|
|
 |
• Содержание выпуска •
• Methods for Optimal Control and Control Theory •
• Software and Hardware for Distributed Systems and Supercomputers •
• Mathematical Foundations of Programming •
• Artificial Intelligence, Intelligence Systems, Neural Networks •
• 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
/r1/pdf.jpg) .
|
Article # 2_2016
13
p.
/r1/pdf.jpg)
PDF |
submitted on 10th
Dec 2015 displayed on
website on 21th Jan
2016 E. Biryaltsev E., P. Bogdanov,
M. Galimov, D. Demidov, A. Elizarov
HPC Platform for Oil&Gas Industry
There is a growing need in integrated software and
hardware solutions for high performance computing, big data
analysis, and 3D realtime visualization in the oil and gas industry.
We describe technical details and results obtained while creating
such an integrated software and hardware solution, and talk about
challenges and directions for further progress in the technological
field. (In Russian).
Key words: general-purpose computing for graphics processing
untis, numerical simulation, distributed computing, GPU cluster, oil
and gas industry, HPC platform. |
|
article citation |
http://psta.psiras.ru/read/psta2016_1_15-27.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2016-7-1-15-27 |
|
Article # 3_2016
18
p.
PDF |
submitted on 11th
Dec 2015 displayed on
website on 21th Jan
2016 Smirnov S.A., Voloshinov V.V.
Effective use of discrete optimization solvers in cloud
infrastructure on the basis of heuristic decomposition of the
initial problem by optimization modeling system AMPL
Let’s take some discrete optimization problem and a
time interval to find its solution by some branch and bound solver.
Can we reduce time of solving if a number of instances of the same
solver will run simultaneously with the same problem? We have got
significant acceleration with open source solvers in
multi-processors cloud computing environment. Our approach is based
on the following:
(1) preliminary decomposition of the given problem into sub-problems
via fixing integer values of some discrete variables selected in
accordance with some heuristic rule implemented as a program in
optimization modelling language AMPL;
(2) parallel solving of sub-problems by a pool of solvers which run
simultaneously and exchange incumbents (bounds of optimal values in
branch-and-bound method) they found in processing of sub-problems.
The approach is attractive due to the relative simplicity of program
implementation. It has been verified on “Traveling Salesman (TSP)”
and “Task-to-Worker scheduling” problems. (In Russian).
Key words: discrete optimization, branch and bound algorithm,
prior heuristic decomposition. |
|
article citation |
http://psta.psiras.ru/read/psta2016_1_29-46.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2016-7-1-29-46 |
|
Article # 5_2016
37
p.
PDF |
submitted on 27th
Jan 2016 displayed on
website on 24th Feb
2016 Isupov K.S., Knyazkov V.S.
Parallel multiple-precision arithmetic based on residue number
system
This paper deals with algorithms of
multiple-precision arithmetic, based on the use of multi module
residue number systems for representing of arbitrary length
significands of floating-point numbers; the exponent is represented
in the binary number system. Such number representation provides a
large dynamic range and allows for effective parallelization of
arithmetic operations on the digits of multiple-precision
significands across RNS modules. This agrees well with the
architectural features of modern parallel computing systems.
Additionally, the attributive information which provides a fast
estimation for the relative value of significand and allows you to
increase the speed of executing complex non-modular operations in
RNS, such as comparison, overflow control, rounding, etc., is
included into the number format. Results of an experimental study on
precision, performance and SIMD efficiency of multiple-precision
algorithms are presented. (In Russian).
Key words: computer arithmetic, high-precision computations,
parallel algorithms, residue number system, SIMD. |
|
article citation |
http://psta.psiras.ru/read/psta2016_1_61-97.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2016-7-1-61-97 |
|
Article # 7_2016
18
p.
PDF |
submitted on 18th
Dec 2015 displayed on
website on 29th Feb
2016 Baranov A.V., Nikolaev D.S.
The use of container virtualization in the organization of
high-performance computing
The article focuses on the applicability of
operating-system-level virtualization (Linux containers) in the
organization of high-performance computing in terms of the isolation
of user tasks. The results of experiments evaluate the overhead cost
of operating-system-level virtualization, the degree of mutual
influence of user tasks and protection from unauthorized access. (In
Russian).
Key words: high-performance computing, operating-system-level
virtualization, Linux containers, LXC, Docker, overhead cost of
virtualization. |
|
article citation |
http://psta.psiras.ru/read/psta2016_1_117-134.pdf |
|
DOI |
https://doi.org/10.25209/2079-3316-2016-7-1-117-134 |
• Содержание выпуска •
• Methods for Optimal Control and Control Theory •
• Software and Hardware for Distributed Systems and Supercomputers •
• Mathematical Foundations of Programming •
• Artificial Intelligence, Intelligence Systems, Neural Networks •
• Mathematical Modelling •
|
.en.html_cmp_strtedge110_vbtn.gif){kind=small}
/_derived/content-02(029).en.html_cmp_strtedge110_vbtn.gif){kind=small}
/_derived/content-03(030).en.html_cmp_strtedge110_vbtn.gif){kind=small}
/_derived/content-04(031).en.html_cmp_strtedge110_vbtn.gif){kind=small}
