The following post provides an introduction to the use of OpenCL™ for programming Parallella. What is OpenCL? OpenCL is an industry standard API for parallel programming co-processors on heterogeneous platforms. Designed primarily for computing with general-purpose graphics processing units (GPUs), the API may be used to access the compute capability of other types of devices including multicore CPUs and other…
Parallella is much more than simply a hardware project, and software that is able to harness the power of massively parallel systems is going to be crucial to achieving our goal of making parallel computing accessible to everyone. This includes everything from languages and frameworks that make such systems easier to program, through libraries and toolkits that provide the building…
Today we completed the publication of the last remaining Epiphany SDK source tree repositories to the Adapteva GitHub account, including all the drivers and library files. The two repositories can be found at: http://github.com/adapteva epiphany-sdk This repo is the master archive that integrates the various Epiphany-SDK components into a release/installation tree. It contains the master eSDK build script (build-sdk.sh) and…
[NOTE: this is a community proposal and not part of the official Parallella hardware architecture!] Introduction We have a real opportunity to build a thriving expansion board ecosystem for Parallella, with the potential to learn from those that went before it and to advance the state of the art. For this reason I have spent some time thinking about how…
Now that the Parallella board design is finally stabilizing we are ready to release rev-zero of the Parallella Reference Manual (PDF). This document should be considered an initial rough draft and there are bound to be some typos and mistakes (only a few eyes have reviewed it so far). We felt it was important to get the information out to…
The screen shot above shows the latest Parallella board placement (after three major redesigns) and we are now within striking distance of meeting the aggressive project design and cost constraints with a set of specifications that we feel really good about. The expressions “Fast, Good or Cheap. Pick Two” has been our constant companion over the last three months. At…
FMC daughter card complete with an Epiphany IV 64-core chip In a post on Adepteva’s website on Tuesday Roman announced that they had completed integration of the Epiphany IV 64-core chip with the Parallella prototyping platform. Explaining how 16 and 64-core packages had been designed to be virtually pin-compatible, allowing the same PCB to be used with only minor changes…
We at Erlang Solutions are very excited about our partnership with the Parallella Project and have formed a small task force to explore what can be done to make it easier to utilise the multicore processing power Parallela board has within Erlang. In this post, Torben (a.k.a The Erlang Priest) will provide a brief overview of the language and explain…
Over the last few months we have received numerous questions regarding the makeup of the Parallella community. Questions that often come up include: “What will the Parallella be used for?” and “Who is buying the Parallella?” Instead of providing everyone with a few simple generic conclusions, we figured it might be helpful to present the raw data compiled from the…
Earlier this week the first reports of Parallella in the wild started to come in, with ubii and ed2k posting details of initial experiences to the forum, and Yaniv quickly putting together a getting started guide for the prototype boards. Looking forward to hearing further details as the remaining prototypes arrive! Over on the Adepteva website Roman posted a guide…