I am a third year Ph.D. student in Computer Science working with the Systems and Networking Group at UC San Diego advised by Professor Amin Vahdat. I received my Bachelor of Science degree in Computer Science from Caltech in 2009. My research interests include distributed systems and parallel computing, cluster computing, and data center networks.
I will be interning at Google in the MapReduce team during the summer of 2011.
I am currently developing the world's fastest sorting system, TritonSort. TritonSort aims to achieve record speeds by focusing on per-disk and per-node efficiency. We measure ourselves in terms of MB/s/disk rather than raw sorting throughput. The goal of TritonSort is to sort data at the speed of the disks, by keeping all disks constantly reading or writing. TritonSort competed in the 2010 Sortbenchmark.org competition for GraySort and MinuteSort in the Indy category.
We are expanding TritonSort to handle general purpose computation. We have an implementation of MapReduce that shares a significant portion of TritonSort's code base. Although we do not make the same fault tolerance guarantees that traditional MapReduce systems provide, we believe that our implementation demonstrates that TritonSort's efficiency improvements go beyond sorting and are applicable to real workloads.
I held a Software Engineering Intern position at Google during the summer of 2010 working in the search infrastructure group with my mentor Alexander Yip.
During summer of 2009, I investigated balanced systems within the MapReduce framework of Hadoop. Goals consisted of analyzing cluster resources during a MapReduce job, identifying bottlenecks, and classifying various types of jobs according to these bottlenecks with the hope of being able to utilize cluster resources more efficiently.
TritonSort: A Balanced Large-Scale Sorting System, Alexander Rasmussen, George Porter, Michael Conley, Harsha V. Madhyastha, Radhika Niranjan Mysore, Alexander Pucher, and Amin Vahdat, Proceedings of the 8th ACM/USENIX Symposium on Networked Systems Design and Implementation (NSDI), Boston, MA, March 2011.
SortBenchmark 2011: TritonSort competed again in the Sort Benchmark competition in April 2011. This year we set 5 records (all wins, no ties): Indy 100TB GraySort, Daytona 100TB GraySort, Indy 60 second MinuteSort, Indy 100TB Joulesort, Daytona 100TB Joulesort. This year we had a general purpose MapReduce implementation with an initial sampling phase that enabled us to take the Daytona (general purpose) records. The SortBenchmark rules were updated this year to include a 100TB JouleSort category, which measures records/joule as the performance metric. As it turns out, focusing on cluster balance and resource utilzation appears to be a viable means for power efficient computing. TritonSort's JouleSort performance is less than an order of magnitude off of the most efficient system in any of the other JouleSort categories. Keep in mind that energy efficiency was not a primary goal of the TritonSort project. You can read our submission document here.
SortBenchmark 2010: TritonSort, the world's fastest cluster based disk-to-disk sorting system, which we submitted to the Sort Benchmark competition in May 2010. TritonSort placed first in the Indy MinuteSort 60 second category and tied for first in the Indy GraySort 100TB category. TritonSort focuses on per-node efficiency and aims to achieve sorting throughput as close as possible to the sequential speed of the disks. See the SortBenchmark submission document here.
In addition to developing computer systems, I enjoy driving cars. In December of 2009, I picked up a 2001 Chevrolet Corvette, and it's been a dream. Every morning I look forward to getting in my car and driving to work. At some point I am going to go to Racelegal and get myself timed. Racelegal is a group of street racing enthusiasts who do monthly sanctioned races in the parking lot of Qualcomm Stadium here in sunny San Diego. If you're in town and interested in cars I would highly recommend checking it out.
