-
- Hyper-threading: Perfect for Neural NetworksArtificial neural networks (ANNs) are used today to learn solutions to parallel processing problems that have proved impossible to solve using...
- NCSA Scientist Backs MICs over GPUsWhile the massively parallel Xeon Phi coprocessor faces supercomputers leveraging Nvidia’s graphic processing units (GPUs), Intel’s many-integrated-core (MIC) architecture will prevail,...
- Haswell Architectural Ups Parallelism AnteIntel has revealed new architectural details about the Haswell micro-architecture and its support for parallel processing. The faster, lower power...
- How Xeon Phi Stacks Up to GPUsXeon Phi lead architect George Chrysos presented comparisons between using Xeon Phi co-processors instead of graphics-processor units (GPUs) at the recent...
- IDF Debuts Parallel Web Apps for 2013Parallel applications will extend their reach to the world wide web in 2013, according to Renée James, senior vice president...
- Networking Fabric Coming to XeonIntel has revealed plans to integrate a network fabric controller into its Xeon microprocessors within the next few years. On Sept....
Michael McCool Interview: Making the Best Use of Intel Compilers
Trends in Parallel Software Development: A Conversation with Intel’s James Reinders
Maximizing Hardware with Parallel Programming
Perceptual Computing: Human Interaction as the Next-Generation User Interface
Stephen Blair-Chappell Interview Part 3: Case studies of Sudoku, star simulation, CERN collider, rewriting Dhrystone
Stephen Blair-Chappell Interview Part 2: Challenges of legacy code, importance of understanding the coding environment
Case Studies
Evaluation Guides
Tech Documents
Tutorials
Videos
White Papers
- Hyper-threading: Perfect for Neural Networks
-
-
Most Recent Posts Blog Archive
Using the Timeline in Intel® VTune™ Amplifier XE
This 10-minute video shows the features of the timeline in Intel VTune Amplifier XE, and gives examples of how to use it to understand your application’s performance.
Leave a comment
Read Full Post
Posted in Tune
Load Balance and Parallel Performance
Load balancing an application workload among threads is critical to performance. The key objective for load balancing is to minimize idle time on threads. Sharing the workload equally across all threads with minimal work sharing overheads results in fewer cycles wasted with idle threads not advancing the computation, and thereby …
Read Full Post
Posted in Tune
Smart Framework Enables Internet-of-Things
A new architectural framework for embedded devices on the quickly growing Internet of Things has been released by Intel with its McAfee and Wind River subsidiaries. The Intelligent Systems Framework (ISF) aims to facilitate easier coordinated use of multiple Atom, Core and Xeon processors in distributed embedded systems. ISF works with …
Read Full Post
Posted in Insights
Hyper-threading: Perfect for Neural Networks
Artificial neural networks (ANNs) are used today to learn solutions to parallel processing problems that have proved impossible to solve using conventional algorithms. From cloud-based, voice-driven apps like Apple’s Siri to realtime knowledge mining apps like IBM’s Watson to gaming apps like Electronic Arts’ SimCity, ANNs are powering voice-recognition, pattern-classification …
Read Full Post
Posted in News
Optimize Your Code for the Latest Intel® Xeon® Processors and Intel® Xeon Phi™ Coprocessor
Learn how and why using Intel® Parallel Studio XE for Linux* offers developers a powerful pathway to programming success. View presentation here.
Read Full Post
Posted in Design
Delivering the Architectural Foundation for Big Data
It’s easy to forget the profound ways that parallel programming can be a key component in solving huge challenges such as visualization and law enforcement. In this big-picture technology insight session from the recent IDF 2012 in San Francisco you’ll take a step back and see how and why Intel helps …
Read Full Post
Posted in Insights
Universal Parallel Computing Research Centers
Intel and Microsoft jointly created the Universal Parallel Computing Research Center (UPCRC) program. UPCRC illustrates some of the features of an ideal URO program: a clear research goal and agenda, the potential for significant strategic impact on Intel and the industry, a long range research horizon, a multi-year funding commitment, …
Read Full Post
Posted in Insights
Bringing Parallelism to the Web
Want a quick understanding on a proven pathway? Download this PDF from the recent Intel IDF in San Francisco. You’ll get a good introduction to parallel programming on the web and a tutorial on Rivertrail parallel extensions for Javascript API. Download presentation here.
Read Full Post
Posted in Build
Threading Building Blocks
In response to some of the emails I’ve received we’ll talk this week about some necessary background in using the Threading Building Blocks. In particular, including Threading Building Blocks header files and whether you need to link to the .lib files. You always need the header files for the code …
Read Full Post
Posted in Build
Related Links
Case Studies
- Bringing Magic to Life- Exocortex and Intel Technologies Work with VFX Studios Gradient Effects and Will Garrett to Achieve Movie Magic
- Massachusetts General Hospital* Achieves 20X Faster Colonoscopy Screening Processing Time
- Designing Quiet Products Faster
- Nik Software Increases Rendering Speed of HDR by 1.3x
Evaluation Guides
- A Guide to Vectorization with Intel® C++ Compilers
- Intel® Cilk™ Plus
- Intel® Threading Building Blocks (Intel® TBB)
- Resolve Resource Leaks
- Add Parallelism
- Boost Performance
- Eliminate Threading Errors
- Eliminate Memory Errors
- Improve Fortran Code Quality with Static Security Analysis (SSA)
- Improve C++ Code Quality with Static Security Analysis (SSA)
Tech Documents
- The Parallel Universe — Issue 11– September 2012
- The Parallel Universe – Issue 10 – June 2012
- The Parallel Universe
- Estimating FLOPS using Event Based Sampling (EBS)
- New Rules for Array Sections in Intel® Cilk™ Plus
- Using the new concurrent priority queue in Intel® Threading Building Blocks
- Writing and Optimizing Parallel Programs — A complete example
Tutorials
- An Introduction to Vectorization with the Intel® C++ Compiler
- An Introduction to Vectorization with the Intel® Fortran Compiler
- How to manage the analysis overhead of Correctness Modeling in Intel® Parallel Advisor 2011
- Using Intel® TBB in network applications: Network Router emulator
- Intel® C++ Composer XE 2011 Getting Started Tutorials
Videos
- Don't Dread Threads Part 1
- Three Things You must Teach: Module 3. Programming with OpenMP (pt.1)
- Three Things You must Teach: Module 2. Shared Memory and Threads (pt.1)"
- Three Things You must Teach: Module 1. Recognizing Potential Parallelism (pt.2)"
- Three Things You must Teach: Module 1. Recognizing Potential Parallelism"
White Papers
- Market Report- The State of Software Development Today: A Parallel View
- The ROI from Optimizing Software Performance with Intel® Parallel Studio XE for Development Organizations
- Two Tools Measure the Performance Scalability of Your Application
- Intel® Compilers for Linux* – Compatibility with GNU Compilers
- Optimizing VLife* Molecular Design Suite Using Intel® Parallel Studio XE