Object storage vendor Scality Inc. is going after the media and entertainment market with its Ring Digital Media...
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
Storage application launched last week.
Ring Digital Media Storage builds on Scality's base Ring Organic Storage software, which helps companies and service providers build storage clouds. For digital media, Scality added support for content delivery networks (CDNs) and media asset management (MAM) software for active directories, archiving and near-line storage. Scality also plans to add support for fast media file transfers later in 2013, according to Director of Product Strategy Philippe Nicolas.
"The idea is to support the broadcast media environments with the Scality Ring," Nicolas said. "We can get data from many media content sources and serve it through HTTP to different CDNs. We can also deliver file requests through HTTP cache."
Scality's object storage uses a distributed architecture to provide concurrent access to data stored on x86-based hardware. An application in a server makes object PUT or GET requests through a setup designed as a ring of storage nodes. Scality uses this distributed architecture to provide highly concurrent access. Ring uses replication and erasure coding for data protection.
"We provide a peer-to-peer, distributed architecture so applications talk to storage without gateways but in multiple paths," Nicolas said.
Scality's move into the digital media industry follows the company's March announcement regarding its Hadoop strategy. Nicolas said customers can now deploy Hadoop on a Scality Ring cluster and use the Scality Scale Out File System (SOFS) in place of the Hadoop Distributed File System (HDFS).
The first step in Scality's Hadoop strategy is to support Cloudera Inc.'s CDH4 and Hortonworks Inc.'s HDP 1.0 Apache Hadoop frameworks.
"Instead of using the Hadoop cluster, you use the Scality Ring storage cluster to run Hadoop jobs natively," Nicolas said. "You use the Scality file system instead of HDFS, which has limitations due to the nature of the product. It uses a name node or metadata server of HDFS, so if you lose a NameNode, it is a single point of failure."