Get Connected |
BlueArc Storage Pools virtualize physical SAN storage and enable automatic growth of file systems as required using pre-defined administrator rules, virtually eliminating storage provisioning. File systems get the benefit of all the disk spindles within the storage pool for optimal performance, but only use capacity as they grow, allowing for improved storage utilization. Storage pools can be shared between clustered Titan and Mercury servers, further optimizing storage management and provisioning.
As most storage administrators don't know exactly which file systems will grow fastest or to what capacity, storage pools solve this by providing a shared storage pool with file system capacity added as needed, automatically, dynamically, and without manual storage provisioning. With storage pools, physical SAN storage becomes truly virtual SAN, allowing for improved utilization and management. Each file system is created with a minimal level of capacity, and is grown at user configurable increments or chunks. As the storage pool is consumed, additional physical storage can be added to the storage pool to allow for continued seamless storage growth. BlueArc servers can scale to support the largest enterprise NAS configurations with support for storage pools up to 512TB and growing. Multiple storage pools using BlueArc's Tiered Storage capabilities are still supported, enabling the administrator to choose the right disk technology for their storage requirements.
BlueArc's storage servers have always created large file systems across multiple RAID controllers to create a high performance back-end SAN that can sustain the throughput and transactional performance of the server. With the addition of storage pools, Titan and Mercury servers can now share storage pools, creating file systems across the entire pool of storage, further improving the performance, scalability, and flexibility of a BlueArc storage system. Also with storage pools, administrators can now monitor file system performance and move file systems instantly between servers for improved workflow and flexibility. All of these separate file systems can be seen as one large unified name space with BlueArc’s Cluster Namespace (CNS) capability. Users can access any file system, directory, or file from any Titan or Mercury node in the cluster, allowing load balancing and data migration to be transparent to users, and freeing administrators from having to "migrate users over to another server" as requirements change and users outgrow their system.
Another benefit of Storage Pools is the ability to support snapshots at the more granular level of an individual file system versus taking snapshots of the entire storage pool. Snapshots save an exact point in time copy of data to protect against file changes or deletions, allowing new data to be written but not over-write the point in time copy. With the ability to support multiple file systems within a storage pool, administrators can now create separate file systems for active working space, which requires many snapshots to preserve user's files, and separate file system for scratch space which does not need the protection of snapshots and may be so dynamic that it would use up disk capacity unnecessarily. This new level allows for improve data protection while minimizing disk utilization.
In real world implementations, storage pools nearly eliminate storage provisioning and improve storage utilization. For example, an administrator with a 10 terabyte storage system may have 4 projects, each requiring a portion of this storage over time, but he has unclear information from the users on the rate of growth or final capacity of their projects. The administrator can simply create four separate file systems (and unify them with Clustered Namespace if desired). Each of the four file systems will start with administrator-defined minimal capacity, incremental chunk size (the amount added during growth triggers), capacity triggers to initiate growth and the maximum capacity if desired. These file systems are then mounted by the users and their capacity will automatically grow per the defined rules. The 10 terabytes will now be more efficiently used than had each project received equal capacity up front, as most projects don't grow smoothly. System monitoring tools provide information on usage, growth and performance of each file system, which can be used to plan out future storage purchases, which can be easily added to the existing storage pool as needed.