Country missing? Please select your nearest region...
1 February 2012 by Penny Jones - DatacenterDynamics
EMC has launched what it said is the first “enterprise proven” scale-out NAS architecture using Hadoop for data analytics.
The EMC Isilon scale-out NAS with Hardoop Distributed File System is designed to enable distributed computing, allowing large data sets to be transported across computing clusters for high density environments.
It makes using Hadoop much easier, according to EMC, removing complexity and making the rollout of Apache-based Hadoop much faster.
“Up until now, early adopters of Hadoop have relied on makeshift storage infrastructure not optimized for Big Data exploration, inhibiting the shared data access critical to robust analysis, and preventing wide adoption of Hadoop in the enterprise,” EMC said.
Many industry experts have predicted Hadoop will be big in the year to come as many data center operators struggle with growing demands for storage and analytics resulting from growing amounts of data.
Savvis’ Cloud Services VP Daniel Patton said much of this demand is being driven by cloud computing.
“Companies are now moving into a world of petabytes,” Patton said.
“It is not easy to move a petabyte of data around, so you get this general idea that it is easier to move the compute to the data than it is the data to compute, so you really want to store that in the cloud.”
He said Hadoop will become more important for cloud providers looking to ensure fast availability of data in years to come.
IDC worldwide storage systems program VP Benjamin Woo said the analyst firm also predicts that companies will accelerate Hadoop adoption and deployment this year.
“Hadoop is becoming a viable competitor to existing storage systems for big data analytics, but fear can stand in the way,” Woo said.
“An investment by an industry leader like EMC casts a vote of confidence for Hadoop environments and makes the technology much more palatable.”
Hadoop is based on Apache software, and moves the emphasis from the delivery of high availability data from the hardware to software layer.