Parallel Computing

PARALLEL DO : problem with speed

HI,

it's my very first visit on this forum and english is not my natural language. I am 61, with a long experience of parallel programming but my first time with FORTRAN + OMP.

I have a do loop in a PARALLEL DO clause. This loop is running 1329 times. 1326 iterations are running under 0.002s but 5 of them are 0.01s long and one of it is 0.1s on a WIN7-64 system. Those 6 iterations seems to be 'at random' but near the beginning.

Why ? Can I do something to avoid (prevent ?) the 6 long iterations ?

Thanks in advance

Jean in France

Examining the serialized memory access effect, in multi-threaded softwares

hello everyone,

 

I am working on a multi-threaded video encoder application (x265).

I need to prove that, while increasing the number of threads can improve the total run-time, after certain number of thread (cores), it will cause in insufficient memory resources. that is to say, concurrent memory accesses from different cores, will lead to a queue of request at the DRAM, so the delay from the memory can affect the performance.

1- what do you think is the best method to get these results?

Parameter 5 was incorrect on entry to cblas_dgemm when running Intel example

 

I have attempted compiling and running the example "Multiplying Matrices Using dgemm" in the online MKL tutorial; the URL is https://software.intel.com/en-us/node/529735.  I am using Visual C++ in Visual Studio 2013.

The code will compile successfully in debug mode, but at runtime, the cblas_dgemm(.) function call fails with the following message output to the screen:

"Intel MKL ERROR: Parameter 5 was incorrect on entry to cblas_dgemm."

SGX - presence of Manageability Engine

The SDK guide mentions monotonic counters and trusted time. These are apparently provided by the tae_service library, which communicates with special architectural enclaves.

Such features must surely require hardware support (non-volatile storage at the least). The prior SGX specifications did not mention any such features.

The SDK guide describes a "manageability engine" in a few short sentences:

Strange Errors by reseting mic configuration

Hi all,

 

I have installed mpss 3.3.3 on my Centos 7.2 machine. After rebuilding the kernel-modules, I could install it without problems.

I had another mpss configuration on the machine which caused problems (in terms of, no functioning). So I removed those packages and reinstalled.

When I want to reset the config / init the default config, I get those errors. 

micinfo displays my Xeon Phi and I can access it via ssh.

I just wanted to know it this Errors are somewhat take influence on my system. And how to fix them

 

Yet another C++11 error (RTTI deactivated ?)

Hello,

The minimalist code found here http://en.cppreference.com/w/cpp/utility/functional/function/target cannot be compiled with icpc version 16.0.1 (gcc version 5.0.0 compatibility)

$  icpc -std=c++11 function.cpp                                                                                                 
function.cpp(9): error: class "std::function<int (int, int)>" has no member "target"
      if (arg.target<std::plus<int>>())

Subscribe to Parallel Computing