RETAIL

Meiosys, a provider of transparent middleware solutions for optimizing and protecting business-critical applications, today announced support for Altair Engineering's PBS Pro (version 5.4) workload management solution. Meiosys and Altair Engineering have successfully integrated MetaCluster HPC -- Meiosys' transparent checkpoint and restart (CPR) solution -- with Altair's flagship batch management system. Both PBS Pro and MetaCluster HPC target high-performance computing customers who demand better availability and manageability of their applications running on clusters and computing farms. "Altair customers will truly benefit from this integration effort," said Michael M. Humphrey, vice president of Altair Engineering's Enterprise Computing Business Line. "Our strategy is to ensure PBS Pro is integrated with best-of-breed technologies that meet our customers' needs. The combination of MetaCluster HPC and PBS Pro V5.4 provides an extremely powerful and flexible solution." "Meiosys is pleased to see our MetaCluster HPC product integrated with PBS Pro, which customers have anticipated," said Hubert Catanese, vice president of sales and marketing for Meiosys. "Altair has an outstanding reputation in several markets, including government, manufacturing, Life Sciences and Energy. By installing MetaCluster HPC, PBS Pro users can protect their long-running jobs transparently and dynamically optimize their hardware and software resources." PBS Pro users can enjoy the benefits of Meiosys solutions without having to modify their application or operating system or learn a new user interface. Meiosys MetaCluster HPC is a unique, transparent middleware layer that supports multi-thread, multi-process, sequential and MPI-based distributed applications. The Need for Checkpoint and Restart Checkpoint and restart are mechanisms for protecting critical applications from computer failures by creating a snapshot -- or checkpoint -- of the complete state of the application at a given point in time. Points can be assigned through policies (daily, weekly or a given time interval) or as needed during long runs. If the system or application fails, it can be restarted from the most recent checkpoint file, without having to start over at the beginning and rerun the entire job. The result is predictable and optimized computing operations.