Audience measurement software maker Quantcast just made a very notable contribution to the open-source community – it released the Quantcast File System (QFS), a “higher performance alternative” for the Hadoop Distributed File System. QFS is fully integrated with Hadoop, uses as little as half the disk space that HDFS would require, and offers better IOPS according to the company.
Quantcast provided an internal case study to validate its platform. The company uses QFS as the primary data store for the 20 petabytes of information it processes on a daily basis. That amounts to about 100 million web destinations and is “well in excess of 500 billion records” aggregated and analyzed each month.
“The sheer size and scale of its data set easily positions the company among the top five data processing organizations in the world, and its success leveraging QFS in production offers a compelling use case,” said Ben Woo, principal analyst and managing director at Neuralytix.
“With the inherent ability to more efficiently store, manage and process data at multi-petabyte scale, QFS will be attractive to technology savvy organizations that require greater storage efficiency and performance from their Hadoop clusters.”
The ecosystem around Hadoop is constantly expanding thanks to an influx of new players that are finding new ways to leverage and enhance the open-source software. A partnership between Hadoop distributor HortonWorks and sqrrl earlier this week was one very notable milestone in this regard.
sqrrl, founded by the NSA’s best and brightest, is working with the Apache Accumolo storage system to enhance the platform’s cell-level security functionality - the ability to customize access and authorization for different users. By partnering up with the startup Hortonworks, sqrrl will be able to offer its customers a new level of security – a commodity high in demand among enterprises that analyze mission-critical and potentially confidential data.