19/04/2017 11:30Taub 301


Omid: Low latency, Scalable and Highly-Available Transactions in Distributed Data Storage

Ohad Shacham

Yahoo Research

We present Omid - a low latency transaction processing service that powers web-scale production systems at Yahoo. Omid provides ACID transaction semantics on top of traditional key-value storage; its implementation over Apache HBase is open sourced as part of Apache Incubator. Omid can serve hundreds of thousands of transactions per second on standard mid-range hardware, while incurring minimal impact on the speed of data access in the underlying key-value store. Additionally, as expected from always-on production services, Omid is highly available. Omid is currently being integrated as Apache Phoenix transaction engine. This talk combines several incarnations of Omid. The people that contributed to all/part of the presented material are: Aran Bergman, Edward Bortnikov, Yoni Gottesman, Eshcar Hillel, Igor Katkov, Idit Keidar, Ivan Kelly, Matthieu Morel, Sameer Paranjpye and Francisco Perez-Sorrosal.

Bio: Ohad Shacham is Senior Research Scientist at Yahoo. He works on scalable big data and search platforms. Most recently, he focused on extending the Omid transaction processing system with high availability and scalability features. Ohad received his PhD in concurrent software verification from the Computer Science department at Tel-Aviv University, under the supervision of Mooly Sagiv and Eran Yahav. Prior to Yahoo, Ohad was a Research Staff Member at IBM Research and a Software Engineer at Intel.