 |
|
Ramakrishna Kotla
|
Research areas:
Distributed systems, fault tolerance, operating systems, storage systems
|
|
|
|
I am a Researcher at Microsoft Research (Silicon Valley). Before moving to MSR, I co-founded Zyzzyva and SafeStore projects with Prof. Mike Dahlin (advisor) and Prof. Lorenzo Alvisi at the University of Texas at Austin, where I received MSE and PhD degrees in Computer Engineering. Prior to that I received B.Tech degree in ECE from IIT Kharagpur in 1998, and worked as an R&D Engineer at Synopsys Inc., between 1998-2001. I received two best paper awards(SOSP and USENIX) and one award paper(IWLS).
My research interests span a broad range of topics related to systems, including distributed systems, operating systems, fault tolerance, file and storage systems, and packet classification.
-
- "xBFT: Byzantine Fault Tolerance with High Performance, Low Cost, and Aggressive Fault Isolation," Ramakrishna Kotla. Dissertation(pdf)
This dissertation solves three key problems of practical Byzantine Fault Tolerance: Performance, Cost, and Isolation. First, Zyzzyva improves performance and reduces cost of BFT replication technique using speculation. Second, SafeStore provides long term data durability by isolating faults aggressively along multiple dimensions. Third, CBASE improves performance of BFT replicated application by exploiting application-level parallelism. Following papers summarize some of these results but more on this can be found here .
- "Zyzzyva: Speculative Byzantine Fault Tolerance" Ramakrishna Kotla, Lorenzo Alvisi, Mike Dahlin, Allen Clement, and Edmund Wong . In Proceedings of 21st ACM Symposium on Operating Systems Principles (SOSP 2007), WA, 2007. Best Paper Award
Paper(pdf) Talk(pdf) Extended report (pdf) Source code
To appear in CACM Magazine as an invited article under Research highlights section (currently under review). New
- "SafeStore: A Durable and Practical Storage System" Ramakrishna Kotla, Lorenzo Alvisi, and Mike Dahlin. In Proceedings of Usenix Annual Technical Conference (USENIX 2007), CA, 2007. Best Paper Award
Paper(pdf) Extended Tech Report(pdf) Talk(pdf) Source code
- "High Throughput Byzantine Fault Tolerance," Ramakrishna Kotla and Mike Dahlin. In proceedings of the International Conference on Dependable Systems and Networks (DSN 2004), DCC Symposium, Florence, Italy, June 2004.
Paper(pdf) Talk(ppt) Source code
-
- "Scheduling Processor Voltage and Frequency in Server and Cluster Systems" Ramakrishna Kotla, Soraya Ghiasi, Tom Keller, and Freeman Rawson. In proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) Paper(pdf)
-
-
"A reconfigurable architecture and associated synthesis methodology for high speed packet classification," Amit Prakash, Ramakrishna Kotla, Tanmoy Mandal, Adnan Aziz. In 11th IEEE/ACM International Workshop on Logic and Synthesis (IWLS 2002), pp 97-102, June 2002, New Orleans, USA . Paper(ps)
Award paper invited for publication in IEEE Transactions on CAD, June, 2003.