Today DDN announced the latest generation of EXAScaler, the company’s high performance Lustre solution. Ideal for HPC and data-intensive Enterprise environments, EXAScaler 3.0-based solutions combine the industry’s fastest storage platforms from DDN with the Intel Enterprise Edition for Lustre 3.0 to deliver the world’s most advanced Lustre appliances.
EXAScaler 3.0 raises the bar for Lustre performance and management,” said Laura Shepard, senior director of products and vertical markets, DDN. “As the world’s most experienced Lustre provider, DDN leverages input from a broad installed base and the Lustre community to deliver the most advanced Lustre solutions to our customers around the globe.”
Developed in close collaboration with Intel and with full Intel and DDN global support, EXAScaler offers the world’s highest sustained Lustre performance, in a converged appliance that is simple to install, scale and manage.
DDN was the developer or lead tester for all the major features in Intel EE for Lustre software 3.0, which is the Lustre core within EXAScaler 3.0. In addition to Intel EE for Lustre core features, EXAScaler 3.0 adds unique, DDN exclusive features such as high availability enhancements that significantly improve mean time to data access, metadata scaling, efficient project quotas, and support for the differentiated storage services framework, among others.
As a strong supporter of the open source community, DDN actively develops and contributes software features, tools and frameworks to help advance future versions of Lustre,” added Shepard.
To further increase the community interaction on Lustre development tools, management frameworks and performance and scaling development, DDN will be open sourcing a number of projects this fall. Community sharing projects will include:
A “Lustre Automatic Test System” that will help developers to run regression tests in parallel on hundreds of test nodes;
A Lustre monitoring framework platform based on industry standards such as Collectd, elastic search capabilities for NoSQL databases and Hadoop, and visualization on Grafana
Parallel copy tools and internal Lustre mechanisms that will move and copy data between OSTs and across namespaces more effectively and in parallel.