Intel Software Network articles Feed http://software.intel.com/en-us/articles/parallel/type/landing-links/ en-us Intel® OpenCL SDK


About Intel® OpenCL SDK

About OpenCL™
OpenCL™ (Open Computing Language) is the first open, royalty-free standard for general-purpose parallel programming of heterogeneous systems. OpenCL provides a uniform programming environment for software developers to write efficient, portable code for client computer systems, high-performance computing servers, and handheld devices using a diverse mix of multi-core CPUs and other parallel processors.

About Intel® OpenCL SDK 1.5
Available for free download, the Intel® OpenCL SDK 1.5 makes it easy for OpenCL developers to design, build, debug, and profile OpenCL applications running on the CPU device.

Intel® OpenCL SDK is the Intel implementation of the OpenCL standard optimized for Intel processors, running on Microsoft* Windows*, and Linux* operating systems. Intel® OpenCL SDK 1.5 is fully conformant with the OpenCL 1.1 specification for the CPU, and with Microsoft* Windows* 7 operating systems

Read More >

OpenCL and the OpenCL logo are trademarks of Apple Inc. used by permission by Khronos.

]]> http://software.intel.com/en-us/articles/opencl-sdk/ Sun, 09 Jan 2011 13:00:00 -0800 http://software.intel.com/en-us/articles/opencl-sdk/#comments http://software.intel.com/en-us/articles/opencl-sdk/ Parallel Programming Visual Computing Media Threading Challenge 2010 Phase 1 - Additional Submitted Entries (Master Level) Threading Challenge 2010 Phase 1 - Entries Submitted (Codes)


Below you will find some of the submitted entries by problem for Phase 1 of the Threading Challenge 2010.  Please feel free to review and join us in the forum dedicated to each problem to discuss.

Master Level

Problem 1 - Hosoya Index             Comment on dedicated forum for Hosoya Index

Entry Submitted by Michael_Uelschen:     Code        Write-up

Entry Submitted by sukhaj:     Write-up

]]> http://software.intel.com/en-us/articles/threading-challenge-2010-phase-1-additional-submitted-entries-master-level/ Wed, 22 Dec 2010 21:00:00 -0800 http://software.intel.com/en-us/articles/threading-challenge-2010-phase-1-additional-submitted-entries-master-level/#comments http://software.intel.com/en-us/articles/threading-challenge-2010-phase-1-additional-submitted-entries-master-level/ Parallel Programming The Difference Between Advisor Survey and Amplifier Hotspots
1. Both tools use the same low-overhead statistical sampling technology
2. Advisor Survey shows time spent in loops in addition to total time while Amplifier Hotspots does not
3. Advisor only provides a top-down call tree to help identify sites for parallelism, whereas Amplifier provides a top-down and bottom-up tree to help tune hot paths as well as individual functions
4. Advisor Survey is integrated into the Advisor Workflow

If you plan on using Advisor to help parallelize your existing application, it is recommended that you use Advisor Survey as the first step.

]]> http://software.intel.com/en-us/articles/the-difference-between-advisor-survey-and-amplifier-hotspots/ Sun, 12 Sep 2010 19:00:00 -0700 http://software.intel.com/en-us/articles/the-difference-between-advisor-survey-and-amplifier-hotspots/#comments http://software.intel.com/en-us/articles/the-difference-between-advisor-survey-and-amplifier-hotspots/ Parallel Programming Intel® Parallel Advisor Intel® Parallel Advisor Knowledge Base Intel® Parallel Amplifier Under the Hood
Hotspots

The first type of analysis, Hotspots, is the most lightweight. Running it only takes about as much time as it takes to run your application normally (outside of Intel® Parallel Amplifier). While your app executes, Intel Parallel Amplifier’s data collector will periodically take samples. For each sample, the collector cooperates with the operating system to interrupt your program and gather data. It records the instruction pointer (IP) for each CPU core that is executing your app, as well as the call stack (saved as part of the O/S’s data structures). Once your application finishes executing, Intel Parallel Amplifier takes all of the IP samples and uses them to figure out how long each function was executing, and it uses all of the call stack samples to create a Call Tree for the whole program. In order to create the hotspots list and the call tree, Intel Parallel Amplifier also uses information about your program’s data and instruction space stored in its program debug database (.pdb file).

