Working with a customer recently with a large data store of 75TB of video data, the video data needs to be accessible in two different locations.
Location one is the primary location which ingests/add's content to the filesystem as well as performs a playout function for the environment.
Location two is a secondary site which needs a copy of the data for DR purposes but they would also like to read the data for local playout functions
The question around using SyncIQ to solve a replication, content distribution quandary that the customer is facing.
SyncIQ creates a secondary copy on the target cluster. This copy is read only and normally not being used, however it is possible to create a read-only share on the secondary cluster pointing to the snapshot directory/pointer of the replica.
If the SyncIQ policy is creating multiple updates per day you can also automate the share definition via command line to point at the new share and the data becomes visible to the SMB share.
The destination filesystem will have a .snapshot_timestamp copy of the share. Using a shell script you can check for the latest snapshot instance and redirect the share definition. Testing this on a Windows SMB mount the windows file explorer updated the contents with the new read-only data.
Handy to use if you only need R/O data at the secondary site.
We have a small development Isilon cluster which I would like to access from our development HPC cluster, and I used SyncIQ ot make a copy of our production HPC data from our production Isilons. The sync policy is manual, so it will not be run on any schedule.
I have two questions, (1) how can I make the copy on target read/write for use with our dev HPC, and (2) is there any way to occasionally overwrite and refresh these files from production?
You can check out some of the functions in recent Superna versions to help customers create "Writable copies of SyncIQ" I haven't personally tested but it looks like it would help with the automation,
ClaudeComputing
1 Message
765
0
Posted April 20th, 2017 16:00
We have a small development Isilon cluster which I would like to access from our development HPC cluster, and I used SyncIQ ot make a copy of our production HPC data from our production Isilons. The sync policy is manual, so it will not be run on any schedule.
I have two questions, (1) how can I make the copy on target read/write for use with our dev HPC, and (2) is there any way to occasionally overwrite and refresh these files from production?
Thanks