Hyve Partners Up for Fast Apache Hadoop Implementation

 
 
By Chris Preimesberger  |  Posted 2011-11-29 Email Print this article Print
 
 
 
 
 
 
 

Another new-generation data analytics package became available to the markets Nov. 28 when Fremont, Calif.-based Hyve Solutions launched its Series 8 high-performance Hadoop computing platform.

Apache Hadoop platforms from Cloudera, Hortonworks, EMC, Hewlett-Packard and several other companies have already hit the markets and are being evaluated for their pros and cons.

Hyve's take on Hadoop is that it has partnered with several other forward-thinking, young companies: Arista Networks, Fusion-io, Solarflare Communications and Zettaset. Hyve says its Series 8 offers enterprise users a fully integrated analytics package that can be sized to current workloads and scaled to meet future demands.

Batch benchmarking test results achieved in October showed that a five-node cluster completed a 1TB Terasort workload in 30 minutes, 16 seconds. The same five-node cluster scaled linearly completed a 4TB Terasort workload in 1 hour, 59 minutes and 11 seconds.

Not bad -- not bad at all. These kinds of workloads used to take days or weeks to process.

Flash-based ioMemory from Fusion-io drives the massive I/O requirement. Each node includes three 1.28TB Fusion ioDrive Duos.

Hyve Series 8 has three shipping configurations: a two-node starter kit (7.2TB); full rack (72TB); and individual 2U expansion nodes (3.6TB).

Networking, compute, storage, software and support are preconfigured with each SKU. Pricing and product availability will be in place by early December 2011, with first shipments of the built-to-order systems beginning in January 2012.

 
 
 
 
del.icio.us | digg.com
 
 
 
 
 
 

Submit a Comment

Loading Comments...
 
 
Manage your Newsletters: Login   Register My Newsletters























 
 
 
 
 
 
 
 
 
 
 
Thanks for your registration, follow us on our social networks to keep up-to-date
Rocket Fuel