Concurrency

The second type of analysis, Concurrency, works in the same way but collects a bit more information. It also records data on the status of each of the application’s threads – running, ready to run, and blocked. Once all the samples have been collected, Intel Parallel Amplifier analyzes the data to determine the percentage of time your application’s threads were active. This is called the Concurrency Level, and it is broken down per function. Ideally, the concurrency level for your app should match the number of processors on the system – this is what Intel Parallel Amplifier calls Fully Utilized.

Locks and Waits

The last type of analysis is called Locks and Waits, and it is the most processor intensive. Your application may take longer to run under Locks and Waits analysis, and here’s why. In addition to collecting the data above, Intel Parallel Amplifier will add instructions to your compiled program. These instructions are placed wherever your program uses threading and synchronization API calls, and their purpose is to measure the wait times elapsed whenever a thread is not active. This timing information is combined with symbol information found in the .pdb file to create a picture of where your application is waiting, and what it is waiting on. Locks and other structures which might cause threads to wait are called Synchronization Objects. After running this analysis Intel Parallel Amplifier will show you a list of these objects along with the wait time for each and the concurrency of your app during the wait.
]]> http://software.intel.com/en-us/articles/parallel-amplifier-under-the-hood/ Wed, 01 Sep 2010 21:00:00 -0700 http://software.intel.com/en-us/articles/parallel-amplifier-under-the-hood/#comments http://software.intel.com/en-us/articles/parallel-amplifier-under-the-hood/ Parallel Programming Tools Intel® Parallel Amplifier Knowledge Base How to use Intel(R) Parallel Amplifier with application compiled with symbols but source files are not available Some users have an EXE file and related DLL files (with corresponding PDB files), but have no sources for EXE or DLLs. Can they use Intel(R) Parallel Amplifier in this situation?

Yes! The Intel(R) Parallel Amplifier is supported on Microsoft* Visual Studio* 2005/2008 - the user should create .sln file, .vcproject file in Solution Explorer.

1. Use wizard to create a new project for existing code

2. Specify project file location and project file name (it could be same as your EXE file)

3. Add existing item (EXE file) to current project

4. Manually create "Debug" folder under project file location.

5. Manually copy all EXE/DLL files and PDB files to the "Debug" folder

6. Ensure that you can run "Debug\Start Without Debugging" in Microsoft* Visual Studio*, then you can run Intel(R)Parallel Amplifier.

]]> http://software.intel.com/en-us/articles/how-to-use-amplifier-with-application-compiled-with-symbols-but-source-files-are-not-available/ Wed, 01 Sep 2010 21:00:00 -0700 http://software.intel.com/en-us/articles/how-to-use-amplifier-with-application-compiled-with-symbols-but-source-files-are-not-available/#comments http://software.intel.com/en-us/articles/how-to-use-amplifier-with-application-compiled-with-symbols-but-source-files-are-not-available/ Parallel Programming Tools Intel® Parallel Amplifier Knowledge Base How to Use the Summary in Intel Parallel Advisor 2011 to Make Informed Decisions
After profiling code with the Advisor's Survey tool, Figure 1 shows the breakdown of time spent in functions and loops in the application.

Survey Report
Figure 1

There are two loops in two different functions that are consuming most of the run time, chainedSeed() and orderedSeed().  Figure 2 shows the code for these functions.

Function Source
Figure 2

Both functions are populating an array in a for loop whose iterations may be parallelizable. Figure 3 shows the functions after adding Advisor Annotations.

Annotated Code
Figure 3
After Advisor Suitability and Correctness modeling tools have been run on the annotated code, the Summary Report (Figure 4) combines the information from both tools, to help users make informed decisions about how to proceed with parallelizing their code.

