JSTOR, a cloud native digital library that contains upwards of 50 million pages of content and serves millions of users annually, has implemented Metafor Software’s machine learning based real-time anomaly detection product to interpret network traffic events by analyzing terabytes of machine data daily. This early detection has decreased JSTOR’s mean time to resolve issues, while ensuring a trustworthy, reliable environment for their customers.
Mountain View, CA, May 15, 2015 –(PR.com)– Metafor Software, a leading provider of real-time anomaly detection technology, and JSTOR today announced that JSTOR has adopted Metafor Software as a core component of its technology operations. JSTOR, a digital library that contains upwards of 50 million pages of content and serves millions of users annually, has implemented Metafor to speed awareness of problems in application performance and site usage, and to enhance its monitoring of key metrics.
The JSTOR website is a cloud native, micro services architecture running on Amazon Web Services (AWS). It contains over one hundred different applications and growing. Using Metafor, JSTOR can now interpret network traffic events by analyzing terabytes of machine data daily.
“Our engineers release changes as much as one hundred times per week, and thousands of higher education institutions around the world provision access to JSTOR through reverse proxies running on campus to authenticate patrons,” said Archie Cowan, Chief Architect at ITHAKA (parent of JSTOR). “This environment is so dynamic that a strong performance and usage anomaly detection capability is critical to ensuring a great experience for our users.”
JSTOR did a trial run of Metafor’s product early on, running tests using historical network data. Metafor detected all of the past incidents that JSTOR had found, as well as some that were more complex that it had not.
“Using Metafor has been like adding a great engineer to our team,” added Cowan. “Now up and running on our real-time network traffic, Metafor has helped us respond in record time to customer-facing issues that are challenging to detect. This early detection means we are resolving issues before any support calls come in.”
“We are thrilled that JSTOR selected Metafor as a strategic technology to support its growing operations,” said John Fowlkes, CEO of Metafor Software. “It’s great to be able to help customers like JSTOR decrease their mean time to resolve issues, while ensuring a trustworthy, reliable environment for their customers.”
To learn more about Metafor’s machine learning based real-time anomaly detection, visit: http://www.metaforsoftware.com
About Metafor Software
Metafor’s real-time machine learning provides early detection of the anomalies that matter in the network. The product automatically analyzes the behavior of logs and metrics of any type from any physical or software defined data source so you can get alerted at the earliest sign of a problem and fix it immediately; no threshold or rule setting required. Using a broad set of non-parametric, unsupervised learning techniques, it detects “unknown unknowns” – the issues you haven’t seen before and don’t have rules or signatures for. Metafor is delivered as either a SaaS / private cloud offering or in on-premises form factors. Metafor was named a 2014 Cool Vendor in Application Performance Monitoring and IT Operations Analytics by analyst firm Gartner. For more information, visit: http://www.metaforsoftware.com/about-us
JSTOR is a not-for-profit digital library that provides access to academic journals, books, and primary sources to nearly 10,000 institutions and millions of individuals around the world. By using JSTOR, people can discover centuries of historical and current content through a powerful set of research and teaching tools while also being assured this content is stable, preserved, and will be available for future generations. JSTOR is a part of ITHAKA, a not-for-profit organization that also includes Portico and Ithaka S+R. For more information, visit: http://about.jstor.org