Quantum Adds Deduplication To StorNext 4.0
Quantum has announced native deduplication capabilities in StorNext 4.0, the latest version of its file sharing and data management software. The system also now automates data tiering, can perform timecode-based partial file retrieval and includes its own replication engine, eliminating the need for third-party solutions. General availability is set for mid-February.
January 22, 2010
Quantum has announced native deduplication capabilities in StorNext 4.0, the latest version of its file sharing and data management software. The system also now automates data tiering, can perform timecode-based partial file retrieval and includes its own replication engine, eliminating the need for third-party solutions. General availability is set for mid-February.
StorNext is built around the StorNext File System, which installs on hosts that share disks, arrays or SANs and permits sharing and simultaneous access to files. At the heart of the system is the MetaData Controller, standing between Linux and Windows clients, and servers running Linux, Mac OS X, Unix and/or Windows. "There's more raw data today and it's being kept around longer, so you need to enable a single file pool to be accessed by multiple OSes simultaneously," says Chris Duffy, StorNext product marketing manager.
Building deduplication into the file system eliminates obvious data redundancies, and the company says it also facilitates a form of tiering. "Deduping in the file system enables you to have a dedupe file repository," says Duffy. "Just drop a file in the dedupe folder, and it has actions performed on it as indicated." Better optimization in a tier-one infrastructure, he says, means deduped data that was once stored only offline can be kept closer at hand.
Also new in StorNext 4.0 is timecode-based, partial-file retrieval. According to storage analyst Tom Coughlin of Coughlin Associates, this will boost productivity, particularly for professionals in the media and entertainment fields. "Metadata is extremely important in professional media applications such as video editing," he says. "Broadcasters will spend many hours describing where a shot was taken, who was in it and who took it, and so much is tied to the time stamp. The time stamp in video apps are the key to pulling up what you need when you need it."
Duffy describes the benefits in terms of hardware and network resource utilization. "This will improve workload and storage optimizations because you don't have to set aside the storage and bandwidth requirements of transferring a 30-minute film but only of the 15-second clip you want to retrieve."
Coughlin, whose firm specializes in storage for the media and entertainment industries, added that the capability also helps support real-time video editing environments, post-production and other media applications. "It gives the IT professional supporting media and entertainment facilities more options in terms of the storage tiering and creating workstations with less costly hardware."
For multi-tiered systems, so-called distributed data movers or DDMs, can move data among tiers to optimize access speed and server horsepower requirements. By deploying multiple DDMs, the workload can be shared across many servers. This also provides a measure of resiliency, because if one DDM fails, another will take its place. Pricing of StorNext 4.0 is based on the number of SAN clients and the amount of data stored. The company says an average installation costs about $2,000 per client plus $1,000 per Tbyte. Optional DDMs cost $15,000 each.
You May Also Like