Fusion-io Accelerates Deutsche Bank Credit Risk Calculation
Performance increases by "up to 60%".
This is a Press Release edited by StorageNewsletter.com on October 14, 2011 at 2:53 pmFusion-io, Inc.
announced that Deutsche Bank has
deployed ioMemory technology to
reduce latency and increase performance by up to 60 percent in its Credit Risk
Calculation and Information System (dbArtos), according to Deutsche Bank AG
Director Dr. Marcus Pratzas.
After evaluation of a variety of
technologies, Fusion ioMemory was selected to accelerate the SAS business
analytics and business intelligence software utilized by the Deutsche Bank
dbArtos system.
"The Deutsche Bank credit risk
portfolio uses SAS business analytics software for credit risk calculation and
information system evaluations, and we recently updated our infrastructure to
meet new and additional requirements for these risk evaluations," said
Dr. Pratzas. "The infrastructure
design update was focused on low latency technologies, and now, Fusion ioMemory
is an important part of the Deutsche Bank calculation engine architecture. With
Fusion-io, we reduced operating costs and achieved considerable performance
improvements up to 60 percent without any changes to the application."
The SAS software used in the Deutsche Bank dbArtos system is designed to help
customers make fact-based decisions to improve investment choices, from
identifying the right product to market to forecasting trends. The impact of
these decisions made it critical for Deutsche Bank to offer the lowest latency
and highest performance possible for the SAS application in order to allow
customers to make informed decisions about financial matters with the best
possible information in a real-time banking environment. Given the importance
of the dbArtos system to Deutsche Bank and its customers, extensive evaluations
were conducted before selecting technology from Fusion-io over traditional
disk-based solutions and SSD offerings.
"In an environment where
milliseconds make a big difference to the bottom line, rapid access to data is
critical to meeting the business analytics requirements of Deutsche Bank
clients," said Neil Carson, Fusion-io Chief Technology Officer. "Latency is significantly reduced when the
large, evolving datasets used in financial analytics are stored in the server
where data is processed and accelerated through a dedicated memory tier. With
Fusion ioMemory, financial organizations like Deutsche Bank can provide their
customers with significantly faster responses to market developments."
Fusion-io integrates flash as a new tier of non-volatile memory, rather than
attempting to integrate flash through legacy disk-based storage protocols. As
demonstrated by Deutsche Bank, this architecture enables ioMemory to deliver a latency advantage driven by fewer context switches and a parallel
construct, which eliminates bottlenecks. Without this architecture, alternative
solutions struggle when trying to achieve anywhere near the same levels of
application acceleration, or infrastructure consolidation.
In addition to improving application performance, Deutsche
Bank also found technology from Fusion-io delivered raw I/O increases that
offered four times the performance of traditional hard disks and SSDs also
evaluated by Deutsche Bank.
Deutsche Bank also observed a significant improvement in overall CPU
utilization, which meant that the ioMemory reduced the amount of time the CPU
waits for data to process. In addition, because Fusion’s solution permitted
Deutsche Bank to minimize the amount of equipment needed to achieve the
necessary performance, the financial institution realized a lower cost of
ownership in comparison to its earlier system due to reduced power, cooling,
management and infrastructure expenses.