#lumosql Minutes

Meeting started by dan at 2022-02-07 10:05:19+0000 (full logs).

Meeting Summary

  1. Prologue (dan, 10:05:19)
    1. ATTENDEE: Dan Shearer (dan, 10:05:39)
    2. ATTENDEE: Gabby (gabby_bch, 10:06:03)
    3. ATTENDEE: Ruben (rubdos[m], 10:06:38)
    4. ATTENDEE: Claudio (Labhraich, 10:06:42)
  2. Background (dan, 10:07:05)
    1. INFO: Ruben has offered a powerful 4-machine cluster at VUB for benchmarking (dan, 10:07:25)
    2. INFO: Claudio and Dan agreed at https://lumosql.org/meetings/lumosql.20220205.1007.html to provide consolidated benchmarks to Gabby (dan, 10:08:31)
    3. INFO: Right now, some benchmark.sqlite files exist on Dan and Claudio's machines (dan, 10:09:06)
    4. LINK: https://lumosql.org/src/lumosql/doc/trunk/README.md "Quickstart: Using the Build and Benchmark System" (dan, 10:11:06)
    5. LINK: https://lumosql.org/src/lumosql/doc/trunk/doc/lumo-build-benchmark.md Full Documentation so far (dan, 10:11:54)
    6. ACTION: Ruben to write Kubernetes infra scripts to run benchmarks (rubdos[m], 10:14:57)
    7. ACTION: Ruben to run those scripts on the VUB cluster (rubdos[m], 10:15:14)
    8. ACTION: Ruben to host the results via HTTPS (rubdos[m], 10:15:20)
  3. Goals for Cluster Use (dan, 10:16:46)
    1. INFO: The sqlite benchmark databases will be run on homogonous hardware so our goal is to compare different configs with identical hardware (dan, 10:17:34)
    2. INFO: SQLite is single-threaded so we probably won't see cluster load-type errors in the data (dan, 10:18:42)
    3. ACTION: Ruben to figure out whether our fourth server contains the same CPU as the three others. We bought that separately and is 2U instead of 1U (rubdos[m], 10:18:53)
    4. INFO: LumoSQL can test a zillion dimensions and will be something really new in the field of databases (dan, 10:19:36)
    5. INFO: I propose we do not test a zillion dimensions this time (dan, 10:19:53)
    6. INFO: LumoSQL benchmarking tries to capture the details of the environment it is running on. This is so we can get benchmarks from random people all over the world. (dan, 10:23:38)
    7. ACTION: Ruben to document servers in the kube-benchmarking suite. (rubdos[m], 10:23:44)
  4. Big Picture (dan, 10:35:06)
    1. INFO: LumoSQL people (Claudio and Dan so far) believe that SQLite as deployed to trillions and LMDB as deployed to billions do not necessarily work completely the way their respective excellent developers think (dan, 10:36:05)
    2. INFO: The scale matters - hardware decisions and electricity consumption when multiplied by these very high numbers, as a start (dan, 10:37:05)
    3. ACTION: Dan to speak to the SQLite and LMDB teams/authors as soon as we have a repeatable test framework (dan, 10:37:52)
    4. INFO: The best result would be if these teams are running the benchmarking themselves and contributing what they get (dan, 10:38:16)
Meeting ended at 2022-02-07 10:41:20+0000 (full logs).

Attendees

  1. Labhraich (aka Claudio) - 23 lines
  2. dan (aka Dan Shearer) - 52 lines
  3. gabby_bch (aka Gabby) - 1 lines
  4. lumosql-meetbot - 4 lines
  5. rubdos[m] (aka Ruben) - 37 lines

Action Items

  1. Ruben to write Kubernetes infra scripts to run benchmarks
  2. Ruben to run those scripts on the VUB cluster
  3. Ruben to host the results via HTTPS
  4. Ruben to figure out whether our fourth server contains the same CPU as the three others. We bought that separately and is 2U instead of 1U
  5. Ruben to document servers in the kube-benchmarking suite.
  6. Dan to speak to the SQLite and LMDB teams/authors as soon as we have a repeatable test framework

Action Items by Attendee

Generated by HCoop Meetbot v0.4.1 (15 Nov 2021)