods for Optimal Control and Control Theory •
• Mathematical Foundations of Programming •
• Software and Hardware for Distributed Systems and Supercomputers •
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
/r1/pdf.jpg) .
|
Article #
39_2017
24 с.
/r1/pdf.jpg)
PDF |
submitted on
11th
Dec 2017 displayed on
website on 28th
Dec
2017 Alexei Adamovich, Andrei Klimov
How to create deterministic by construction parallel programs?
Problem statement and survey of related works
One of the main problems that make parallel
programming unreliable, labor-intensive, error-prone, and programs
difficult to debug, is the non-determinism of processes and results
of computation, when several runs of the same program with the same
input data can produce different results because of different order
of interaction of parallel processes. In connection with the rapid
growth in the complexity of programs for supercomputers, the idea of
parallel computations with determinism, guaranteed by language and a
programming system has become more popular in the last decade and is
becoming more vital.
The problem of how to make parallel programming as deterministic as
possible is analyzed. An overview of some approaches to solving it
is given. The task of developing a system that provides an
opportunity to write both deterministic and nondeterministic code
with guarantees to the application programmer that his program is
deterministic, is discussed. (In Russian).
Key words: parallel computation models, deterministic
programs, functional programming, object-oriented programming. |
|
article citation |
http://psta.psiras.ru/read/psta2017_4_221-244.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2017-8-4-221-244 |
|
Article #
40_2017
18 с.
PDF |
submitted on 04th
Dec 2017 displayed on
website on 28th
Dec
2017 Boris Shabanov, Aleksey
Ovsiannikov, Anton Baranov, Sergey Leshchev, Boris Dolgov, Dmitriy
Derbyshev
The distributed network of the supercomputer centers for
collaborative research
The article is devoted to the project of the creation
of the distributed network of the supercomputer centers for
collaborative research. The key technique for improving the
availability and efficiency of the supercomputer centers is the
integration of the centers into a joint distributed network. This
technique allows to redistribute the workload between resources by
the redirecting of the user’s jobs from one center job queue to
another depending on the instant resource usage.
The pilot project of the supercomputer centers distributed network
is based on Joint Supercomputer Center in Moscow and Siberian
Supercomputer Center in Novosibirsk. The centers communicate by the
secure high-performance link. The results of the communication
environment research are presented. (In Russian).
Key words: supercomputer center, center for collaborative
research, secure communications, distributed computing. |
|
article citation |
http://psta.psiras.ru/read/psta2017_4_245-262.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2017-8-4-245-262 |
|
Article #
41_2017
21 с.
PDF |
submitted on
16th
Dec 2017 displayed on
website on 28th
Dec
2017 Y. V. Shevchuk, E. V. Shevchuk,
A. Y. Ponomarev, I. A. Vogt, A. V. Elistratov, A. Y. Vakhrin, R. E.
Yarovicyn
Etherbox: a protocol for modular sensor networks
Etherbox is an application level protocol for sensor
networks that achieves the flexibility needed for modular sensor
networks by representing all commands and data transferred in the
form of virtual machine bytecode. We introduce the Etherbox protocol
and compare it with MQTT-SN and CoAP, then consider the software
architecture of sensor nodes and the controlling computer. (In Russian)
Key words: sensor network, IoT, Etherbox, MQTT, MQTT-SN, CoAP. |
|
article citation |
http://psta.psiras.ru/read/psta2017_4_263-283.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2017-8-4-221-244 |
|
Article #
42_2017
19 с.
PDF |
submitted on
16th
Dec 2017 displayed on
website on 28th
Dec
2017 Y. V. Shevchuk, E. V. Shevchuk,
A. Y. Ponomarev, I. A. Vogt, A. V. Elistratov, A. Y. Vakhrin, R. E.
Yarovicyn
Etherbox: a protocol for modular sensor networks
Etherbox is an application level protocol for sensor
networks that achieves the flexibility needed for modular sensor
networks by representing all commands and data transferred in the
form of virtual machine bytecode. We introduce the Etherbox protocol
and compare it with MQTT-SN and CoAP, then consider the software
architecture of sensor nodes and the controlling computer.
Key words: sensor network, IoT, Etherbox, MQTT, MQTT-SN, CoAP. |
|
article citation |
http://psta.psiras.ru/read/psta2017_4_285-303.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2017-8-4-285-303 |
|
Article #
43_2017
13 с.
PDF |
submitted on
07th
Dec 2017 displayed on
website on 28th
Dec
2017 Nikolay Dikarev, Boris Shabanov,
Aleksandr Shmelev
Execution of sorting algorithms on vector dataflow processor
Sorting is a common problem in
computer science and there is a lot of algorithms, both sequential
and parallel to solve this problem. Therefore, several wellknown
sorting algorithms were selected for testing Vector Dataflow
Processor(VDP). In this paper we analyze advantages and
disadvantages of the VDP for solving the sorting problem and compare
performance based on the simulation results with the best
traditional processors. (In Russian).
Key words: supercomputer, vector processor, dataflow
architecture, performance evaluation, merge sort, bitonic sort. |
|
article citation |
http://psta.psiras.ru/read/psta2017_4_305-317.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2017-8-4-305-317 |
|
Article #
44_2017
8 с.
PDF |
submitted on
28th
Oct 2017 displayed on
website on 28th
Dec
2017 Vladimir Roganov, Valerii Osipov,
German Matveev
Integration of Ray Tracing system Tachyon and T-system with an
open architecture (OpenTS)
This paper describes integration of
Ray Tracing system Tachyon of Intel®
Threading Building Blocks library and T-system with an open
architecture (OpenTS). (In Russian).
Key words: dynamic parallelization, T-system with an open
architecture, OpenTS, T++ language, parallel programming, tachyon,
Ray Tracing, Intel TBB. |
|
article citation |
http://psta.psiras.ru/read/psta2017_4_319-326.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2017-8-4-319-326 |
• Содержание выпуска •
• Supercomputing Software and Hardware •
• Artificial Intelligence, Intelligence Systems, Neural Networks •
• Information Systems in Culture and Education •
• Methods for Optimal Control and Control Theory •
• Mathematical Foundations of Programming •
• Software and Hardware for Distributed Systems and Supercomputers •
|
/_derived/content-01(032).en.html_cmp_strtedge110_vbtn.gif){kind=small}
/_derived/content-02(033).en.html_cmp_strtedge110_vbtn.gif){kind=small}
/_derived/content-03(034).en.html_cmp_strtedge110_vbtn.gif){kind=small}
.en.html_cmp_strtedge110_vbtn.gif){kind=small}
