In this video, Torben Kling-Petersen from Seagate describes new storage innovations for high performance computing. At SC15, Seagate announced a major expansion of its HPC product portfolio including the ClusterStor HPC hard disk drive designed for Big Data applications.
New ClusterStor products include:
- ClusterStor L300 – Engineered Solution for Lustre. The L300 integrates with Seagate’s new 3.5-inch ClusterStor HPC drive specifically designed for the high-performance computing market, establishing the company’s commitment to full HPC solutions. The L300 delivers up to 112 gigabytes-per-second (GB/sec) throughput per rack when integrated with Seagate’s ClusterStor HPC drives, a performance improvement of 77 percent compared to previous ClusterStor systems. This level of performance density per rack unit helps customers reduce data center space, power and cooling costs while simplifying management and equipment requirements. Additionally, the solution optimizes random and sequential I/O workloads across the entire storage cluster, improving workflow productivity.
- ClusterStor G200 – Engineered Solution for IBM Spectrum Scale (GPFS). This HPC-class storage solution is designed to to help data-driven enterprises manage massive storage growth. The G200 integrates with Seagate’s new 3.5-inch ClusterStor HPC drive specifically designed for the HPC market, highlighting the company’s commitment to industry-leading HPC storage solutions.
- ClusterStor A200 – Engineered Solution for Tiered Archives. As Tier 2 storage expressly designed for active archive and HSM applications, the ClusterStor A200 is a first-of-its-kind tiered archive storage system for high performance computing (HPC). The A200 helps reduce storage and operational costs by up to 50 percent compared to tier-one storage platforms. The A200 was designed to complement the rest of Seagate’s ClusterStor family of scale-out storage systems. It allows customers to non-disruptively migrate designated data off of the performance-optimized, primary storage tiers while keeping it online for fast retrieval. This avoids a common problem in shared HPC environments where the organization is forced to choose between having all of the data available to make the best analysis versus the time required to retrieve data from tape. Performance of the primary storage is often improved by migrating data and freeing up space for more efficient data layout. The pre-configured ClusterStor A200 solution includes an automatic policy-driven hierarchical storage management (HSM) system and near limitless scale-out capacity.