Our first Zoom DistSys reading group meeting
We did our first Zoom DistSys reading group meeting on April 1st, Wednesday 15:30 EST. We discussed the Gryff paper.
I didn't have much Zoom experience, and this was very experimental reaching out to the world at large to run a reading group with whomever is interested.
20 people attended. As I was introducing the format, one person starting writing chat messages, saying "this is so boring", etc. He had connected with a phone, and the video was showing him walking probably in a market. This should have been a red flag. The meeting participants asked me to remove him, because he was pinging them and bothering them as well. That was our troll.
I had taken measures to stop zoom-bombing, since I had heard this was an issue.
The meeting took 90 minutes. The presentation from the slides was for 30 minutes. The general discussion was 25 minutes, and we used 25 minutes for the breakout rooms for focused deeper discussion on selected questions. And we had a 10 minutes of wrap up at the end, where we summarized discussion from the breakout sessions.
This meeting was the first time I used the breakout room session. I was not sure how well the breakout rooms would work, but it turned out great. I used automatic assigning of participants to the room and mentioned that I would switch users between rooms if requested. We had 3 breakout rooms, and around 5 people per room. This enabled us to meet each other in the breakout rooms, and we had a more relaxed and productive conversation. One participant in my breakout room had joined from India (2:30 am local time) and another had joined from Germany (11:30 pm local time). It was nice meeting and discussing with people passionate about distributed systems around the globe.
In the wrap up after the breakouts, I performed a poll and learned that 90% of the participants has read the paper before the meeting. This improved the quality of the discussion. It is not easy to understand a paper just by watching a 25 minute presentation.
Here are the documents for offline viewing. But you had to be there; there is a big advantage for live participation.
On the whole, the Zoom meeting went better than I expected. After we get more experience with the mechanics of Zoom, I think things will run better.
I am now hopeful that we can make this meeting sustainable. I hope we will be able to get enough volunteers for presenting papers. The paper presentation is only for 20-30 minutes, because we assume participants read the paper before the presentation. Volunteering to be a presenter is a good commitment to make for learning more about a specific paper/topic. We ask the presenter to give us a summary 4-5 days before the presentation. This gives enough time for others to get a head start for preparing for the discussion.
Here are the next papers, we will discuss in the upcoming meetings:
I didn't have much Zoom experience, and this was very experimental reaching out to the world at large to run a reading group with whomever is interested.
20 people attended. As I was introducing the format, one person starting writing chat messages, saying "this is so boring", etc. He had connected with a phone, and the video was showing him walking probably in a market. This should have been a red flag. The meeting participants asked me to remove him, because he was pinging them and bothering them as well. That was our troll.
I had taken measures to stop zoom-bombing, since I had heard this was an issue.
- Only the hosts and cohosts could share screen.
- I made two co-hosts to help with moderation.
- I had selected the option to disallow joining after removal.
The meeting took 90 minutes. The presentation from the slides was for 30 minutes. The general discussion was 25 minutes, and we used 25 minutes for the breakout rooms for focused deeper discussion on selected questions. And we had a 10 minutes of wrap up at the end, where we summarized discussion from the breakout sessions.
This meeting was the first time I used the breakout room session. I was not sure how well the breakout rooms would work, but it turned out great. I used automatic assigning of participants to the room and mentioned that I would switch users between rooms if requested. We had 3 breakout rooms, and around 5 people per room. This enabled us to meet each other in the breakout rooms, and we had a more relaxed and productive conversation. One participant in my breakout room had joined from India (2:30 am local time) and another had joined from Germany (11:30 pm local time). It was nice meeting and discussing with people passionate about distributed systems around the globe.
In the wrap up after the breakouts, I performed a poll and learned that 90% of the participants has read the paper before the meeting. This improved the quality of the discussion. It is not easy to understand a paper just by watching a 25 minute presentation.
Here are the documents for offline viewing. But you had to be there; there is a big advantage for live participation.
- We used the Google Doc for questions and breakout room discussion, and it worked fine.
- Below is the meeting video. I forgot to start record on Zoom, so the video starts halfway in to the slides. (Going forward, I set the option to start recording automatically.)
On the whole, the Zoom meeting went better than I expected. After we get more experience with the mechanics of Zoom, I think things will run better.
I am now hopeful that we can make this meeting sustainable. I hope we will be able to get enough volunteers for presenting papers. The paper presentation is only for 20-30 minutes, because we assume participants read the paper before the presentation. Volunteering to be a presenter is a good commitment to make for learning more about a specific paper/topic. We ask the presenter to give us a summary 4-5 days before the presentation. This gives enough time for others to get a head start for preparing for the discussion.
Here are the next papers, we will discuss in the upcoming meetings:
- Wormspace: A modular foundation for simple, verifiable distributed systems (SOCC'19)
- Fine-Grained Replicated State Machines for a Cluster Storage System (NSDI'20)
- Scalog: Seamless Reconfiguration and Total Order in a Scalable Shared Log (NSDI'20)
- Unifying consensus and atomic commit (VLDB19)
- Stable and consistent membership at scale with rapid (ATC18)
- Mergeable Replicated Data Types (OOPSLA'19)
Comments