* [ ] Debnandini: communicate that we’re doing an IMBH search to the CBC chairs
- Divya: Not done yet
* [ ] Debnandini: communicate that we’re doing an IMBH search to the CBC chairs
- Debnandini: For next week, I'll have a presentation at the CBC call and circulate slides beforehand.
## Agenda / Minutes
* Announcements (5 minutes)
- Please check the rota for next week's call
- Confirmation of next week's focus session
- We will be doing review but please put things for focus sessions.
- Do we want to restructure the West call?
- We will be doing review but please put things for focus sessions.
- Do we want to restructure the West call?
- Rachael: There's overlap with weekly sprint. If you have any suggestions or comments, send a message in chat and we can discuss.
- Please double-check what accounting tags you're using. Make sure you're using online/offline and production/dev correctly
* Last week's East call
- Prathamesh: Victoria, Kipp, and Chad discussed injection file structure.
* Quick updates (45 minutes)
- Operations (5 minutes)
- LL CBC operations
- Leo: HTTP was hanging over the weekend. I'm going to meet with others to discuss it. We also had issues with scald jobs hanging, causing messages about all jobs being down. Another issue was fixed where emails reporting that all jobs are down were not being sent. Jacob had a problem with influx queries. We also got an event.
- Rachael: An issue was made fro the HTTP issue, but can you make one for the influx test?
- Leo: Will do.
- Divya: Over the last week, me, Leo, Prathamesh, and Chad have been looking into SSM significance assignment in Charlie. The variance seems to be smaller than the variance of signal model in All-Sky, which shouldn't be the case. If anyone wants to help in the next day or two, I'm putting information here: https://git.ligo.org/gstlal/projects/-/issues/782
- Shio: Some responder positions for future rotas are blank. Sign up if you can!
- LL IDQ operations
- Rachael: This is running great. We have the first 4 offline chunks ready.
- O4 Dev (30 minutes)
- MDC for O4b prep
- Prathamesh: We set up an MDC for practice even though it will end at the end of this week. We have a plan for what to do for the next MDCs. We changed the naming convention to follow the computing group convention, making the current one MDC23. After this we have 4 MDCs. If anyone has O4b features they want to test, let us know ASAP. We're going to add bankchisq and snr optimizer for MDC24. The end of this (November 23rd) is the soft deadline for testing new features. MDC25 is ending Jan 3 and is the hard deadline for new features. MDC 26 will be data collection, sensitivity estimates, and comarisons with O4a. MDC27 will be for finishing review and emergency patches.
- Leo: I think LR-based sampling can be ready by MDC25. The KDE rework might be a long shot but we want to push for that.
- Prathamesh: Thanks, we can work on those.
- Rachael: It would be good to make a diagram showing when different analyses will come online and be decomissioned, just like the one for O4a.
- Rachael: Ron, is it okay that we launched an MDC using real time nodes, but not specifying that they're gstlal online jobs?
- Ron: That's fine, but there might be a delay for jobs to start if the cluster is under high load.
- Low latency integrated testing and Monitoring
- Template bank
- Injection file format
- Victoria: Chad and I are looking into a descrpancy between waveforms generated using the O3 and O4 formats.
- LR background sampling
- Leo: I've been setting up online rerank tests for bankchisq. This should be ready in a couple of days.
- offline DAG
- DQ dev
- HM search
- Exploratory development (5 minutes)
- Cody: I haven't had time to finish fixing the problem where jobs don't write their snapshot to disk when they shutdown. Can anyone take that over?
we may not have restarted MDC analyses and kept using the same rundir and analysis.
13:56:21 From Rachael Huxford To Everyone:
Reacted to "we may not have rest..." with 👀
13:56:27 From Rachael Huxford To Everyone:
Replying to "we may not have rest..."
Oh that’s very true
13:56:30 From Leo Tsukada To Everyone:
Revival of renee and esme?
13:58:11 From Rachael Huxford To Everyone:
Replying to "Revival of renee and..."
I would save those for pre-engineering runs tbh just to not confuse things
13:58:18 From Leo Tsukada To Everyone:
Replying to "Revival of renee and..."
fair
13:58:20 From Rachael Huxford To Everyone:
Replying to "Revival of renee and..."
But there are plenty of other characters 😛
13:58:38 From Shio Sakon To Everyone:
The mdc that WE called mdc12 (= the last one before ER/O4a) probably corresponds to “Past 19” in https://wiki.ligo.org/Computing/DASWG/O3EndToEndReplay
The MDC that the o4b team is calling mdc23 is “past 23” in the link.
13:58:47 From Rachael Huxford To Everyone:
Reacted to "The mdc that WE call..." with 👀
13:58:51 From Rachael Huxford To Everyone:
Reacted to "The mdc that WE call..." with 👍
14:00:09 From Amanda Christine Baylor To Everyone:
👍🏻
14:00:15 From Divya Singh To Everyone:
Replying to "Revival of renee and..."
Jane and Alec
14:00:26 From Rachael Huxford To Everyone:
Replying to "Revival of renee and..."
I was thinking of the brothers
14:00:55 From Rachael Huxford To Everyone:
Replying to "Revival of renee and..."
Emmett and Jasper
14:01:47 From Urja S. Shah To Everyone:
Reacted to "Emmett and Jasper" with ❤️
14:03:15 From Divya Singh To Everyone:
Sad.
14:03:23 From Richard George To Everyone:
Reacted to "Sad." with 😱
14:03:26 From Shio Sakon To Everyone:
Reacted to "Sad." with 😱
14:03:35 From Richard George To Everyone:
Reacted to "Sad." with 🧛
14:03:56 From Shio Sakon To Everyone:
Ah the git timeline thing!
14:04:41 From Jim To Everyone:
just give us more cluster
14:04:51 From Shio Sakon To Everyone:
Replying to "Ah the git timeline ..."
I don’t remember the link …
14:04:51 From Amanda Christine Baylor To Everyone:
What about Nemo?
14:05:03 From Cody Messick To Everyone:
Replying to "What about Nemo?"
hahahahaha
14:05:05 From Cody Messick To Everyone:
Replying to "What about Nemo?"
good joke
14:05:30 From Amanda Christine Baylor To Everyone:
Replying to "What about Nemo?"
🐡
14:05:41 From Cody Messick To Everyone:
Reacted to "just give us more cl..." with 👆
14:06:44 From Duncan Meacher To Everyone:
whats wrong with nemo?
14:07:04 From Amanda Christine Baylor To Everyone:
Reacted to "whats wrong with nem..." with 👆🏻
14:09:25 From Divya Singh To Everyone:
Replying to "Revival of renee and..."
Jane and Alec have psychic powers 🧠
14:09:27 From Divya Singh To Everyone:
Reacted to "Emmett and Jasper" with ❤️
14:12:29 From Prathamesh Joshi To Everyone:
done for reranks
14:13:06 From Divya Singh To Everyone:
This is Charlie related: What is the best way to choose the autocorrelation length? Ryan mentioned it depends on the sample rate. But I’m not sure of the dependence.
14:13:26 From Cody Messick To Everyone:
@Divya: I'd look at whatever is used for BNS
14:13:43 From Cody Messick To Everyone:
and then compesnate for different sample rates
14:14:05 From Cody Messick To Everyone:
i.e. if BNS is 701 at 2048 Hz (350 on either side), then at 1024 Hz you'd want to use 351 (175 on either side)
