HPC Storage That Delivers on Performance, TCO and Insight
HPC challenges demand more from storage today. They demand more performance, more capacity … and more dollars. They demand more than legacy storage technologies and HPC budgets can deliver.
Cray® ClusterStor™ HPC storage meets these demands with the industry’s highest-performing and most efficient data storage technologies at the lowest overall TCO. ClusterStor parallel file systems balance the value equation by delivering the exact performance, speed, scalability, data protection and availability your requirements and budget demand.
More performance, better ROI. Enterprise-level performance with more capacity, fewer drives, less need for IT support and more data access.
Time and money saved. System design minimizes required hardware and maximizes efficiency among components. You don’t pay in lost time or money for technology you don’t need.
Business-boosting insights. ClusterStor technology eliminates maintenance issues, performance gaps, lost data and I/O issues which speeds your time to insight.
Three Solutions, Zero Challenges
ClusterStor L300 Storage System
The ClusterStor L300 storage system is an all-HDD Lustre solution. It achieves performance requirements with the lowest number of HDDs, enclosures and racks by maximizing the performance of each storage device. The use of all HDDs makes it well suited for environments using a few applications with large, sequential I/O workloads.
ClusterStor L300N Storage System
The Cray ClusterStor L300N storage system is a hybrid SSD/HDD solution with flash-accelerated NXD software that redirects I/O to the appropriate storage medium. It delivers cost-effective, consistent performance on mixed I/O workloads while shielding the application, file system and users from complexity through transparent flash acceleration.
ClusterStor L300F Storage System
The all new ClusterStor L300F scalable storage unit provides the opportunity to add Flash storage pool creating a truly hybrid system. The L300F is designed and optimized to overcome the latency experienced by rotating media - the remaining IOPS bottleneck for Lustre.
Unlike other Flash solutions, the ClusterStor Lustre-based L300F requires no specialized training, tuning or benchmarking. Lustre users with high IOPS needs can dramatically reduce the run time of their applications resulting in more iterations and faster time to insights.
In addition to the benefits adding a flash pool delivers, the ClusterStor L300F provides:
Consistency - IOPS performance is consistently delivered throughout the system. Unlike flash systems designed for high capacity the L300F doesn’t leave performance stranded behind the system infrastructure
Linear Scalability – Performance is scaled linearly as more ClusterStor L300Fs are added to the ClusterStor configuration assuring predictable performance all the way to exascale.
High availability – By using dual port SAS, the ClusterStor L300F delivers redundancy throughout the storage system, eliminating susceptibility to path failures.
Robust management – ClusterStor Manager helps administrators reduce the learning curve and accelerate proficiency. View for ClusterStor provides an end-to-end view of Lustre jobs, network status and storage system performance.
Like the rest of the ClusterStor storage line, the L300F is fully integrated, rigorously tested, optimized and supported by Cray’s dedicated storage team.
Slow and inefficient job performance, lost productivity, too much time searching for answers with cumbersome diagnostic tools and logs, and not enough time for analysis — these are the challenges administrators face every day when trying to understand and diagnose storage performance problems. Adding to the frustration is a lack of visibility into what applications are doing to the system, which makes optimization all but impossible without specialized tools. What administrators — and ultimately users — need are quick ways to isolate performance problems and visibility into what applications are doing all the time.
Cray® View for ClusterStor™ was designed to address these challenges. As the industry’s first complete Lustre® storage performance analysis application, Cray View for ClusterStor enables administrators to get the insight they need to maximize their resources, optimize system throughput, and provide users with the best results and experience. With Cray View for ClusterStor, you have all the information at your fingertips, from a single view, to identify storage performance issues whether they’re from application inefficiencies, a busy system, or another systemic issue that impacts applications. With always-on metrics, Cray View for ClusterStor moves these problems from the realm of black magic to practical application. After all, if you can’t measure it you can’t improve it.
Cray View for ClusterStor includes:
Job Runtime Variability: Real-time and historical views of data to help you understand what is impacting user jobs
Event Correlation: A unified view of the system, providing administrators with the ability to correlate systemic events that impact performance
Trend Analysis: Data-driven analysis and visualization from historical data can help identify trends that can then be used to shape changes to the system
Alerting: Threshold engine enables customized alerts based on any metric
The easy-to-use View for ClusterStor interface and visualization capabilities provide always-on insight into what applications are doing, without impacting performance. With detailed visibility into how resources are utilized by applications, administrators are finally enabled to make impactful optimizations
The result? Faster time to resolution, improved productivity, elimination of bottlenecks, and better overall performance.
Cray® View for ClusterStorTM is the industry’s rst complete Lustre® storage performance analysis application. Its always-on metrics give you the insight you need to maximize your resources, optimize system throughput, and provide users with the best results and experience — without impacting performance.
ClusterStor L300N storage systems with NXD flash acceleration gives organizations Cray-strength HPC storage with encrypted FIPS drives. You get unprecedented encrypted storage performance for data ingest, extraction and access with purpose-configured solutions for your most demanding applications.
Cray ClusterStor storage systems balance the value equation by providing enterprise-level performance with more capacity, fewer drives, less IT support required, and more access to data — all of which leads to the lowest TCO.
High-performance computing (HPC) is growing rapidly — and so is the amount of data that needs to be accessed and stored. Are you investing properly to make sure your HPC storage system can handle it, now and in the coming years?
This ebook is designed to help you understand the differences between the different types of storage — parallel, object and network attached storage (NAS) — in order to identify what is best for your budget and the performance you need … today and in the future.
High performance computing (HPC) is at the center of modern life sciences research. From genomic sequencing to large-scale genomic analysis to precision medicine, HPC plays a critical role in advancing many life sciences disciplines.
The oil and gas (O&G) industry has been undergoing fundamental changes in recent years. While declining oil prices have taken a toll, shifting regional demands and increased adoption of alternative energy are also affecting the industry in terms of profitability and business viability. To better compete, the O&G sector must adopt new technologies and methodologies to improve yield and achieve better operational efficiency.
High performance computing (HPC) plays a critical role in transforming manufacturing by accelerating innovation and establishing a competitive advantage. Consider this: 40 years ago airplane design was typically done by manually drawing the blueprints and laborious physical tests. New generations of airplanes often took years to come to the market. Today, most aerospace design work is done with simulation software such as CAE and CFD applications. This is possible thanks to the maturity of the modeling software, as well as the availability of HPC compute and storage systems that can run those parallel applications effectively.
Sept. 25, 2017 -- Global supercomputer leader Cray Inc. today announced it has completed the previously announced transaction and strategic partnership with Seagate centered around the addition of the ClusterStor high-performance storage business.
Alliance Technology Group is a North American systems integrator & IT solutions and services company. Alliance specializes in IT infrastructure, storage, systems, networking, physical security and cybersecurity.
Penguin Computing is a leader in developing open, Linux-based cloud and HPC solutions. They offer a comprehensive portfolio of products, from Linux servers and workstations to integrated, turnkey HPC clusters and cluster management software.