Leveraging filesystem snapshots to provide copies of production data for testing
Developers hate working with test data. It isn't representative of the working dataset. However, providing multiple full copies of production is slow to refresh and consumes lots of disk. This talk will show how you can use filesystem snapshots to produce copies of your production environment without service interruption and with minimal overhead.
Some of the advantages of using the proposed strategy over a plain dump/restore include:
* Nightly refreshes of the copies of prod with mere seconds of dev downtime
* Provide multiple copies using only slightly more than the space of one copy
* Revert a copy back to the nightly refresh almost instantly
* Provide a new copy almost instantly, using almost no additional disk
* No production downtime involved
A few methods of snapshotting will be discussed including LVM2 (the Linux Volume Manager), btrfs, and SAN/NAS.