December 4, 2015 - Last Seagate Product Announcement
The L300 integrates with Seagate’s new 3.5-inch ClusterStor HPC drive, designed specifically to meet the demands of high-performance computing environments.
When combined with the HPC drives, the L300 delivers up to 112 gigabytes-per-second throughput per rack, a performance improvement of 77 percent compared to previous ClusterStor systems.
The L300 offers a disk pool management system so that users can easily manage multiple types of media – including Seagate’s flash memory technology – from one central database. Additionally, the L300 optimizes random and sequential I/O workloads to improve productivity.
The L300 integrates with next-generation networks built for exascale computing, including Intel’s Omni-Path and the Mellanox EDR 100Gb/s Infiniband.
The French Alternative Energies and Atomic Energy Commission (CEA) intends to use ClusterStor L300 for its next-generation 25-petaflop supercomputer.
: HPC storage is no longer limited to academia; industries ranging from life sciences and automotive to aerospace design and financial services are seeking new ways to efficiently manage massive amounts of data. The G200 – which comes ready to install – brings scale-out HPC storage to those applications relying on mission-critical, IBM Spectrum Scale-based deployments.
Like the ClusterStor L300 mentioned above, the G200 integrates with Seagate’s new 3.5-inch ClusterStor HPC drive, designed specifically to meet the demands of high-performance computing environments. The drive offers up to 4 terabytes of storage capacity, a sequential data rate of 300 megabytes per second, and 35 percent faster random performance than any other drive of its size.
Initial customers include Atos and Dell.
: The A200 solves the common dilemma in HPC environments where users must choose between expensive primary tiers for best data analysis, and less-accessible—and more time-consuming—archiving and retrieval methods, such as tape. The A200 helps reduce storage and operational costs by up to 50 percent compared to tier-one storage platforms.
The A200 is a pre-configured solution that includes an automatic policy-driven hierarchical storage management (HSM) system and near limitless scale-out capacity. It includes a single, easy-to-use administrator interface, as well as network-based erasure coding to provide data protection with up to 11- nines of durability. This allows A200 to achieve extremely fast rebuilds of failed drives while prioritizing the most critical elements, improving data availability and maximizing storage space.
The A200 is ideal for HPC applications that involve quickly analyzing fast-changing data, such as prediction models in weather, as well as applications in life sciences, oil and gas, and geospatial intelligence. |