Most Recent Tune Posts RSS



tbbwebinar

Get More From Heterogeneous Computing with Threading Building Blocks (TBB)

TBB, Meet Heterogeneous Computing Computing platforms are becoming increasingly heterogeneous with the combination of CPUs, GPUs, FPGAs, co-processors, and more, but programming to take advantage of these heterogeneous environments in a single application remains a challenge. Watch this webinar to hear Jackson Marusarz of the Intel Developer Product Division present …

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

Using Intel Performance Primitives (IPP) for Fast Fourier Transforms (FFTs)

Video Presentation shows how easy Fourier Transforms are, and how fast they can execute Fourier Transforms are difficult to do, and the code usually takes a lot of time to execute. But Intel’s Performance Primitives take away the hassle, according to this video by Slashdot Media Contributing Editor Rick Leinecker

Read Full Post Posted in Tune | Tagged , | Leave a comment
post10141604

Vectorization For The Masses

ingle instruction multiple data (SIMD) vectorization drives performance gains Clock speeds have been fairly constant since around 2005. This fact begs the question: how have we experienced such significant performance improvements since then? Most people think that performance improvements rely solely on increasing clock speeds. But that is not necessarily …

Read Full Post Posted in Tune | Tagged , | Leave a comment
colfax2

Performance Optimization for Xeon Phi: Video Workshop

  Colfax Research 2 hour hands-on-workshop offers best practices, demonstrates techniques As more organization look to utilize computing resources and big data to become more competitive, development teams are being tasked with adopting machine learning tools at an increasing pace. With that in mind, Intel partner Colfax Research has developed …

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

Using OpenMP to Fine Tune Vectorization

Adopting OpenMP can have significant payoffs, for vectorization and more Everyone wants their programs to execute fast and smooth. For instance, Microsoft Word does complex image manipulation easily without noticeable delay. The march of software toward even greater levels of performance helps satisfy our need for speed. This blog talks …

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

OpenMP: Past the Basics

So, you’ve started tinkering with OpenMP to help parallelize your code. Now what?  This video by Slashdot Media Contributing Editor Rick Leinecker points OpenMP newbies in the right direction to go beyond parallelizing for loops, and demonstrates how to avoid data race conditions while you’re doing it.  

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

Parallelizing Binary Searches

Binary searches are orders of magnitude faster than linear searches. In this tutorial, Slashdot Media Contributing Editor Rick Leinecker shows how to speed all manner of searches in your code by paralellizing using OpenMP

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

Avoiding Data Races with Reducers

Fixing what happens when simultaneous threads modify the SAME memory.

Parallelizing code drastically speeds execution, but if you’re not careful data race conditions can produce very unwanted results. See how Slashdot Media Contributing Editor Rick Leinecker gets at the problem and fixes it automatically with a reduction clause.

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

Using Cilk Plus to Speed Parallel Code

Move Over, OpenMP: Cilk Plus struts it stuff. What’s the best way to parallelize for loops? In this blog, Slashdot Media Contributing Editor Rick Leinecker looks at Cilk Plus as an alternative to OpenMP Using Cilk Plus Rick Leinecker, August 2016 Many of my blogs that talk about parallelization use …

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

Timing C++ Execution: Do Compilers Matter?

Execution speed can be impacted by choice of language AND compiler In this instructional video, Slashdot Media Contributing Editor David Bolton show us how to get the most performance out of C++ code by paying attention to compiler features that are in tune with hardware features.

Read Full Post Posted in Tune | Tagged , | 1 Comment