Tip:
Highlight text to annotate it
X
Hello, I'm Andrey Tychkin with EMC Isilon.
In this video we'll talk about the deduplication capabilities of OneFS 7.1 and the SmartDedupe
feature.
SmartDedupe is a licensed module of OneFS that allows us to gain space savings by deduplicating
common data.
SmartDedupe is non-intrusive to protocol operations because it runs as a post-process job and
the SmartDedupe job runs at low priority to minimize production impact to the cluster.
SmartDedupe can deduplicate files over 32k in a given path.
The deduplication occurs at the block level, but we can submit files and directories to
be deduplicated at a specific directory path.
When the SmartDedupe jobs first runs, it looks through the files in a given path and builds
an index of all the blocks in the path.
It then looks through that index searching for duplicate blocks.
Once a pair of blocks is found, it looks at all the adjacent blocks and it deduplicates
duplicate content by creating a copy of the data and then updating the blocks with pointers
to a copy of that data.
Let's take an example.
Let's say we have an enterprise IT organization that uses Isilon for their Home directory
storage and they're considering SmartDedupe as a tool to make their storage more efficient.
The storage administrator will first run a SmartDedupe Assessment job on their Home directory
path.
The assessment job does not require a license and will produce a detailed report showing
potential percentage space savings, the number of file and folders in a given directory path,
and the time it took to run the job.
After reviewing the report, our storage administrator finds notable space savings and activates
SmartDedupe on their cluster.
Now, they can either run the job manually or he can schedule the job to run the same
as any job in OneFS.
For most clusters, we recommend to start a new SmartDedupe job every ten days.
If you have questions, or want to implement OneFS 7.1 features in your environment, please
contact your account team.
Thank you for watching.