Quantum Adds Scalability, Big Data Support To StorNext

Quantum on Thursday rolled out major enhancements of its StorNext file system aimed at handling big data and archive applications.

The new software focuses on managing the large datasets generated in media and entertainment, oil and gas exploration, life sciences, and video surveillance. These datasets consist of a large number of objects that are randomly accessed by multiple users and made available for analysis and processing.

The new StorNext 4.3 includes changes to a mySQL database for managing large numbers of objects and storage of over one billion files.

Additional performance enhancements in StorNext focus on tiering and archiving, and faster creation and storage of files. Access to files over the LAN has also been improved by use of a proprietary LAN protocol.

[ Read about Quantum's StorNext M660 metadata appliance. See Quantum Unveils Appliance For Big Data. ]

A new tier of storage, the Archive Vault, works on massive quantities of data that is written once and read rarely. This data needs to be removed from active tiers of storage but remain closely managed in an archive inside a tape library. The vaulting of data to tape and the return to the library are semi-automated, requiring operator assistance. Tapes in the Active Vault are checked for data integrity.

When large files are ingested form cameras, sensors, or other devices, they are written to a StorNext disk system and archived immediately to tape. File data is removed from disk, but its metadata still appears on the StorNext file system. Data is truncated up to 1.5 times faster and archived 20% faster than previous versions.

The Archive Vault feature is available in the Quantum StorNext AEL 6000 and Scalar i6000 libraries.

The large files generated by compute-intensive analytical applications such as genetic sequencing also require accelerated access over the LAN. To meet this requirement, Quantum has developed the Distribute LAN Client, which provides data access up to 50% faster than that of traditional CIFS and NFS protocols.

Within StorNext, files can be organized by project, with the team members involved. Directory quotas can be enabled to help manage storage capacity by project.

High-resolution video files can be ingested and stored in unsequenced order. When files are stored in sequence read performance improves, so StorNext 4.2 has added an Allocation Session Reservation feature, which stores files in sequence. The company has also added a tool called Vidiomap, which reorders the sequence of files to make them easier and faster to restore.

Finally, metadata migration among hardware platforms has been improved.

StorNext 4.3 is available now.

Deni Connor is founding analyst for Storage Strategies NOW, an industry analyst firm that focuses on storage, virtualization, and servers.

Extending core virtualization concepts to storage, networking, I/O, and application delivery is changing the face of the modern data center. In the Pervasive Virtualization report, we discuss all these areas in the context of four main precepts of virtualization. (Free registration required.)