Most Recent Tune Posts RSS



Explicit Multithreading: The Pros and Cons

First, before detailing the pros and the cons, we need to establish some baseline understandings with regard to Explicit Multithreading (XMT). To start with, let’s look at sequential code execution, best described by the von Neumann model. In this model, single instructions pointed to by the instruction counter are fetched, …

Read Full Post Posted in Tune | Tagged , , | Leave a comment

Modernize Your C++ Code

Many software developers are faced with having to update their original code at one time or another. On two occasions in my life I’ve had to revisit code I wrote a few years earlier. The first was in 1996 for code I developed during 1989-1990. The second time was in …

Read Full Post Posted in Tune | Tagged , | Leave a comment

Dodge Dangers of Explicit Multithreading

With all processors now having multiple cores even on mobile devices, an application that seeks high-performance computing needs to use multiple processes or threads. One way to do this is use OpenMP or Intel TBB but these confine you to their specific programming model. The temptation is to program multiple …

Read Full Post Posted in Tune | Tagged , , | Leave a comment

Offload Your Code to Your GPU: How to Get Started

At the time of this writing, most desktop computers have video cards with multiple processor cores that support advanced graphics. But most of the time they are idle, just waiting for a graphically intense program to run. In the past, it has taken a bit of effort to take advantage …

Read Full Post Posted in Tune | Tagged , , | Leave a comment

5 Tips for Speeding Up Your Code Using TBB

Intel Threading Building Blocks is a library to help your application use parallelization without you having learn all the subtleties of threading and to avoid the pitfalls. If you use TBB to speed up your software, below are a few tips to help you on your way. You don’t always …

Read Full Post Posted in Tune | Tagged , , | Leave a comment

OpenMP vs. MPI: Which Is Better?

Unlike SIMD, where vectorization has multiple data items affected by a single instruction, in SPMD (Single Program Multiple Data) multiple processors run the same program but operate on different subsets of the data. There are two different ways that you can implement SPMD which depend upon the underlying architecture. If …

Read Full Post Posted in Tune | Tagged , | Leave a comment

Intel HPC Code Modernization Road Show Teaches Tuning for Xeon Phi

As multicore processors such as Xeon and Xeon Phi expand their presence in both on-premises and Cloud infrastructure, the need to tune your code to get the most bang for your processor buck becomes increasingly important. To that end, Intel is hosting a developer conference series that offers insights into …

Read Full Post Posted in Tune | Tagged , , , | Leave a comment

Parallel Computing in the Age of Big Data

The age of Big Data is here. With zettabytes of daily Internet traffic, analysis of even a small subset involves Big Data. The age old paradigm of sequential programming no longer suffices, it is essential to move to a parallel programming mode to service the needs of Big Data applications. …

Read Full Post Posted in Tune | Tagged , , | Leave a comment

Unintended Consequences of Parallel Computing: How to Avoid Pitfalls

In this blog, we’ll take an in-depth look at parallel computing, and examine some of the pitfalls that can thwart efforts for great code to create applications that sing. Parallel Is Not Always Faster Than Sequential There are times when changing code from sequential to parallel will slow it down. …

Read Full Post Posted in Tune | Tagged , , , | Leave a comment

Intel’s James Reinders on Multicore, How to Future-Proof Your Code

Is your business getting the most out of all the cores in your processors? When you consider the rapid advances in today’s Xeon Phi coprocessors, the chances are much of your code would benefit from modernization. In this video, Intel Chief Evangelist James Reinders speaks to Colfax HPC research engineer …

Read Full Post Posted in Tune | Tagged , , | Leave a comment