Veritas Storage Foundation Cluster File System enables concurrent file access from multiple servers to provide a flexible, high performance, and highly available platform for shared data in a SAN environment. It's built for commercial transactional workloads and provides a mature single file system schema that is cache-coherent and POSIX-compliant. A cluster-wide volume and file system configuration allows for simplified management. Organizations also benefit from an integrated cluster volume manager that presents every node in the cluster with the same logical view of shared device configurations.
Benefit from storage consolidation
IT organizations that have information-sharing requirements across multiple applications can minimize the need to create redundant copies of information by implementing Veritas Storage Foundation Cluster File System. With the integration of a cluster file system and cluster volume manager, IT organizations can architect applications so that applications residing on multiple server nodes have concurrent access to the same information through shared storage, thus minimizing the need to replicate, copy, or move information. With Veritas Storage Foundation Cluster File System, organizations can achieve near linear scalability by adding up to 32 nodes, all with concurrent access to a single storage pool, greatly increasing the benefits of physical storage consolidation via a storage area network.
Improve application performance and scalability
If an organization selects a scale-out approach, a single cluster file system across all nodes is essential to coordinate read/write access to the storage pool and ensure data integrity. Cluster File System does this and also ensures that performance isn't compromised, with its advanced file-locking capability providing parallel access to the same file from multiple nodes. A critical enabling technology is the advanced file-locking architecture, Multiple Transaction Server, that distributes file ownership across all nodes in the cluster (supporting up to 32 nodes), ensuring near linear scalability without requiring unnecessary metadata communication throughout the cluster.
Minimize application downtime
Veritas Storage Foundation Cluster File System is tightly integrated with Veritas Cluster Server by Symantec, providing a comprehensive application failover solution minimizing both planned and unplanned downtime across all nodes in the cluster, up to 32. In the event of an application failure or a node failure, the application is dynamically migrated to an available node in the cluster without application or business interruption, and eliminates any single point of failure. Furthermore, cluster ownership is dynamically redirected in the event of the "primary node" failure. An additional benefit to a cluster file system architecture is that, as all nodes in the cluster have visibility to a shared storage pool, there is no need to manually mount storage to unique nodes in the event of application or node failures - minimizing application downtime, both planned and unplanned.
Ensure data consistency
In an application architecture where multiple server nodes have access to the same storage pool, cache coherency is of the utmost importance, ensuring data consistency and integrity. There must be a single version of all files in a cluster that are visible to all nodes in the cluster; if each node has "its version" of the file (particularly during writes), it results in corrupt information. Veritas Storage Foundation Cluster File System uses the Veritas Global Lock Manager (GLM) to make writes throughout the cluster atomic. In doing so, even though application performance is maximized by having files cached at the local node level, GLM ensures that during any application write, there is only a single version of that file, preventing data corruption.
Reduce storage costs
With the dynamic storage tiering capabilities of storage foundation, unimportant or out-of-date files can be moved to less expensive storage devices without changing the way users or applications access those files. Policies can be created that will move files based on date created, last time accessed, owner, size, or name. These files will be dynamically moved without having to be taken offline. Because Cluster File System can span multiple volumes, applications/databases can be stored on multiple tiers of storage hardware. This move is completely transparent to the users and applications that own the files - they don't need to know the files have moved. As data is moved to different storage tiers, the policies are centrally managed and dynamic, and support a heterogeneous server and storage infrastructure that require no application, database, or backup/recovery policy modifications.
Ensure data integrityProduct Highlights
When multiple systems/nodes have access to data via shared storage, the integrity of the data depends on internode communication ensuring that each node is aware when other nodes are writing data. When the coordination between the nodes fail, it results in a "split brain" condition - a situation in which two servers try to independently control the storage, resulting in application failure or data corruption. I/O fencing is the method of ensuring the integrity of critical information by preventing data corruption, and Veritas Storage Foundation Cluster File System has implemented I/O fencing using the industry-standard SCSI-3 persistent group reservation technology. This allows a set of systems to have temporary registrations with the disk and coordinate a write-exclusive reservation with the disk containing the data. With I/O fencing, Symantec ensures that errant nodes are "fenced" and do not have access to the shared storage, while the eligible nodes continue to have access to the data.
- Simplify storage administration