What conferences might I attend that focus on Research Computing and Facilitation?

Research computing is a rapidly growing and developing profession, critical to universities, laboratories and organizations working toward integrating high performance computing (and other compute technologies) with researcher needs. Besides working with existing resources, research computing professionals are often the innovators when new methods and new ideas are needed to find the way forward.

It seems likely that groups of research computing professionals have formed in order to productively share experiences and theories, discuss plans and funding opportunities, engage in multi-institution projects; in short, to embrace all of the crucial elements that comprise a successful professional community.

Thus my question; what are the conferences relevant to Research Computing? Which ones are the big, all-encompassing events, which are more detailed, which are midsize? Are there cross-discipline conferences, and if so, which ones, and what do they concern? Might one gathering favor industry, another benefit academia?

I look forward to discussion around conference opportunities!

I believe the Practice and Experience in Advanced Research Computing (PEARC) conference that begins happens next week is probably the largest gathering of practitioners. There are a few virtual communities that you can participate in as well. One of them is CarCC which hosts monthly video seminars and has a discussion list.

https://www.pearc.org/
https://carcc.org/

I started writing this here, and realized I had too much to say, and wrote it up more properly ->

https://vsoch.github.io/2019/conferences/

I’ll leave the original text here for ease of reading.


I have a different point of view around this whole conferences thing - I don’t have a direct answer to the question about ā€œWhat conferences are greatā€ but rather what I hope for the future.

First let’s discuss the pros of going to a conference. As is stated in the question, there are numerous, from networking (career and growth opportunities) to discovering new ideas, to collaboration. But what about the cons? Well, conferences are:

  • expensive, usually between $300-$600 for a plane ticket depending on where you live, always bordering around high hundreds for a few nights in a hotel, and then maybe a few hundred for the registration. For tech conferences the registration fees go over $1000. This means that even the cheapest conferences will set a group back maybe $300, more realistically it’s around $1000.
  • Sometimes passive. What I mean to say is that there are so many talks and events, it’s easy to go and sit like a baked potato. I guess it’s not so bad to sit and enjoy yourself, but there is opportunity cost of the time. Tickets aren’t being answered, users helped, etc.

The last point isn’t really so important - everyone could deserve to take some time. The thing that is troubling is that there is a stark contrast between extra funds that a traditional research computing group has and the costs, period. Attending a conference is a privilege, because it means someone else can do the work for you, and (typically) your group can pay your way. The question I’m interested in is Who do we not see at conferences? Who doesn’t get the benefit of learning and networking? The answer are the groups that maybe need it most.

I don’t think it’s fair to be critical without making suggestions for change, so now I’ll share those thoughts. I also want to point out my bias - I’m a remote worker for over 2 years and I’ve run the gamut from attending conferences in person, to not being able to attend, to having quasi participation remotely. The change that I want to see is more initiative taken to make conferences ā€œRemote first.ā€ Here is how that would look:

  • An agenda is made, just as it was before, for remote participation. For multiple events happening at once, you just have multiple call ins. A call in is akin to a room.
  • For smaller group sessions, you have ā€œbreak outā€ sessions on the phone. Something like zoom can do this for you. There is both audio / video and chat. I would go as far to say that it’s a more comfortable environment (for some) to speak than raising a hand in a big conference hall.

Now here are the benefits:

  • Broad inclusion: It changes from an event of privilege to one of all inclusion. (Most) HPC and research computing folks have an internet connection and computer.
  • Temporal Freedom: In terms of time, it’s much easier to attend select sections without leaving the home base. The participants (if needed) can still provide support to their users without taking a week vacation. You save all the time that would incur from travel, taking cabs, etc.
  • Global Participation if it’s virtual, anybody can attend easily, even outside of the US, both as participants and speakers.
  • Comfortable Not everyone is comfortable in the environment that a conference affords. The talks might be ok, but events can be loud, crowded, and overwhelming. With virtual participation, people can be comfortable.

I think it’s probably easy to get access to a lot of conferences, and not think twice about the people that cannot go. And the negative effects on the environment. Others are noticing, and the HPC world should too. Here are some examples of other initiatives that have taken notice:

A lot of prominent scientists are also putting down their foot and saying ā€œI won’t be traveling, but I’ll give a talk remotelyā€ and (imho) that’s a way to lead to change:

There is another tweet I can’t find again where, instead of traveling, a scientist gave a remote talk, and when he finished, a bottle of expensive wine showed up at his door, which was much cheaper than having to fund his travel there.

The components that we lose are the big social ones - grabbing lunch with colleagues, or hanging out. But maybe we don’t have to - why not have sessions over zoom that are entirely social?
My point isn’t that conferences are bad, but the practice is dated, hurts the environment, and leaves a lot of folks out. I think it’s probably time to try and do better, even if just starting small (making them remote friendly), and asking how we can do better.

The hope for this topic was not necessarily to debate the pros and cons of conferences, but to bring forward the options as they exist today. Certainly there is a place for discussing the merits of bringing people together in person, and I respect those opinions. But regardless of how and where they happen, it would be good to know what meetings are taking place currently in the research computing world. From local and regional to international.

For example, RMACC, which happens in May in Boulder, CO; PEARC, occurring now in Chicago; SC, scheduled for November in Denver; are some that come to mind. It would be especially helpful to hear of the smaller-scale opportunities.

1 Like

The president of MĆ©xico, López Obrador, who slashed funding for basic and applied research in MĆ©xico on May 03, calls traveling to conferences ā€˜political tourism’ according to Science, vol 365, pages 301 and 305.

I just attended IEEE HPEC conference, which is held in the Boston, Massachusetts area. It was less expensive compared to other conferences. While I’m a system administrator, it was a good opportunity to hear about the different types of problems being solved with Research Computing. I made some good contacts and learned about this website!

The focus of that conference seems to be secure hardware, FPGAs, GPUs, AI, etc. There was an education session that I was interested in since we are looking for ways to educate our system administrators and users alike on our systems.

1 Like

James Hetherington has started a list here: https://github.com/jamespjh/eResearch-meeting-list

In addition to the monthly CaRCC calls on several tracks of topics (https://carcc.org/people-network/) that were previously mentioned, there are also Campus Champions monthly calls: https://www.xsede.org/community-engagement/campus-champions

+1 what @cmaimone linked! it’s not advertised yet (I think he’s going to put it at a more official URL) but there is a humanly digestible version here: https://jamespjh.github.io/eResearch-meeting-list/