Native Support for Data Virtualization in NFS 4.2
Posted in news
Primary Data today announced that our open standards Parallel NFS (pNFS) contributions have been accepted into the NFS 4.2 standard. These enhancements give Linux clients and application servers native support for DataSphere’s data virtualization and mobility capabilities. Built from Primary Data’s previous NFS contributions, these open standards have been integrated into the Linux operating system kernel by Primary Data Principal System Architect and Linux NFS client maintainer Trond Myklebust.
Enhancements to the Parallel Network File System (pNFS) Flex File layout enable clients to provide statistics on how data is being used, and the performance provided by the storage resources serving the data. DataSphere makes this information intuitively visible to customers through the UI and reports. Customers can use this information to create Service Level Objectives in DataSphere, a software defined platform that virtualizes and non-disruptively moves data to automatically match evolving data needs with different storage resources as business needs change.
Following is a summary of key NFS 4.2 enhancements, many of which are featured in DataSphere:
· Server-Side Clone and Copy enables cloning and copying of files by the storage server. DataSphere customers will use this feature to optimize and offload snapshot and clone operations.
· Application Input/Output (I/O) allows applications to inform the storage server of expected I/O behavior
· Sparse Files allows files to be space efficient and use placeholders rather than zeros to improve storage efficiency
· Space Reservation allows storage servers to reserve space without writing data to protect against unexpectedly running out of capacity
· Application Data Block (ADB) enables block on file storage management implementations
· Labeled NFS enhances security by enabling clients to enforce data access rights
· Layout Enhancements deliver significant improvements to pNFS capabilities, supporting data mobility and I/O stats per file. DataSphere uses this information to make intelligent data mobility decisions.
Trond Myklebust is excited to make these features openly available to the Linux community, stating, “As the world’s most popular operating system was even recently embraced by Microsoft, Primary Data is excited to be advancing NFS 4.2 as a new open standard, and integrating these enhancements into the Linux kernel. These contributions deliver support for advanced data management across different storage types, helping to ensure Linux continues to evolve as the most powerful, feature-rich OS available thanks to the work of the global community.”
Have questions? Contact us at firstname.lastname@example.org.