How to Make Solving I/O Bottlenecks Look Easy

We live for a good customer challenge. And you can’t get a much better challenge than one that turns into a new product with broad application. The challenge came to us by way of Los Alamos National Laboratory (LANL) and the National Nuclear Security Administration (NNSA). LANL launched their “Trinity” supercomputer in 2015. Based on Cray® XC™ series architecture, Trinity is the first installation of the NNSA Simulation and Computing program’s initiative to develop and deploy a series of advanced technology systems. Together they’re exploring compute technology so they can provide platforms with higher performance for predictive capability — all while solving national security challenges. In short, NNSA and LANL are working ... [ Read More ]

Is Lustre Ready for Flash Storage?

Flash is the flashiest new storage technology, but is Lustre® ready for it? The short answer is “yes.” We’re optimizing Lustre for flash — including server-side configuration, setup and IOPS performance tuning. Peter Bojanic and our panel of Cray storage experts will go in-depth in our upcoming webinar, “Will Lustre Ever Be Ready for Flash?”, at 9 a.m. PT on June 12, 2018. Balancing performance and economy Spending your storage budget strategically on the optimal balance of capacity versus performance storage gives you more resources to put to work solving mission-critical problems. Flash is expensive for capacity storage, but it easily beats HDDs on $/IOPS and $/GB/sec. HDDs, on the other hand, are an economical choice for ... [ Read More ]

5 Answers to Your Lustre Storage Questions

We recently held a webinar for the Lustre® community — now available on demand — to address questions about the convergence on a single canonical Lustre release stream and what that means for future feature development. Some of you asked questions that we ran out of time to cover in our Q&A session. Our storage experts answer them here: 1. Any tips on how to identify usage patterns on previous Lustre versions that would benefit from Lockahead in 2.11? Our 2017 CUG paper, "Lustre Lockahead: Early Experience and Performance Using Optimized Locking," addresses that very topic. Lockahead shows significantly improved write performance for collective, shared-file I/O workloads. Initial tests show write performance improvements of more ... [ Read More ]

New ClusterStor L300N System Addresses the #2 Roadblock in HPC

Hyperion identified storage access time as the #2 roadblock in HPC. The newly announced ClusterStor™ L300N hybrid storage system with intelligent NXD software is purpose-engineered to address the mixed I/O bottleneck that many HPC users are facing as they run more and more applications on the same system. In fact, this software innovation delivers up to 1,000 percent I/O workload acceleration over traditional HPC storage systems. The NXD I/O manager monitors incoming I/O in real time and optimizes data access by directing large-block writes directly to HDDs while small writes are sent to SSDs. Once the small writes have been coalesced on the SSDs, the blocks are flushed to the HDDs. This allows completely transparent, efficient ... [ Read More ]

Cray Analytics and Alluxio – Wrangling Enterprise Storage

For business to not just survive — but to flourish — it’s become imperative to make decisions with near immediacy, continuously pivot strategy and tactics, and merge streams of inquiries into meaningful action. Executing requires high-frequency insights — the competitive advantage in today’s frenetic business landscape. Together with Alluxio, Inc., we enable businesses to gain the competitive advantage with faster time to insights with our integrated solution of Cray high-performance analytics platform and Alluxio’s memory-speed virtual storage system — Alluxio Enterprise Edition. For businesses on the cusp of innovation and seeking that information advantage, Cray has fused supercomputing with an open, standards-based framework to ... [ Read More ]