(13)
Overview: Enterprise data centers are struggling with large data sets with real implications for application performance, cost and competitiveness. While memory capacity has grown, distributed data center architectures require that many servers have access to the same data sets, often TBs in size and too large to fit in memory of any one server.
RNAcache provides a solution directly targeted at the root cause of this bottleneck; memory limitations. Because RNAcache dynamically scales to store the entire working data set in a shared, virtualized memory pool that appears to each server to be local memory, data does not need to be moved around the data center. Alternatives like faster storage, data replication, or scaling out marginally improve performance in isolated components. RNAcache delivers 20 – 100x faster results for applications that access large data sets to run simulations or business modeling. Examples include clustered applications, high-powered modeling for predictive analytics and high volume internet sites serving thousands of simultaneous users querying TBs of data (shopping, travel, social networks).
RNAcache delivers performance at scale, transparently allowing CIOs, application developers, data center architects, and business owners to achieve dramatic performance improvements without adding additional hardware, complicated software frameworks, and the associated maintenance burden to their data centers.
Data Center Survival Guide
It's a jungle in there. Today's data center is highly complex and hard to manage. That drives up costs, power and cooling requirements and provisioning time,...
Learn how innovative thinking can benefit your data center
In this white paper from The Bathwick Group, courtesy of IBM, you’ll learn the role of innovation in the data center.
|
