The advent of non-volatile memory (NVM) will fundamentally change the dichotomy between memory and durable storage in database management systems (DBMSs). These new NVM devices are almost as fast as DRAM, but all writes to it are potentially persistent even after power loss. Existing DBMSs are unable to take full advantage of this technology because their internal architectures are predicated on the assumption that memory is volatile. With NVM, many of the components of legacy DBMSs are unnecessary and will degrade the performance of data intensive applications.
In this talk, I will discuss our work in developing new high- performance DBMSs that are able to take advantage of NVM's unique properties. I will begin by discussing the development of different storage and recovery methods for OLTP workloads. Our experimental evaluation on a NVM hardware emulator show that these NVM-optimized methods achieve up to 5.5X higher throughput than their traditional counterparts while reducing the amount of wear due to write operations by up to 2X. These methods are also able to restore the database to a correct state nearly instantaneously after a crash. I will then discuss the early development of N-Store, a new hybrid DBMS designed specifically for NVM.
Joy Arulraj is a second year Ph.D. student in the Computer Science department at CMU. He is advised by Prof. Andy Pavlo. His general research interest is in database systems. Recently, he has been looking at exploiting non-volatile memory technologies to improve the performance of database systems. Before coming to CMU, Joy received is M.S. in Computer Science from the University of Wisconsin, Madison and is B.E in Computer Science from the College of Engineering, Guindy (Anna University).