There is a sign-up schedule for this event that can be found at

http://apps.cs.utexas.edu/talkschedules/cgi/list_events.cgi

Type o

f Talk: Faculty Candidate

Speaker/Affiliation: Haryadi S. Gunawi/Unive

rsity of California, Berkeley

Talk Audience: UTCS Faculty, Graduate

Students, Undergraduate Students, and Outside Interested Parties

Dat

e/Time: Tuesday, March 27, 2012, 11:00 am

Location: ACES 2.302

nHost: Mike Dahlin

Talk Title: Towards Reliable Storage Systems: From

OS-Level File Systems to Cloud Storage

Abstract:
Three trends will d

ominate the storage systems of tomorrow: increasingly massive amounts of da

ta, the incredible growth of software complexity, and the increasing use

of cheap and less reliable hardware. These trends present us with a huge ch

allenge: How can we promise users that storage systems work robustly in spi

te of their massive software complexity and the broad range of hardware fai

lures that can arise? Addressing this question is not straightforward as cu

rrent approaches scatter recovery code in thousands of lines of intricate,
low-level C code. As a result, reliability problems are often found in cu

rrent storage systems.

In this talk, I will present how we build a ne

w generation of more robust and reliable storage systems via simpler design

s and powerful testing frameworks. Specifically, I will first describe I/O
Shepherding and SQCK, new online and offline reliability frameworks for O

S-level file systems, with which we advocate a higher-level strategy where
the logic of reliability policies can be described clearly and concisely.

I will then present my recent work in improving cloud storage reliability w

ith FATE and DESTINI, a powerful failure testing service and a framework f

or declarative recovery specifications. Finally, I will close this talk wi

th my future plans in the area of cloud storage and new storage technology.

Bio:
Haryadi Gunawi is currently a postdoctoral fellow at the Un

iversity of California, Berkeley. He received his Ph.D. from the Universit

y of Wisconsin, Madison in 2009. His current research focuses on operating
systems and large-scale distributed storage systems. Beyond that, his res

earch experience also spans cross-disciplinary areas such as software engin

eering, databases, and networking. He has won numerous awards including a

n Honorable Mention for the 2009 ACM Doctoral Dissertation Award, a co-win

ner of the 2009 departmental best thesis award, and the 2010 NSF Computing
Innovation Fellowship. He was also a lead author on an awarded NSF proposa

l under the CISE Data-intensive Computing program.