Summary Report
Figure 4

Figure 4 shows that both sites have the possibility of near-linear speedup on two cores (Maximum Self Gain = 1.99x).  It also shows that the ChainedSeedSite has a correctness issue, while the OrderedSeedSite is free of correctness issues. The suitability and correctness data summarized here aid the decision making process when considering how to move forward with adding parallelism.  Parallelizing both sites would improve speedup similarly. However, it may be less trouble to parallelize the OrderedSeedSite because there are no glaring correctness issues to fix.  Having this program-wide performance and correctness information and using it properly, can greatly reduce wasted time and effort to parallelize code that is very difficult or almost impossible to thread.

1. By installing or copying all or any part of the software components in this site, you agree to the terms of the Intel Sample Source Code License Agreement.
2. By accessing and using the Intel Corporation Web Sites and Materials, you acknowledge and agree to abide by the following Terms of Use.
]]> http://software.intel.com/en-us/articles/how-to-use-the-summary-in-intel-parallel-advisor-2011-to-make-informed-decisions/ Mon, 30 Aug 2010 21:00:00 -0700 http://software.intel.com/en-us/articles/how-to-use-the-summary-in-intel-parallel-advisor-2011-to-make-informed-decisions/#comments http://software.intel.com/en-us/articles/how-to-use-the-summary-in-intel-parallel-advisor-2011-to-make-informed-decisions/ Parallel Programming Intel® Parallel Advisor Intel® Parallel Advisor Knowledge Base How to Find Promising Sites for Parallelism Using Intel® Parallel Advisor 2011 Intel® Parallel Advisor 2011 (Advisor) along with the other Intel® Parallel Studio 2011 tools lay out a multi-step process to aid developers in transitioning their serial code to efficient and correct parallel code. This article will focus on the first step of the process: How to determine where to add parallelism in an application.

Advisor provides an easy to use GUI interface as a plug-in to Microsoft® Visual Studio. The first step in using Advisor is to run the Survey tool, which will help determine where most of the time in the application is being spent. These “Hotspots” are good starting points when deciding where to add parallelism in an application. Figure 1 shows a screenshot of some profile data generated by Advisor on a K-Nearest Neighbors application.

Survey Report

Figure 1

Figure 1 shows that the majority of the program (59.3%) is spent in a function called std::::vector() and 20.9% in a loop in the method called KNN::distance. Navigating to the vector function reveals that it is part of the Standard Template Library (STL). The best return on investment will most likely be to focus on the KNN::distance method as opposed to the STL. With one mouse click, Advisor will automatically navigate to the source code (definition) of this method. Figure 2 shows the breakdown of time spent in KNN::distance.

Distance Function
Figure 2

Looking at this method, it can be seen that a single call should not take very long. It only calculates the distance between two points and it spends quite a bit of time returning the value in addition to the calculation.

If this was all the information that Advisor presented, it may be difficult to continue parallelizing because the bulk of the work doesn’t appear to be very conducive to parallelism. However, the Survey Report provides much more detailed information.

Selected Loop
Figure 3

Figure 3 revels that the KNN::predict method is making calls to distance and the Survey Report also shows that these calls are made in the body of a loop (see the highlighted "loop" line).

This loop identifies another possible site for parallelism that still focuses on the Hotspots of the code. This site appears to be a loop with independent iterations (the execution of one iteration isn’t dependent on a previous iteration), which is a prime candidate for parallelization.

The way Advisor encourages the use of profile information can be very useful in uncovering the not-so-obvious locations where introducing parallelism may greatly improve performance. The Survey Report can present call path information that may reveal better locations to introduce parallelism while still focusing on the Hotspots. The timing breakdowns also reveal interesting characteristics about applications that may not be obvious, such as the heavy overhead of vector operations.

After locating ideal sites for parallelism, the next steps in the Advisor Workflow are to insert Advisor Annotations to gather Suitability and Correctness data about the proposed parallelism to determine the easiest and most efficient way to parallelize an application.

]]> http://software.intel.com/en-us/articles/how-to-find-promising-sites-for-parallelism-using-intel-parallel-advisor-2011/ Sun, 29 Aug 2010 19:00:00 -0700 http://software.intel.com/en-us/articles/how-to-find-promising-sites-for-parallelism-using-intel-parallel-advisor-2011/#comments http://software.intel.com/en-us/articles/how-to-find-promising-sites-for-parallelism-using-intel-parallel-advisor-2011/ Parallel Programming Intel® Parallel Advisor Intel® Parallel Advisor Knowledge Base Intel® Threading Challenge 2010 - Contest Prizes


CONTEST PRIZES:


There should always be a reward for solving a puzzle better than anyone else!    

