Most Recent Verify Posts RSS

Allocate Memory Effectively in TBB

Threading Building Blocks (TBB) includes drop-in replacement functions for the C-type memory allocation functions, as well as replacements for the C++-type allocators. Jeff Cogswell introduces them. When you’re using Threading Building Blocks and writing an app that makes use of the basic memory allocation functions, there’s a feature whereby you …

Read Full Post Posted in Build | Leave a comment

Cray Enters New Territory with Chapel Parallel Language

Supercomputer manufacturer Cray is developing a general-purpose, parallel programming language called Chapel that supports Intel architecture. Jeff Cogswell gives you a quick tour. When people think of Cray, they usually think of supercomputers. But something else is coming out of the company: A language for parallel programming that is portable …

Read Full Post Posted in Build | Leave a comment

Explicit Vectorization Explained

Vectorizing your code improves performance, which can also save power. The faster an application can compute CPU-intensive regions, the faster the CPU can be set to a lower power state. With resources at a premium, being able to improve performance while saving on energy costs is very attractive indeed. In …

Read Full Post Posted in Verify | Leave a comment

How TBB Goes Atomic

Threading Building Blocks includes several atomic functions. But how does it make these atomics work? Jeff Cogswell looks under the hood of TBB to find what assembly code it uses.   Threading Building Blocks  contains several atomic operations that allow you to run lockless parallel applications. In order to accomplish …

Read Full Post Posted in Verify | Leave a comment

Parallelizing Tachyon Ray Tracer

Parallel Studio includes a sample program called Tachyon, which is an open-source ray tracer. Intel has included versions of the code that add parallelization, greatly speeding it up. Jeff Cogswell shows you how to try it out. When you install Parallel Studio XE, you get several sample applications that live …

Read Full Post Posted in Verify | Leave a comment

How Numerical Integration Adds Up for Parallel Coding

In the second part of a look at the intersection between scientific applications and parallel programming, Jeff Cogswell explores numerical integration as a prime example of how scientific applications can use parallelization in their mathematical functions. Last time, I talked about how software such as SeisSol, a seismology simulator, uses …

Read Full Post Posted in Verify | Leave a comment

Applying Scientific Apps to Parallel Programming

To come up with a highly parallel scientific app for modeling seismic activity, developers turned to an important mathematical technique for solving differential equations. Jeff Cogswell discusses how similar functions work well with parallel programming. We recently ran a news story on Go Parallel about a high-performance application called SeisSol. …

Read Full Post Posted in Verify | Leave a comment

Vectorization Gets Explicit with Intel’s Updated Parallel Tool

Intel Parallel Studio XE 2015′s explicit vectorization feature allows programmers to define SIMD and other loops that should split and run in parallel on separate compute nodes. Intel has released a new version of its HPC application development suite with support for more hardware and development languages and a simple …

Read Full Post Posted in Verify | Leave a comment

IDF14: Tech Sessions with Parallel Programming Experts

The upcoming Intel Developers Forum (IDF) Sept 9-11 in San Francisco is chock-full of sessions that should whet the appetite for the parallel processing and HPC community. Below is a look at just a sample of what you can expect when you attend this year’s IDF. Technical sessions include: • …

Read Full Post Posted in Design | Leave a comment

Charm++ Parallel Programming Language Cools HPC Power Problems

The Charm++ parallel programming model supports load balancing between cores while adjusting the frequencies of the cores during high temperature processes. Researchers at the University of Illinois at Urbana-Champaign’s Parallel Programming Laboratory have improved their Charm++ parallel programming model and runtime’s load balancing system to adjust the balancing based on …

Read Full Post Posted in Build | Leave a comment