4 ms·

Yev from Backblaze here -> at a conference today (JNUC2017 for those curious) but I'll be checking in here to answer questions should any arise!

• linuxguy2 9y ago
  These reports are always well written and interesting reads. Thanks for continually producing them!
  • atYevP 9y ago
    Thank you! Andy (the guy that writes them) will be happy to hear the kudos!
• srcmap 9y ago
  Do you have any info on what temperature your drives are running on normally? Love to see some performance stat to see if read /write IOP, MB/second, etc change over time on per model base. Other than complete failure, do you collect check the any other early warning failure data such as ECC change over the life time of the drive? --- Would be a fun big data exercise for an intern to work on.
  • atYevP 9y ago
    Hey there! We've written this about temperature before: https://www.backblaze.com/blog/hard-drive-temperature-does-it-matter/ https://www.backblaze.com/blog/hard-drive-temperature-does-i.... As for the performance other than total failure, we do keep an eye on SMART stats -> https://www.backblaze.com/blog/hard-drive-smart-stats/ https://www.backblaze.com/blog/hard-drive-smart-stats/ and https://www.backblaze.com/blog/what-smart-stats-indicate-hard-drive-failures/ https://www.backblaze.com/blog/what-smart-stats-indicate-har... so we do monitor their general health as well.
    • srcmap 9y ago
      Nice info. Thanks. Also what's the usage patterns for your HDDs? read/write/spin down/sleep time in term of life time of the HDD. I assume mostly write, some read. But also very low percentage is active?
      • atYevP 9y ago
        Mostly it's write - get fairly full - then sit until read :D
  • haikubrian 9y ago
    BrianB from Backblaze here. Yes, we collect temperature data. If you download the dataset, you'll find a daily sample of the SMART data (including temperature) for every drive.
    • atYevP 9y ago
      Hey Brian! /wave