Live data from Hacker News

Viewing profile — mbjorling

mbjorling

HN member
Joined
Sun, Apr 14, 2013, 7:26 PM UTC
HN karma
72
Public activity
14 items

About mbjorling

https://www.linkedin.com/in/matiasbjoerling/

Recent public activity

  1. comment
    Comment #44739165

    Are there any storage-related roles? Will Oxide redefine storage as well?

  2. comment
    Comment #31497590

    There is more technical information at zonedstorage.io which also offers drives for academia and open-source projects. https://zonedstorage.io/docs/community/devices

  3. comment
    Comment #20850851

    A presentation of the SSD benefits is available here: https://m.youtube.com/watch?v=9yVWb3rbces (Full disclosure - my talk at OCP 2019)

  4. comment
    Comment #20850813

    True - the great thing is that we have been optimizing for this type of interface for the last decade. I.e., due to the benefits of making writes sequential (both for HDDs and SSDs…

  5. comment
    Comment #20850788

    As SSDs get denser, moving to QLC/PLC, the number af writes to the same place on media gets to the low hundreds. By using zones, the total amount of available host writes increases…

  6. comment
    Comment #20850708

    Half of the world bits from HDDs are estimated to be on SMR in 2023 - the gains are significant when deploying at scale. For SSDs, it gets even more fun as zones aligns with the ch…

  7. comment
    Comment #19643144

    It is worth mentioning that the Linux kernel has a new kernel API (io_uring) that changes the whole argument around using libos designs. With the new io_uring library (available wi…

  8. comment
    Comment #12233432

    I like to follow The morning paper by Adrian Colyer. He writes a summary of an influential CS paper each day and sends it out on his e-mail list. https://blog.acolyer.org/

  9. comment
    Comment #7091168

    One thread issuing IOs: A reduction of 2x in the IO path latency isn't unusual. The overhead of the code path drops from 5us to around 2us. When there's multiple IO threads, the ga…

  10. comment
    Comment #7090423

    This statement should be seen as there's no way to scale the old block layer to new devices. To current SSDs, its already useful, in that it decreases latency and CPU usage for cur…

  11. comment
    Comment #6733232

    and the merge message http://www.phoronix.com/scan.php?page=news_item&px=MTUxNDQ

  12. story
  13. comment
    Comment #5547957

    It's hard for people to get data on flash chips / prices without signing an NDA. He probably found the data from various places and stitched it together from that.

  14. comment
    Comment #5547950

    It's one way to look at it. However, having an object interface toward the SSD does not solve the problem of variability that the author mentions. The variability is caused by the …