Breakdown of the prizes available are below but check out our official contest rules for more details and judging and prizes!


PHASE 1 & 2 Individual Problem Prize Breakdown:


The TOP 3 point earners in each problem for each level will win!  You can only participant in one level and one problem at a time!  Entries submitted by an individual to both levels a the same time voids both entries! 


Apprentice Level:                                                                               Master Level:

   

1st - One (1) $50.00 (U.S. Dollars) Visa Gift Card                             1st - One (1) $150.00 (U.S. Dollar) Visa Gift Card

2nd - One (1) $35.00 (U.S. Dollar) Visa Gift Card                              2nd - One (1) $100.00 (U.S. Dollar) Visa Gift Card

3rd - One (1) $25.00 (U.S. Dollar) Visa Gift Card                               3rd - One (1) $50.00 (U.S. Dollar) Visa Gift Card



PHASE 2 Grand Prizes:


The individual problem prize structure, as indicated above for Phase 1, will be the same for Phase 2.  What will be different is your opportunity to win a grand prize by participating in more than one problem in Phase 2. 

Apprentice Level:  The grand prize, (1) an Apple iPOD touch* (valued at approximately $200.00 US dollars), will be awarded to the individual with the highest number of cumulative points from the 3 highest scored problems within phase 2, Apprentice Level.

Master Level: The grand prize, (1) a Laptop (valued at approximately $600.00 US dollars), will be awarded to the individual with the highest number of cumulative points from the 3 highest scored problems within phase 2, Master Level.

So let the games begin.  Choose Your Path In This Year's Threading Challenge and Thread Your Way to these Prizes!


* Other names and brands may be claimed as the property of others

]]> http://software.intel.com/en-us/articles/intel-threading-challenge-2010-contest-prizes/ Mon, 24 May 2010 21:00:00 -0700 http://software.intel.com/en-us/articles/intel-threading-challenge-2010-contest-prizes/#comments http://software.intel.com/en-us/articles/intel-threading-challenge-2010-contest-prizes/ Parallel Programming Intel® Threading Challenge 2010 - Meet the Judges


Back again in 2010 by popular demand:




Dr. Clay Breshears

Dr. Clay Breshears is the author of The Art of Concurrency (O'Reilly, 2009) and has been with Intel since September 2000. He started as a Senior Parallel Application Engineer at the Intel Parallel Applications Center in Champaign, IL, implementing multithreaded and distributed solutions in customer applications. Clay is currently a Course Architect for Intel Innovative Software Education specializing in multi-core and multithreaded programming and training. Clay has served as a judge for the Intel Threading Challenge contests since September 2007.

Before joining Intel, Clay was a Research Scientist at Rice University helping Department of Defense researchers make best use of the latest High Performance Computing (HPC) platforms and resources. Clay received his Ph.D. in Computer Science from the University of Tennessee in 1996, but has been involved with parallel computation and programming for over twenty-five years.


Watch this page for addition of guest judges through-out the contest! ]]> http://software.intel.com/en-us/articles/intel-threading-challenge-2010-meet-the-judges/ Mon, 24 May 2010 21:00:00 -0700 http://software.intel.com/en-us/articles/intel-threading-challenge-2010-meet-the-judges/#comments http://software.intel.com/en-us/articles/intel-threading-challenge-2010-meet-the-judges/ Parallel Programming Intel® Threading Challenge 2010 - Winners

Winners from both Apprentice and Master levels will be announced on this page!


THREADING CHALLENGE 2010 - GRAND PRIZE WINNERS ANNOUNCED!


Congratulations to the Threading Challenge 2010 Grand Prize Winners!


