Announcing the Release of CPE (ClusterTech Parallel Environment) Version 3.0

Announcing the Release of CPE (ClusterTech Parallel Environment) Version 3.0

Hong Kong – 29 December, 2011

Cluster Technology is pleased to announce the release of the ClusterTech Parallel Environment (CPE) version 3.0. This is a major release featuring improved performance and new features.

Enhanced Platform Support

In addition to Linux, CPE now supports clusters running Windows HPC Server. A new CPE wizard is fully integrated with Visual C++, and facilitates the development of Windows-based applications. The command line interface has also been changed to simplify the process of building CPE applications on Linux. Support for both 32-bit and 64-bit application development on Windows and Linux platforms is provided.

Improved Performance of ParaConnect

The new version of ParaConnect has adopted the open-source Boost.Serialization libraries for serialization of structured data objects. Compared to version 2.2, the transfer of large data objects (i.e. over tens of Megabytes) is about 10 times faster. The internal structure and protocol of ParaConnect has also been optimized and invocation of remote methods (i.e. RMI) is about 1.7 times faster than version 2.2. ParaConnect now provides logs to improve system visibility and aid troubleshooting.

New Job Distribution Library

A new library has been introduced to facilitate the distribution of batch of jobs on a cluster. The new job distribution (JD) library has built-in dynamic load balancing and automatic resubmission of jobs upon failure. Together, these features simplify the development of robust parallel solutions.

New (Rewritten) Finite-Difference Library

The finite-difference (FD) library has been completely rewritten to efficiently solve linear homogeneous partial differential equations (PDEs). The library supports Implicit, Explicit, Crank-Nicholson and 4 variants of Alternating Direction Implicit (ADI) schemes (Douglas, Craig-Sneyd, Modified Craig-Sneyd and Hundsdorfer-Verwer). Standard boundary conditions (including Dirichlet, Neumann and second order derivative boundary conditions) and standard discretization schemes (including centered and upwind discretization) are also supported.

For more information, please visit http://www.clustertech.com/products/cpe.html

About Cluster Technology Limited (ClusterTech)
Founded in 2000, Cluster Technology Limited specializes in the provision of advanced computing technology solutions. ClusterTech utilizes Cluster, Grid, Cloud and other high performance computing (HPC) technologies to improve operational efficiency, address large-scale computing demand, develop Business Intelligence, and ensure high availability. ClusterTech, headquartered in Hong Kong, opened branches in Mainland China starting 2003, to provide one-stop advanced computing professional services including pre-sale testing, system design, software and hardware integration and optimization for clients in China. In 2009, ClusterTech was awarded the "Hong Kong Awards for Industries - Technological Achievement Certificate of Merit."

ClusterTech has served the advanced computing needs of over a hundred multinational and national level institutes including:

Banking & Finance: HSBC, Bank of China, DBS (Singapore), Fubon Bank (Hong Kong, Taiwan), ING (North America), AMTD Financial Planning Ltd., Guangdong Investment Limited, Pan Asian Mortgage, etc.
Government: The Chinese National Marine Data & Information Service, Chinese Meteorological Administration, China Earthquake Administration, 14 institutes of the Chinese Academy of Sciences, Environmental Protection Department of the Government of the HKSAR, Planning Department of the Government of the HKSAR, etc.
Oil and Gas: China National Offshore Oil Corporation (CNOOC), China National Petroleum Corporation (CNPC), China Petrochemical Development Corporation (CPDC), PetroChina, China Petroleum & Chemical Corporation (SINOPEC) , etc.
Manufacturers: Changchun FAW Group Corporation, China Sanjiang Space Group, etc.
Retail and Service Industries: Shenzhen Yijing National Cartoon & Animation Industry Base, PCCW (Hong Kong), The Hong Kong Jockey Club, ITV(Britain), etc.
Education and Research: 14 institutes of The Chinese Academy of Sciences, Tsinghua University, China University of Technology, etc.

For more information, please visit http://www.clustertech.com.