Columbia University Deploys Isilon NAS

EMC Corporation announced that Columbia University’s Center for
Computational Biology and Bioinformatics has deployed EMC Isilon scale-out NAS to support the
advanced, computing-intensive demands of its ground-breaking research programs.

Using the EMC Isilon X-Series and NL-Series Platform Nodes combined with the
EMC Isilon SmartPools and SmartQuotas software applications, C2B2 has increased system performance and flexibility, streamlining
researchers’ access to data, while reducing operational expenses and big data
management complexities.

"We were using a traditional NAS system that struggled
to support the huge amounts of input/output demands on the 400 CPUs in our
computing infrastructure," said John Lowell Wofford, director of IT
services for Columbia
University’s C2B2.
"We knew that we’d soon outgrow that number by at least ten times. After
switching to Isilon, we no longer had to worry that our system couldn’t handle
our research demands. We knew that we could independently scale capacity and
performance, so that we buy only what we need, when we need it."

The Isilon storage system that C2B2 employs now supports
some 4,000 CPUs, which can handle the heavy I/O and data analysis demands of
C2B2’s research into such areas as computational biophysics and structural
biology. In addition to C2B2, the Isilon system also supports the storage needs
of Columbia University’s
Herbert Irving
Comprehensive Cancer
Center, the Institute for Cancer
Genomics, and the J.P.
Sulzberger Columbia
Genome Center.
By leveraging Isilon system capabilities these research centers reduce cost and
IT complexity by sharing a computing and storage infrastructure that’s flexible
enough to match a range of tasks.

Genomic databases, such as the ones C2B2 employs, are
typically directories with an enormous amount of files that require a heavy
amount of namespace reads to index. Wofford notes that 40 percent of the
read-write requests on the C2B2 system are namespace-related. The Isilon
X-Series’ intelligent use of SSDs for metadata and file-based storage speeds
namespace-read performance, providing a "huge improvement" over
C2B2’s previous computing/NAS system infrastructure.

The Isilon system enables C2B2 to speed researchers’ access
to data and streamline big data system management. "We’re processing big
sets of genomic data, doing molecular biophysical simulations and sequence
analysis, which can lead to new drug discoveries and advances in basic
science," said Wofford. "Despite the fact that we have nearly one
petabyte of data, the EMC Isilon OneFS operating system is so easy to manage,
we don’t need a dedicated storage administrator. Aside from the cost savings,
we’re able to free up our IT people to focus on other initiatives, such as our
virtualization infrastructure."

"Columbia
University’s Center for
Computational Biology and Bioinformatics is leading the way in the scientific community with its ground-breaking research programs," said Sam Grocott,
vice president of marketing, Isilon. "We are happy to provide the storage
foundation for its computing-intensive demands, so C2B2 can focus on the
important discoveries at hand and not on managing the headaches of complex big
data infrastructure."