[hpc-announce] Call for papers: IEEE CG&A Special Issue on High Performance Visualization and Analysis

[Wes Bethel]

CFP URL: http://www.computer.org/web/computingnow/cgacfp3

IEEE Computer Graphics and Applications

IEEE CG&A Special Issue on High Performance Visualization and Analysis

Final submissions due: 1 September 2015
Publication date: May/June 2016

In the 27 years since the groundbreaking report by McCormick, DeFanti, 
and Brown that coined the phrase “visualization in scientific 
computing,” we have witnessed a dramatic growth in our ability to 
collect and generate data. Concurrently, computing technology has 
rapidly evolved from single-processor systems to large scale, 
multi-petaflop systems comprised of 10Ks to 100Ks processors, with 
processors having upwards of 100s of cores per chip. The confluence of 
larger HPC systems, data sets of unprecedented size and complexity, and 
complex lines of inquiry, gives rise to diverse and difficult research 
challenges and opportunities for visualization and analysis that were 
only dimly visible at the dawn of the field of scientific visualization.

We define high performance visualization and analysis as those methods 
that are, by their design, capable of taking advantage of modern 
computational platforms, either in whole or in part. “In whole” refers 
to techniques that are capable of effectively using all computational 
resources on today’s largest computational platforms. “In part” refers 
to techniques that are specifically designed and implemented to take 
advantage of new processor or system architectures in one way or another.

The upcoming Special Issue of IEEE Computer Graphics and Applications 
will focus on High Performance Visualization and Analysis (HPVA). For 
this special issue, we solicit papers presenting original research that 
span a diversity of visualization and analysis topics including:

New algorithms and methods for knowledge discovery suitable for use on 
modern computational platforms, methods that leverage the extreme-scale 
concurrency of these platforms to solve a problem of extreme scale or 
complexity.
Examples of new methods for visualization and analysis that are designed 
to take advantage of new architectural features, such as deepening 
memory hierarchies, extreme-scale concurrency, etc.; methods that 
overcome the challenges inherent to modern HPC platforms where, for 
example, it is increasingly expensive to move data through the memory 
hierarchy and increasingly intractable to save full-resolution data to 
persistent storage for subsequent analysis.
Case studies/applications of HPVA methods to solve knowledge discovery 
problems in physical or social science, engineering, medicine, etc. 
where there is a thematic element of size and/or complexity that is made 
tractable through the use of new scalable methods making use of modern 
HPC-class platforms.

See http://www.computer.org/web/computingnow/cgacfp3 for additional 
information.

-- 
Wes Bethel -- voice (510) 486-7353 -- fax (510) 486-5812 -- vis.lbl.gov