gridMathematica is a parallel technical-computing system for science, engineering, finance and business analysis including a collection of algorithms in an integrated system optimised for multi-processor machines, clusters, grids and super computers. Providing a network-managed pool of 16 computation kernels, gridMathematica can be shared by a group of Mathematica users locally and can run on remote hardware to combine the power of multiple computers.

It comprises computational kernels: one manager kernel and a pool of worker kernels, that work as a cohesive unit, co-ordinated by the manager and communicating via MathLink technology. It scales from an ad hoc collection of PCs running overnight jobs to a dedicated departmental cluster or an enterprise-wide super computer. Broad platform support and 64-bit optimisations andsupport for Gigabit and high-speed data networks are provided. It integrates seamlessly with many cluster management systems via the Cluster Integration Package supporting Altair PBS Professional, Microsoft Windows Compute Cluster Server, Platform LSF and Sun Grid Engine.

Every copy of Mathematica includes the capability for instant parallel computing at no additional charge. In single-machine configurations, Mathematica includes the ability to compute across four local processor cores and can be extended to make use of eight or more cores with the purchase of Mathematica Core Extension.

gridMathematica is an integrated extension system for increasing the power of Mathematica licences. Each gridMathematica Server gives Mathematica users a shared pool of 16 additional network-enabled Mathematica computation kernels for running distributed parallel computations over multiple CPUs.

Users do not need to change their existing parallel code, they are just required to make gridMathematica Server available, and parallel programs can automatically use the additional CPU power.

Features

• Grid deployment of all of Mathematica's functionality: including fast numerical routines, image processing, statistics, and finance capabilities. It supports remote access to GPUs and the distributed on-the-fly generation and compilation of parallel C code.
• A high-level parallel programming language: automates much of the communication, synchronisation, data transfer, and error recovery that usually makes grid computing so difficult to set up. With automatically serialised data transfer, users can send any structured data and programs to remote machines without needing to configure a common file system.
• Support for HPC standards, including cluster managements systems: Altair PBS Professional, Microsoft Windows Compute Cluster Server, Microsoft HPC Server, Platform LSF, and Sun Grid Engine and includes Wolfram Lightweight Grid software for systems without cluster management. Gigabit and high-speed networks are supported, and both CUDA and OpenCL GPU hardware are supported.
• Local hardware: all additional kernels run on each user's own machine
• High performance cluster: a dedicated cluster of machines running a specialised grid-management tool
• Ad hoc grids: a network of different machines available for parallel computation.

gridMathematica provides innovative parallel computing functionality, with features such as the following:

• Parallelisation at the Mathematica language level
• Support for multiprocessor machines, clusters, and grids
• Machine-independence - user code is completely portable
• High-performance MathLink communication protocol optimised for all common configurations
• Scheduling takes processor speed and communication latency into account
• User-programmable scheduling for problem-specific adaptation
• Automatic failure recovery and reassignment of stranded processes
• Support for tracing and debugging
• Speculative parallelisation for non-deterministic problems
• Parallel applications can be simulated and tested on a personal computer

gridMathematica supports all Mathematica functionality, including the collection of advanced algorithms for numeric and symbolic computation, discrete mathematics, statistics, data analysis, graphics, visualisation, and general programming, as well as all Mathematica application packages.

 Licensing
Each gridMathematica unit comes with one controlling kernel process, eight concurrent worker kernel processes, one front-end process and one Parallel Computing Toolkit. One MathLM licence manager is also provided for each cluster or grid. Any number of units can be combined for larger grids.

Hardware Specifications:

• Processor: Intel Pentium III 650 MHz or equivalent
• Disk Space: 4GB
• System Memory (RAM): 512MB required; 1GB+ recommended

Platform Availability:

gridMathematica is available for all common Linux and Unix systems, Windows, and Mac OS X. For more information, see the supported platforms for Mathematica. Computers in the cluster do not have to run the same operating system. For a cluster consisting of separate computers, the machines need to be able to communicate with each other over TCP/IP.

Windows:

• Windows 7
• Windows Vista
• Windows XP*
• Windows HPC Server 2008
• Windows Server 2008
• Windows Server 2003
• Windows Compute Cluster Server 2003

Mac:

• Mac OS X 10.6 Intel
• Mac OS X 10.5 Intel

Linux:

• Ubuntu 7-10
• Red Hat Enterprise Linux 4
• CentOS 5
• Debian 5
• openSUSE 11

Networks

• Required: TCP/IP
• Supported: Ethernet, Gigabit Ethernet, and Firewire (IEEE 1394)
• Also supported: Myrinet, Infiniband, Quadratrics

gridMathematica uses standard TCP/IP to communicate between nodes, but it is independent of the actual network infrastructure. As a result, gridMathematica can take advantage of infrastructures such as Myrinet and fiber-based networks to increase bandwidth and decrease latency during high-communication computations over the more common Ethernet and copper-based networks.

Other Requirements
You need to be able to perform remote execution to launch Mathematica on computers in the cluster. Standard ways of doing this are through RSH, REXEC, or SSH.

Components
Mathematica, Parallel Computing Toolkit, MathLM