Tweets

You blocked @glcst

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @glcst

  1. Jan 31

    I've been dealing with SQL and CQL for so long, that I know instinctively read the names of the io_uring data structures as "structured completion entry" and "cassandra completion entry".

    Undo
  2. Jan 30

    Getting started with io_uring for sockets for seastar. Good that I am doing this per-merge of the storage bits, because some design decisions will have to change to accommodate both more elegantly. Soon seastar will need only one system call

    Undo
  3. Jan 29

    impeach cmake!

    Undo
  4. Jan 29

    I lost (am losing) so much time with cmake these past days, it's not even funny

    Undo
  5. Jan 29

    Wondering how are the early results for the io_uring backend for seastar? 50% faster in the first benchmark (workload is small 512-byte reads with iodepth of one, competing for dispatch time against a CPU-bound constant workload)

    Undo
  6. Jan 29

    Discord is migrating from Cassandra to Scylla. It caused them high latencies because they had to read more from their current Cassandra cluster to siphon the data. Easily solved by slowing down. But wait until you see how low those latencies will be once the migration finishes!

    Undo
  7. Jan 28

    Initial implementation of io_uring for seastar posted. Need to do more testing, polish some details, and find some annoying bugs. But looking good!

    Undo
  8. Jan 24

    I meant to say those are in the kernel.

    Show this thread
    Undo
  9. Jan 24

    BTW, , liburing does not seem to have any test using non-vectored writes, and when I try to use it, it crashes with a coredump. The last lines of the backtrace are io_write() -> io_setup_async_rw(), and the faulty addr is 0x0000000000000088

    Show this thread
    Undo
  10. Jan 21

    Scylla 3.2 is a great release, finally bringing LWT-style transactions to Scylla and bridging the last big gap between Scylla and Cassandra. We're also going above and beyond with a table-based CDC implementation

    Undo
  11. 21 Dec 2019

    Stopping now for some time off to clear my mind. Eager to come back in January!

    Undo
  12. Retweeted
    18 Dec 2019

    . Editor's Choice! On Scylla: "It’s the rare product that exceeds my expectations." supercharged

    Undo
  13. 5 Dec 2019
    Undo
  14. 5 Dec 2019

    With availability of hardware like EC2 M6g instances and startups like Nuvia, I do consider that this is a game-changing for Arm servers

    Undo
  15. 20 Nov 2019

    The concepts in this article are valid for pretty much every database, although are things we see a lot our users struggling with. How to determine the best number of connections and threads to use for your clients? Should I retry requests that timed out? Answers below!

    Undo
  16. 18 Nov 2019

    If you missed the Scylla Summit, this is a summary of what happened there. It was a great conference, and we announced great things for the times ahead

    Undo
  17. Retweeted
    9 Nov 2019

    San Francisco has a number of very fit, technically-brilliant billionaires. It has a crime and safety problem, and it elects politicians that exacerbate this problem. If any place was going to create Batman, it's San Francisco.

    Undo
  18. Retweeted

    On November 12, please join us at Python Meetup to hear our very own Bruno Brodarić and Marija Zelić and learn on "Big Data ETL using Python, Kafka and ScyllaDB". More info:

    Undo
  19. Retweeted
    7 Nov 2019

    . and ScyllaDB combined the fastest baremetal instances with the fastest database to scale a cluster capable of reading 1,000,000,000 rows per second!

    Undo
  20. 7 Nov 2019

    That is right: we + packet just demonstrated Scylla reading 1 BILLION ROWS per second at just 83 nodes.

    Undo

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·