| |
|
|
|
|
|
|
 |
|
Requirements |
|
Platform |
|
 |
| |
|
|
|
|
|
|
CT Parallel Environment, CPE |
|
Hardware |
|
Operating System |
|
|
Free download of evaluation copy of its mainsteam product Clustertech Parallel Environment.
(CPE) with a 4-CPU-core parallel license for a 90-day evaluation period... |
|
| • |
CPU: Intel x86 (32-bit /64-bit) |
| • |
Memory: 512mb or above |
|
|
| • |
Red Hat Linux Enterpise 4 (recommend) |
| • |
Centos Linux 4 (recommend) |
| • |
AIX for HPC |
|
|
|
| |
|
Platform |
|
|
|
|
| |
|
| Optional: |
| • |
SSH |
| • |
Torque or PBS Pro |
|
|
|
|
|
|
|
|
|
|
|
|
 |
|
|
Prerequisite |
|
Compiler |
|
|
Combining object-oriented programming and performance math libraries.
Features
| • |
Elegant object-oriented interface. Users write formulae-like C++ code statements. |
| • |
Transparent BLAS/LAPACK integration. Delegation to highly-optimized BLAS and LAPACK libraries automatically. |
| • |
Performance. Extensive optimization techniques ensure high performance. |
Fields of Application
| • |
Scientific and engineering computations. |
| • |
Time-critical linear algebra-related applications. |
Benefits
| • |
Resulting code is highly readable and maintainable. |
| • |
Increased productivity. |
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
| |
|
|
Prerequisite |
|
Compiler |
|
|
The lack of a tight integration of MPI with C++ Standard Template Library hinders developers from writing
efficient parallel programs using message-passing in C++. STL-MPI library fills this gap by providing a
transparent and unified mechanism in C++ to transmit STL containers in an object-oriented and type-safe manner. Features
| • |
Supports STL containers and nested containers for data transport |
| • |
Conforms to the MPI features, naming conventions and semantics |
Benefits
| • |
Facilitates the development of parallel programs in C++ |
| • |
Optimizes for high performance data transport in a cluster |
| • |
Eases porting of conventional MPI C++ applications |
|
|
|
|
| • |
gcc-c++ 3.4 or above |
| • |
Microsoft visual studio .Net 2003 or above |
|
|
 |
| |
|
|
|
|
|
|
| |
CT Parallel Random Numbers Generator (PRNG) |
|
Dependencies |
|
Compiler |
|
|
The CLUSTERTECH Parallel Random Number Generator (CT-PRNG) is based on Mersenne Twister which has a period of 219937-1. It generates multiple independent streams simultaneously across multiple GPU servers and uses a jump-ahead feature to guarantee the quality of the output..
Features
| • |
Supports uniform, Bernoulli, Geometric, Exponential, Normal and Lognormal distributions. |
| • |
Jump-ahead feature in initialization phase to guarantee the quality of the random numbers output. |
|
|
|
|
| • |
gcc-C++ 4.1.2 or above |
| • |
Intel C++ Compiler 11.0 or above |
|
|
|
| |
|
|
|
|
|
|
