Tuesday, September 10, 2013

[DMANET] [Algorithms] Algorithms (ISSN 1999-4893) - Special Issue: "Algorithms for Multi Core Parallel Computation" - Special Issue Announcement

Dear Colleagues,

Prof. Dr. Ken Hawick is serving as Guest Editor for a Special Issue
of the online open access journal, Algorithms (ISSN 1999-4893,
http://www.mdpi.com/journal/algorithms). We would like to announce
this opportunity for you to submit either a research or review article.

Algorithms for Multi Core Parallel Computation
http://www.mdpi.com/si/algorithms/algorithms_multi_core_parallel_computation/


Dear Colleagues,

Multi-cored CPUs and many-cored accelerators such as GPUs and
FPGAs are becoming commonly affordable and available platforms
for parallel computing. Developing highly-efficient algorithms and
applications for these platforms is an exciting area and one that presents
even greater challenges as core counts increase. Algorithms that make
optimal use of hybrid many-cored processors and accelerators also
stretch the capabilities of existing parallel software and tools.
Approaches vary from classic data-parallelism algorithms deployed
on massively-cored accelerators to thread or message-managed task
communication on many-core central processing units. The close-coupling
potential of having many cores on a single chip or device presents
opportunities for developing new multi-core algorithms that must of
necessity go beyond the parallel supercomputer algorithms of yester-year.

Hybrid algorithms making use of a whole spectrum of different parallel
granularities are possible on platforms with multiple accelerators driven
by multiple CPU cores. Recent developments have taken the number
of economically feasible CPU cores well into double figures. It is likely
that CPU and GPU solutions will continue to approach the many-core
problem from different ends of the granularity scale, but with interesting
hybrids emerging and with a complex trade-off space where they meet
in the middle. Algorithms that can exploit this topical granularity spectrum
will grow in importance. Software that embodies such algorithms will
have a profound impact in a number of application markets and industries
including: simulation and modelling; games and animation; and many
others where many cores can be kept busy.

Quality articles describing innovative and original work in algorithms that
use many-cored devices are sought for a special issue on "Algorithms for
Multi-Core Parallel Computation." Reported work on algorithms using
topical new and emerging devices such as:
multi-cored CPUs; many-cored GPUs; or other CPU/accelerator
combinations using many cores, is particularly sought.

Prof. Dr. Ken Hawick
Guest Editor

Keywords:

* many-core algorithms
* many-core devices
* performance accelerators
* multi-accelerator algorithms
* multi-cored CPU
* algorithms

The deadline for submission is 30 September 2013, but you
may send your manuscript at any time before then. All submissions
are peer-reviewed and accepted papers will be published immediately.
If you plan to contribute, please send a short abstract to the
Editorial Office (algorithms@mdpi.com).

Algorithms is an open access journal, which has no income from
selling subscriptions. To defray the editorial and production costs,
the journal levies an Article Processing Charge (APC) for accepted papers.
Please refer to http://www.mdpi.com/journal/algorithms/apc/ for more
information. The charge currently is of 300 CHF (Swiss Francs) per
accepted paper.

We look forward to hearing from you.

On behalf of
Prof. Dr. Ken Hawick

Best wishes,
Chelly Cheng
Assistant Editor

---
Ms. Chelly Cheng
MDPI AG
Assistant Editor
Algorithms Editorial Office
Postfach, CH - 4005 Basel, Switzerland
Office: Kandererstrasse 25, 4057 Basel
Tel. +41 61 683 77 34 (office)
Fax. +41 61 302 89 18 (office)
E-mail: algorithms@mdpi.com
http://www.mdpi.com/journal/algorithms
**********************************************************
*
* Contributions to be spread via DMANET are submitted to
*
* DMANET@zpr.uni-koeln.de
*
* Replies to a message carried on DMANET should NOT be
* addressed to DMANET but to the original sender. The
* original sender, however, is invited to prepare an
* update of the replies received and to communicate it
* via DMANET.
*
* DISCRETE MATHEMATICS AND ALGORITHMS NETWORK (DMANET)
* http://www.zaik.uni-koeln.de/AFS/publications/dmanet/
*
**********************************************************