[08:58:35] <elukey>	 I found this whitepaper about the Hops feature store
[08:58:36] <elukey>	 https://uploads-ssl.webflow.com/5e6f7cd3ee7f51d539a4da0b/5e6f7cd3ee7f519fdfa4dadb_feature%20store%20whitepaper%201-0.pdf
[08:58:59] <elukey>	 it talks in more detail about how the use spark etc.., and they mention an integration with kerberos
[09:03:31] <elukey>	 https://docs.hopsworks.ai/latest/
[09:24:41] <elukey>	 ok so https://docs.hopsworks.ai/latest/integrations/spark/ makes it a bit more clear
[09:25:21] <elukey>	 IIUC the Feature store can use Spark running on Hadoop, but it offers its own Hive Metastore 
[09:25:33] <elukey>	 that may use S3/HDFS/etc.. for its external tables
[09:25:44] <akosiaris>	 elukey: We are doing the rdb switchover for ORES in a few
[09:25:54] <elukey>	 akosiaris: perfect!
[09:26:16] <akosiaris>	 I am hoping that this will avoid an outage, I 'll do it in a rolling restart way
[09:27:51] <elukey>	 akosiaris: so basically pointing all the celery workes to another rdb?
[09:27:55] <elukey>	 (n00b question)
[09:29:33] <akosiaris>	 yup
[09:29:40] <akosiaris>	 redis 6, debian bullseye and new hardware
[09:29:53] <akosiaris>	 it's not just ORES of course, it will just happen along the rest of the apps
[09:31:19] <elukey>	 ack!
[09:33:12] <elukey>	 back to the spark thing, I am confused about hive, it may not be integrated
[09:33:22] <elukey>	 I think I'll send an email to them to ask for clarifications
[09:35:44] <akosiaris>	 the all spark is definitely integrated. Deeply into Optimus's chest
[09:40:30] <elukey>	 I really hope not :D
[10:10:31] <elukey>	 ok https://community.hopsworks.ai/t/feature-store-hardware-requirements/466
[10:10:41] <elukey>	 let's see what answers we get :)
[11:21:50] <wikibugs>	 10Machine-Learning-Team, 10artificial-intelligence, 10editquality-modeling, 10Turkish-Sites: Update Turkish Wikipedia's labeling campaign for 2020 - https://phabricator.wikimedia.org/T257359 (10kevinbazira) Thanks to @elukey, I was able to get access using both `ssh ores-misc-01.ores-staging.eqiad1.wikimed...
[11:45:56] <wikibugs>	 10Machine-Learning-Team, 10artificial-intelligence, 10editquality-modeling, 10Turkish-Sites: Update Turkish Wikipedia's labeling campaign for 2020 - https://phabricator.wikimedia.org/T257359 (10elukey) @kevinbazira on `ores-misc-01`  the root partition is full :(  ` (venv) kevinbazira@ores-misc-01:~/editqu...
[12:53:22] <elukey>	 kevinbazira: o/
[13:00:54] <wikibugs>	 10Machine-Learning-Team, 10artificial-intelligence, 10editquality-modeling, 10Turkish-Sites: Update Turkish Wikipedia's labeling campaign for 2020 - https://phabricator.wikimedia.org/T257359 (10kevinbazira) Thank you so much @elukey, I moved my work to /srv/home/kevinbazira and cleared /home/kevinbazira/....
[13:01:22] <kevinbazira>	 elukey o/
[13:03:08] <elukey>	 ah it worked! didn't see the update, goooood!
[13:04:44] <kevinbazira>	 Yep, it took some time to complete the process but it did work. Thanks to you :)
[13:13:57] <elukey>	 \o/
[16:01:34] <wikibugs>	 10ORES, 10artificial-intelligence, 10articlequality-modeling, 10drafttopic-modeling, 10Machine-Learning-Team (Active Tasks): ORES deployment - Spring 2021 - https://phabricator.wikimedia.org/T278723 (10Halfak) @Majavah after some thought, I think it would be great if you would look into Envoy for Beta....
[16:38:11] <wikibugs>	 10ORES, 10artificial-intelligence, 10articlequality-modeling, 10drafttopic-modeling, 10Machine-Learning-Team (Active Tasks): ORES deployment - Spring 2021 - https://phabricator.wikimedia.org/T278723 (10Halfak) For reference, this is where ORES references Envoy (localhost:6500).
[21:33:13] <wikibugs>	 10Lift-Wing, 10artificial-intelligence, 10Machine-Learning-Team (Active Tasks): Create a tutorial for deploying a model on toolforge - https://phabricator.wikimedia.org/T281317 (10ACraze)
[21:38:04] <wikibugs>	 10Lift-Wing, 10artificial-intelligence, 10Machine-Learning-Team (Active Tasks): Create a tutorial for deploying a model on toolforge - https://phabricator.wikimedia.org/T281317 (10ACraze)
[21:39:10] <wikibugs>	 10Machine-Learning-Team: Proof of concept: dockerized model served over CloudVPS - https://phabricator.wikimedia.org/T265066 (10ACraze) 05Open→03Declined Talked about this with @calbon today and determined that toolforge is probably a better place to direct folks to for deploying community models at this tim...
[22:02:04] <wikibugs>	 10Lift-Wing, 10artificial-intelligence, 10Machine-Learning-Team (Active Tasks): Create a tutorial for deploying a model on toolforge - https://phabricator.wikimedia.org/T281317 (10ACraze)
[22:50:24] <wikibugs>	 10Lift-Wing, 10Machine-Learning-Team (Active Tasks): Naming convention for the model storage structure - https://phabricator.wikimedia.org/T280467 (10ACraze) @klausman I think most of our models should be ok with minute-granularity. The only case I can think of where it could be problematic is if we train two...