Master Level Grand Prize Winner: 


Dmitriy Vyukov - 650 Points

(Best 3 out of 4 Phase 2 Problem Scores)


Master Grand Prize Award: Laptop Computer



Apprentice Level Grand Prize Winner:

Duncan Hopkins - 390 Points

(Best 3 out of 4 Phase 2 Problem Scores)

Award:  Apple iPod Touch


(View all the Phase 1 Winning Entries and Phase 2 Winning Entries)

Other Phase 1 Entries Submitted:   Master Level and Apprentice Level

Other Phase 2 Entries Submitted    Master Level and Apprentice Level

***********************************************************

THREADING CHALLENGE 2010 - PHASE 2 WINNERS

PHASE 2 PROBLEM 4 WINNERS ANNOUNCED:

Apprentice Level: P2:A4 - Maximal Paths

1st Place:     changqingchen
2nd Place:    Vilya Harvey
3rd Place:    crockercaria

See how this problem, Maximal Paths, was judged from Clay's forum post HERE ...

Master Level: P2:M4 - Betweenness Metric

1st Place:     Dmitriy Vyukov
2nd Place:    Rui Diao
3rd Place:    akki

See how this problem, Betweenness Metric, was judged from Clay's forum post HERE ...


**********************************************************************

PHASE 2:PROBLEM 3 WINNERS ANNOUNCED:

Apprentice Level: P2:A3 - Reduced Deck Pai Gow Poker

1st Place: Vilya Harvey
2nd Place: archie314
3rd Place: changqingchen

See how this problem, Reduced Deck Pai Gow Poker, was judged from Clay's forum post HERE ...


Master Level:  P2:M3 - Cat and Mouse

1st Place:      akki
2nd Place:     DmitriyVyukov
3rd Place:      RuiDiao

See how this problem, Cat and Mouse, was judged from Clay's forum post HERE ...


************************************************************

PHASE 2:PROBLEM 2 WINNERS ANNOUNCED:

Apprentice Level: P2:A2 - Hailstone Numbers

1st Place:    duncanhopkins 
2nd Place:   archie314 
3rd Place:    Rui Diao

See how this problem, Hailstone Numbers, was judged from Clay's forum post HERE ...

Master Level:  P2:M2 - Multi-list

1st Place:      Dmitriy Vyukov
2nd Place:     akki
3rd Place:      gk4v07

See how this problem, Multi-list, was judged from Clay's forum post HERE ...


**********************************************************************

PHASE 2: PROBLEM 1 WINNERS ANNOUNCED

Apprentice Level: P2:A1 - Bonev Sequences

1st Place:    duncanhopkins 
2nd Place:   andim 
3rd Place:    xiangbo_x

See how this problem, Bonev Sequences, was judged from Clay's forum post HERE ...


Master Level:  P2:M1 - Taxi Paths

1st Place:      Rui Diao
2nd Place:     mdma
3rd Place:      akki

See how this problem, Taxi Paths, was judged from Clay's forum post HERE ...


***************************************************************************

THREADING CHALLENGE 2010 - PHASE 1 WINNERS

PROBLEM 2 WINNERS ANNOUNCED:

Apprentice Level:  2 - Prime Array

1st Place:    doraemonok
2nd Place:   archie314
3rd Place:   jbosh

Get more on how this problem was judged from Clay's forum post HERE.....


Master Level:  2 - Skip List 

1st Place:     Dmitriy Vyukov
2nd Place:    mdma
3rd Place:    BradleyKuszmaul

Get more on how this problem was judged from Clay's forum post HERE.....


*********************************************************


PROBLEM 1 WINNERS ANNOUNCED:

Apprentice Level:  1 - Connected Components

1st Place:    pbialas
2nd Place:   jbosh
3rd Place:   groberts82

Get more on how this problem was judged from Clay's forum post HERE.....


Master Level:  1 - Hosoya Index 

1st Place:     Dmitriy Vyukov
2nd Place:    BradleyKuszmaul
3rd Place:    denghui0815

Get more on how this problem was judged from Clay's forum post HERE.....


 

]]> http://software.intel.com/en-us/articles/intel-threading-challenge-2010-winners/ Mon, 24 May 2010 21:00:00 -0700 http://software.intel.com/en-us/articles/intel-threading-challenge-2010-winners/#comments http://software.intel.com/en-us/articles/intel-threading-challenge-2010-winners/ Parallel Programming