TO RESULTS FOR 1,000



Industry: Academic Research

“The challenge of managing varied research needs is accommodating both very large parallel I/O jobs and millions of small, random read requests without imposing performance penalties on anyone. We needed high-performance storage that could address data volume, velocity and variety.”

Mike Shuey
Research Infrastructure Architect, Purdue University

Purdue University is a global leader in research, discovery and innovation with one-of-a-kind research and educational opportunities to catalyze and facilitate research breakthroughs for global impact. In supporting the needs of up to 1,000 researchers working on several hundred, concurrent research projects, the university’s IT team must ensure uninterrupted access to centralized compute, network and storage resources. As each research project requires access to scratch storage for project files along with archival storage for longer-term data retention, the decision was made to purchase one large, site-wide file system that could be accessed from multiple HPC systems. With such broad technology and connectivity requirements, Purdue’s new system needed to sustain the highest levels of performance for all researchers, regardless of the size or demands of the project.

Read their success story to learn how high-performance DDN® storage with SFX® technology has helped the Purdue University:

  • Help academic researchers achieve quicker time to results with data inquiry results accelerated by 900%
  • Dramatically reduces I/O bottlenecks by providing up to 300x more IOPS than rotating media and performs at speeds much closer to RAM



  • Meeting the needs of up to 1,000 researchers working on several hundred projects drove the decision to deploy one large, centralized data repository powered by high performance storage
  • Big data variety, velocity and volume created the need for highly versatile, scalable storage
  • Diverse workloads required highly flexible storage to accommodate large parallel I/O jobs and many small, random read requests


A centralized, high-performance storage solution for Purdue University’s Data Depot service, driven by DDN’s SFA12KX® storage platform with SFX. Additional sites use DDN EXAScaler® parallel file system appliance with SFX® cache acceleration.


  • DDN’s SFA® storage platform with SFX technology has enabled Purdue to better address the needs of all researchers, including non­traditional groups and new HPC users
  • Reduced the time for simple data queries from two minutes to two seconds, which enables quicker time to results
  • With DDN, Purdue handles big data sets at a price point that’s closer to lower-cost hard disk drives



  • With DDN SFX, nine out of ten random read I/O requests are served from cache, which delivers across-the-board performance improvements
  • Handling random data access or reads of millions of small files on dedicated solid-state modules removes the performance penalty on data-intensive storage requests
  • DDN’s storage versatility has resulted in the addition of a new EXAScaler appliance to meet the needs of a new compute cluster