Most Recent Insights Posts RSS

Increase Performance Potential of Your Apps

Vectorization plays a huge role in speeding up programs and Intel Cilk Plus allows a developer to do explicit vectorization using Intel Cilk Plus SIMD directive, Array Notation and SIMD-enabled functions. Find out more about this technology and see a case study of Array Notation applied to Discrete Cosine Transform …

Read Full Post Posted in Insights | Leave a comment

Parallelization: Finding Frustration-Free Examples

Sooner or later, most programming experts will find themselves having to teach concepts to other programmers, usually informally between coworkers. That includes parallel programming, which is difficult to learn, let alone teach to others. Jeff Cogswell shares some of his experiences. When you spend a lot of years doing any …

Read Full Post Posted in Build | Leave a comment

Cilk Plus: Vectorized or Multicore or Both?

A common focus for Cilk Plus is the multicore aspect. But Cilk Plus loops can also be vectorized. Jeff Cogswell explores how these fit together. Over the past couple of years, I’ve covered a lot about vectorization, and Cilk Plus programming. If you’ve turned on the diagnostics for the compiler, …

Read Full Post Posted in Design | Leave a comment

Xeon Phi vs. GPU Programming: Which Is Better?

While covering the Xeon Phi MIC architecture, the question of how this compares to targeting the high-core GPUs available sometimes comes up. Jeff Cogswell tackles the question and weighs the tradeoffs. A reader recently contacted me and asked the difference between programming for a GPU and programming for the Xeon …

Read Full Post Posted in Verify | Leave a comment

Best Ways to Learn Xeon Phi Programming

There are many different places to learn about Xeon Phi coprocessor programming, starting with introductions to parallel programming, on up to advanced topics. Jeff Cogswell explores a few of them here.

Read Full Post Posted in Insights | Leave a comment

Taking Multicore to the Extreme

While exploring multicore programming, you can accomplish a great deal with a quad-core processor. But what if you need even higher performance? There’s always the Xeon Phi coprocessor. In this blog, Jeff Cogswell explains how the Xeon Phi fits together with the main processor on your computer. When you understand …

Read Full Post Posted in Tune | Leave a comment

Exploring the Different Parts of Intel Parallel Studio XE

In this video, Jeff Cogswell presents a high-level overview of the different parts of Parallel Studio XE and how to find them in Visual Studio.

Read Full Post Posted in Insights | Leave a comment

3 Easy Ways to Target Vectorization Tech with C++ Compiler

The Intel C++ compiler can easily target different processors. To really know what’s happening, you need to look at the assembly code. But before you can do that, you have to make sure you configure the compiler correctly. In this blog, Jeff Cogswell talks about the three ways you can …

Read Full Post Posted in Design | Leave a comment

Intel’s 256-Node Network-on-a-Chip Runs Like a Cluster

Designed to make high-performance multicore processors more powerful and efficient, Intel Labs has revealed a “network-on-a-chip” that provides an interconnect fabric for future multicore chips. This article details Intel’s design as well as a new ultra-efficient graphics chip based on the company’s newest 22nm tri-gate system-on-a-chip. Read More!

Read Full Post Posted in Insights | Leave a comment

Introducing the Intel Math Kernel Library

In this video, Jeff Cogswell introduces the Intel Math Kernel Library (MKL) and provides a high-level overview of the different Mathematics features available in it.

Read Full Post Posted in Insights | Leave a comment