Ellexus is excited to announce the release of Mistral, our ground breaking product that will balance shared storage across a high performance computing cluster. Developed in collaboration with ARM’s IT department, 2015 was a year of intense production – 2016 is the year that Mistral will take HPC by storm.
Mistral addresses the noisy neighbour problem: when a small number of jobs overload the network or file system in a compute cluster with shared storage. Sometimes this problem is caused by rogue jobs that have been submitted to the cluster by mistake. Other times the cluster may simply by overloaded with a high number of I/O hungry jobs. In either case, the performance of all the jobs on the cluster can be affected and the cluster may even be brought down completely, bringing hundreds of engineers to a standstill and causing critical deadlines to be missed at a cost of tens or hundreds of thousands of pounds.
Mistral works by monitoring application I/O and I/O performance across a cluster in order to identify rogue jobs and hotspots, and can automatically throttle I/O problem jobs and applications. By deploying Mistral, a company can gain an in-depth understanding of how users are accessing the storage as well as preventing disastrous data access patterns.
Mistral is now in full production mode and is already being deployed by our customers. The product was launched to a keen waiting list of international organisations and we have no doubt that it will make real differences to their cluster performance.
For more information about Mistral and to request an evaluation of the software, contact Rosemary at firstname.lastname@example.org.