This week Adam talks with Andy Klein from Backblaze about hard drive reliability at scale.
Leave us a comment
Changelog++ members get a bonus 15 minutes at the end of this episode and zero ads. Join today!
Sponsors:
Featuring:
Show Notes:
We love 45Drives. If you need an intro to them, let us know.
Something missing or broken? PRs welcome!
Timestamps:
(00:00) - This week on The Changelog
(01:45) - Sponsor: DevCycle
(04:32) - Start the show!
(07:01) - 10 stories from 10 years
(14:39) - Software at the storage layer
(17:18) - S.M.A.R.T. predictive analysis
(20:49) - Digging into S.M.A.R.T. stats
(24:08) - Causes of High Fly Writes
(25:00) - Collecting and storing S.M.A.R.T. data
(29:10) - Sponsor: Sentry
(33:19) - Many DCs around the world
(37:24) - Burn-in testing
(40:45) - Dollars per failure and leveraged buying
(43:35) - The Backblaze Storage Pod
(51:13) - Storage servers are HEAVY!
(57:11) - 52Us, 12 high @ 150lbs each
(59:34) - Sponsor: Postman
(1:03:12) - How does backblaze buy drives?
(1:09:12) - Determining the size of the drive
(1:12:08) - What is your capacity/usage limit?
(1:14:39) - How should homelabers buy drives?
(1:25:43) - SSDs as boot drives
(1:30:41) - HDD vs SSD cost
(1:33:26) - How Adam buys drives
(1:36:20) - Closing out the show
(1:38:26) - Outro
Create your
podcast in
minutes
It is Free