Intel® Many Integrated Core Architecture

different performance on nativ and offload modes


  • host: Xeon E5-2690, 2,9 GHz, 64 Gb
  • mic: Intel Xeon Phi 7110X, 1,1 GHz, 8 Gb
    • OS Version                      : 2.6.32-220.el6.x86_64
    • Driver Version                 : 5889-16
    • MPSS Version                 : 2.1.5889-16
    • Host Physical Memory    : 65868 MB
  • ICC: 14.0.1 (composer_xe_2013_sp1.1.106

I took the test from this subject ( and modified it to run in offload mode.

OpenMP 4.0 Fortran -> $omp target map(to:x) Does not copy scalars

Hello. Ever since I tried switching over to openMP 4.0 from LEO my code was not working properly. Finally figured out that during the offload transfer the map clause copies arrays but DOES NOT COPY SCALARS. 

The only way to get scalars copied to the device is to call target update to(scalar)

Can someone please explain this behaviour?

Using icc 15.0.1

EDIT: Scalars created in the same file are copied, but scalars imported from a different file using USE do not get copied. Arrays get copied no matter where they originate from.

Better Concurrency and SIMD On The HIROMB‐BOOS­‐Model 3D Ocean Code

By utilizing the strengths of the Intel® Xeon Phi™ coprocessor, the  chapter 3 High Performance Parallelism Pearls authors were able to improve and modernize their code and “achieve great scaling, vectorization, bandwidth utilization and performance/watt”. The authors (Jacob Weismann Poulsen, Karthik Raman and Per Berg) note, “The thinking process and techniques used in this chapter have wide applicability: focus on data locality and then apply threading and vectorization techniques.”.

  • Developers
  • Code Modernization
  • Intel® Many Integrated Core Architecture
  • Parallel Computing
  • Vectorization
  • Webinar: IDF LIVE - Parallel Programming Pearls

    Unable to join us at the Intel Developer Forum in San Francisco this August? We have you covered. This session dives into real-world parallel programming optimization examples, from around the world, through the eyes and wit of enthusiast, author, editor and evangelist James Reinders.

    When: Wed, Aug 19, 2015 11:00 AM - 12:00 PM PDT

    running mpi program in mic


           I have compile OpenFOAM for mic architecture. I am able to run the program without any trouble. But I am having trouble with running openfoam in parallel. The way OpenFOAM works in parallel is you decompose the mesh in to sub domains and put the folders in to processor 1 2 etc folders 

    But when I issue the command to run the program in parallel, it is creating two seperate instances of the same program.

    export I_MPI_MIC=1

    mpiexec.hydra -np 2 ./pisoFoam

    I am adding some of the first lines of output

    Optimizing Legacy Molecular Dynamics Software with Directive-based Offload

    Directive-based programming models are one solution for exploiting many-core coprocessors to increase simulation rates in molecular dynamics. They offer the potential to reduce code complexity with offload models that can selectively target computations to run on the CPU, the coprocessor, or both. In this paper, we describe modifications to the LAMMPS molecular dynamics code to enable concurrent calculations on a CPU and coprocessor. We demonstrate that standard molecular dynamics algorithms can run efficiently on both the CPU and an x86-based coprocessor using the same subroutines.

  • Developers
  • Linux*
  • lammps
  • molecular dynamics
  • many-core
  • Code Modernization
  • Intel® Many Integrated Core Architecture
  • Xeon Phi 5110P on Dell Precision T3600 Workstation installation


    has anybody successfully installed a Xeon Phi 5110P on a Dell Precision T3600 Workstation?

    My card is in a PCIe v3 x16 slot but it doesn’t show up in BIOS or “lspci”. Blue LEDs on the card are blinking irregularly, 6pin and 8pin PCIe power connectors are plugged in (my first error ;-). I flashed the BIOS to revision A14, so that I was able to set the “PCI MMIO Space Size” BIOS Parameter to "Large". I also removed my Nvidia Tesla C2075 from the board, because together they took more power than the system can provide (my second error ;-)

    Using micnativeloadex, how to use args ?


    I am a very beginner with the Intel Phi, and I am trying to do something that is maybe not possible.

    I have this binary file called upcDemo

    And I run it this way: upcrun -n 12 upcDemo 

    (This will run the program on 12 threads)

    I have tried many syntax with micnativeloadex...but I got errors, here is what I tried:

    Subscribe to Intel® Many Integrated Core Architecture