Installing like a network switch, our products use deep-packet inspection to keep track of file activity instead of a traditional overlay file system or messy stub files. If servers and clients are working with a “D:” drive or specific mount point name, no matter where the data has been migrated, servers and clients continue to work with the same “D:” drive or mount point name. In other words, ZERO workflow changes.
Metadata is data describing data. A rough analogy would be looking at thumbnails of your pictures in a directory. Metadata tells applications the status and state of data. Things like: last access time, file size, when was the last time a file was saved, or is a file open or closed.
As data growth rates have exploded, metadata requests being made to storage systems have skyrocketed. It is common to have over eighty percent of all transactions flowing in and out of a storage system to be metadata requests. The problem is storage systems are not very efficient at responding to metadata request because they were optimized to move data in and out – not respond to queries about the state of data. As metadata activity increases storage performance plummets.
Outperform an all-flash array with your existing storage. The NSC-055s Network-based Metadata Accelerator enhances the performance of existing storage by off-loading metadata processing. Being as much as 10x faster than the fastest all-flash storage array processing metadata, our customers typically experience a significant reduction in workload processing times. Processing costs are reduced and users experience improved response times.
Metadata is served on the network directly out of memory
Today, visibility, mobility, and automation are missing across different storage tiers in data management.
When you can’t see what data is active and what data is inactive, a lot of inactive data ends up staying on expense storage. Our NSC-110 Network-based Storage controllers contain an in-memory metadata map that knows the state and location of all data. This rich data store makes it simple to automatically place data on the most cost-effective tier.
Transparently migrate inactive data off your existing storage to low-cost cloud or object storage. The NSC-110 Network-based Storage Controller also off-loads metadata processing while continuously migrating inactive data to a cloud or low-cost storage tier without requiring a slow gateway, stub files, or a complex new file system. With the NSC-110 all migrated data keeps its always on profile and appears local for instant recall. Get the best of both worlds with our controllers: Active data performance is significantly improved and inactive data storage costs are dynamically reduced.
Inactive data is continuously migrated to low-cost cloud storage.