SAN FRANCISCO--(BUSINESS WIRE)--Tapping into the minds and experience of the world’s leading storage researchers, Acunu today released version 1 of the Acunu Data Platform, the first reinvention of the storage stack in more than 20 years. The platform and its open-source core are being showcased today at the GigaOM Structure Conference in San Francisco.
The next evolution
The Acunu Data Platform takes advantage of the dual evolutions enabling big data applications: The rise of non-relational or NOSQL databases such as Cassandra and analytical platforms such as Hadoop, and the performance increases of commodity hardware on which these databases run.
Acunu developed its platform from the ground up to leverage developments such as distributed servers in the cloud and the huge throughput increases enabled by SSDs.
“New technologies have given us tremendous potential for handling big data,” says Tim Moreton, Acunu’s CEO. “But up until now, the storage stack that allows applications to communicate with hardware has been stuck in the 90s. Acunu is a Linux kernel module that enables you to make much better use of the commodity hardware you already have.”
What it isn’t, what it is, what it does
Unlike others in the big data market, Acunu is not simply building tools or supporting existing open-source platforms. The company is completely reengineering the storage platform for use with a variety of data platforms. Today’s release supports Cassandra and an S3-compatible store, and Acunu is already porting Voldemort and memcached. Additional data platforms will be supported in future releases.
Acunu delivers higher performance and more predictable behavior than Apache Cassandra on the same commodity hardware. Specific advantages of using Acunu include:
- Eliminates garbage collection delays and increases throughput by up to 50X compared to vanilla Cassandra.
- Provides algorithms that increase performance without requiring the tuning needed with Cassandra.
- Operates in the Linux kernel, rather than Java, eliminating the need to tune the Java garbage collector.
- Simplifies management and monitoring across the cluster by operating through a single web-based user interface.
- Provides one-click snapshots and clones at scale, enabling new code to be tried out across the full dataset rather than through sample code. This increases development speed and reduces the risks of putting new code into production.
Put to the test by hit TV show
A beta program involving more than 30 organizations preceded today’s release of the Acunu platform. One of those organizations is Live Talkback, which developed the Buzz Off game for “Britain’s Got Talent,” Simon Cowell’s TV show watched by more than 10-million viewers a week. The web application enables viewers to vote in real time to “buzz off” acts they don’t like.
“The volume of peak traffic handled by the Buzz Off application for ‘Britain’s Got Talent’ is the equivalent of 130 billion requests a month, which would make it a global top-10 website,” says Malcolm Box, CTO of Live Talkback. “We think Acunu software has a unique approach that will enable us to dramatically increase throughput, reduce time spent tuning, and integrate easily with other tools.”
To find out more about Acunu’s role in the Buzz Off application, visit Malcolm Box’s blog.
Licensing, pricing and support
The Acunu Data Core is open source under the GPLv2. The rest of the Acunu Data Platform will be offered at a range of support and pricing levels, beginning with Acunu Standard Edition, available free of charge.
For more information, visit http://www.acunu.com.
Product and service names mentioned herein may be the trademarks of their respective owners.