Get Link | Sync TOC | << | >>
User's Guide for Intel® Math Kernel Library 11.1 Update 3 for Linux* OS
Legal Information
Getting Help and Support
Introducing the Intel® Math Kernel Library
What's New
Notational Conventions
Related Information
Getting Started
Checking Your Installation
Setting Environment Variables
Scripts to Set Environment Variables
Automating the Process
Compiler Support
Using Code Examples
Before You Begin Using the Intel® Math Kernel Library
Structure of the Intel® Math Kernel Library
Architecture Support
High-level Directory Structure
Layered Model Concept
Contents of the Documentation Directories
Linking Your Application with the Intel® Math Kernel Library
Linking Quick Start
Using the -mkl Compiler Option
Using the Single Dynamic Library
Selecting Libraries to Link with
Using the Link-line Advisor
Using the Command-line Link Tool
Linking Examples
Linking on IA-32 Architecture Systems
Linking on Intel® 64 Architecture Systems
Linking in Detail
Listing Libraries on a Link Line
Dynamically Selecting the Interface and Threading Layer
Linking with Interface Libraries
Using the ILP64 Interface vs. LP64 Interface
Using the SP2DP Interface
Linking with Fortran 95 Interface Libraries
Linking with Threading Libraries
Sequential Mode of the Library
Selecting the Threading Layer
Linking with Computational Libraries
Linking with Compiler Run-time Libraries
Linking with System Libraries
Building Custom Shared Objects
Using the Custom Shared Object Builder
Composing a List of Functions
Specifying Function Names
Distributing Your Custom Shared Object
Managing Performance and Memory
Improving Performance with Threading
Threaded Functions and Problems
Avoiding Conflicts in the Execution Environment
Techniques to Set the Number of Threads
Setting the Number of Threads Using an OpenMP* Environment Variable
Changing the Number of Threads at Run Time
Using Additional Threading Control
Intel MKL-specific Environment Variables for Threading Control
MKL_DYNAMIC
MKL_DOMAIN_NUM_THREADS
Setting Environment Variables for Threading Control
Calling Intel MKL Functions from Multi-threaded Applications
Other Tips and Techniques to Improve Performance
Coding Techniques
Hardware Configuration Tips
Managing Multi-core Performance
Operating on Denormals
FFT Optimized Radices
Enabling Low-communication Algorithm in Cluster FFT
Using Memory Functions
Memory Leaks in Intel MKL
Redefining Memory Functions
Language-specific Usage Options
Using Language-Specific Interfaces with Intel® Math Kernel Library
Interface Libraries and Modules
Fortran 95 Interfaces to LAPACK and BLAS
Compiler-dependent Functions and Fortran 90 Modules
Mixed-language Programming with Intel MKL
Calling LAPACK, BLAS, and CBLAS Routines from C/C++ Language Environments
Using Complex Types in C/C++
Calling BLAS Functions that Return the Complex Values in C/C++ Code
Obtaining Numerically Reproducible Results
Getting Started with Conditional Numerical Reproducibility
Specifying the Code Branches
Reproducibility Conditions
Setting the Environment Variable for Conditional Numerical Reproducibility
Code Examples
Coding Tips
Example of Data Alignment
Using Predefined Preprocessor Symbols for Version-dependent Compilation
Working with the Cluster Software
Linking with ScaLAPACK and Cluster FFTs
Setting the Number of Threads
Using Shared Libraries
Building ScaLAPACK Tests
Examples of Linking with ScaLAPACK and Cluster FFT
Examples for Linking a C Application
Examples for Linking a Fortran Application
Using Intel® Math Kernel Library on Intel® Xeon Phi™ Coprocessors
Automatic Offload
Automatic Offload Controls
Setting Environment Variables for Automatic Offload
Compiler Assisted Offload
Examples of Compiler Assisted Offload
Linking for Compiler Assisted Offload
Using Automatic Offload and Compiler Assisted Offload in One Application
Running Intel MKL on an Intel Xeon Phi Coprocessor in Native Mode
Using ScaLAPACK and Cluster FFT on Intel Xeon Phi Coprocessors
Examples of Linking with ScaLAPACK and Cluster FFT for Intel® Many Integrated Core Architecture
Examples of Linking a C Application
Examples of Linking a Fortran Application
Threading Behavior of Intel MKL on Intel MIC Architecture
Improving Performance on Intel Xeon Phi Coprocessors
Configuring Your Integrated Development Environment to Link with Intel® Math Kernel Library
Configuring the Eclipse* IDE CDT to Link with Intel® MKL
LINPACK and MP LINPACK Benchmarks
Intel® Optimized LINPACK Benchmark for Linux* OS
Contents
Running the Software
Known Limitations
Intel® Optimized MP LINPACK Benchmark for Clusters
Overview
Usage Modes for Intel® Xeon Phi™ Coprocessors
Contents
Compile Options Specific to the Intel Optimized MP LINPACK Benchmark
Building the Intel Optimized MP LINPACK Benchmark
Ease-of-use Command-line Parameters
Heterogeneous Intel Optimized MP LINPACK Benchmark
Running the Intel Optimized MP LINPACK Benchmark
Running the Benchmark on One Node
Expanding the Benchmark to Two or More Nodes
Running Hybrid Offload Binaries
Improving Performance of Your Cluster
More Details of the Intel Optimized MP LINPACK Benchmark
Configuring the Hybrid Offload Version of the Intel Optimized MP LINPACK Benchmark
Environment Variables for the Hybrid Offload
Optimizing the Result on a Cluster
Options to Reduce Search Time
Appendix A: Intel® Math Kernel Library Language Interfaces Support
Language Interfaces Support, by Function Domain
Include Files
Appendix B: Support for Third-Party Interfaces
FFTW Interface Support
Appendix C: Directory Structure In Detail
Detailed Structure of the IA-32 Architecture Directories
Static Libraries
Dynamic Libraries
Detailed Structure of the Intel® 64 Architecture Directories
Static Libraries
Dynamic Libraries
Detailed Directory Structure of the lib/mic Directory