SC22 Proceedings

The International Conference for High Performance Computing, Networking, Storage, and Analysis

Birds of a Feather Archive

Research Storage Strategies Revisited: Cloud, Capacity, Cost ... and Other Things Keeping You Up at Night

Authors: Vas Vasiliadis (University of Chicago, Globus), Rachana Ananthakrishnan (University of Chicago, Globus)

Abstract: At SC21, over 50 attendees participated in robust discussion of strategies for managing storage in advanced computing environments. At the urging of many, we plan to continue the conversation this year, focusing on two themes: (1) progress made on creating unified storage environments for research computing and (2) strategies and tactics for dealing with the end of “unlimited free" cloud storage. In particular, policy and pricing model changes in cloud storage offerings have placed substantial pressure on RCD organizations to migrate to alternative storage solutions—a task that can be daunting, given the scale and diversity of data involved.

Long Description: This session will bring together campus IT leaders, research computing professionals and storage providers, to engage in a dialogue on the challenges they face and the strategies they are adopting to address them. Our goals are to (a) hear from some of last year’s panelists about progress made on their research storage strategies, and (b) facilitate a discussion about the challenges institutions are facing as they deal with policy changes at major cloud storage providers. On the second point, we anticipate robust discussion and look forward to the audience sharing their experiences as they handle petascale migration of data from systems such as Google Drive.

As with last year’s BoF, a diverse group of institutions who recognize the relevance of this topic have already expressed interest in sharing their insights. We are happy to welcome back three of our previous panelists:

- Erik Deumens, Research Computing Senior Director at the University of Florida - Curtis Hillegas, Associate CIO, Research Computing at Princeton University - Preston Smith, Executive Director of Research Computing at Purdue University - Alexander Tzanov, Director, CUNY HPC Center at the City University of New York

And we expect participation from the following research computing and data experts:

- Stanford University (Invited speaker: Ruth Marinshaw, CTO - Research Computing) - University of Oklahoma (Invited speaker: Henry Neeman, Director of the OU Supercomputing Center for Education & Research) - University of Michigan (Invited speaker: Brock Palen, Director Advanced Research Computing) - Harvard University (invited speaker: Scott Yockel, University Research Computing Officer)

The panelists represent a diverse range of institutional research environments, which we believe will make the session appealing to an even broader audience. We have also invited representatives from cloud storage providers to participate in the discussion and expect this will further enrich the dialogue.

We plan to explore approaches for dealing with cloud storage, data generated by instruments, as well as data security, privacy and the everchanging world of compliance. The discussion will be guided by the following questions:

- What factors have the greatest influence on how you plan, deploy and manage research storage? - How is your institution handling the data deluge from instruments such as cryo-EM? - How is cloud storage being used by researchers? - Did you (or do you plan to) migrate between different classes of cloud storage? - What hurdles do you anticipate as you evolve your storage environment?

We expect that attendees will broaden their understanding of the issues described above and learn about approaches they may consider when faced with similar challenges at their own institutions.


Back to Birds of a Feather Archive Listing