Distributed Indexing and Locking : In Search of Scalable Consistency
Mahdi Tayarani Najaran, Primal Wijesekera, Andrew Warfield, Norman C. Hutchinson · 2011
Indexing and locking are two important components of a Relational Database Management System (RDBMS), which pose as potential bottlenecks when scaling. We present novel distributed alternatives to both of these components which are based on a spatial partitioning tree designed for scalable low-latency information storage and retrieval. The proposed solutions provide strong consistency, fault-tolerance and low latency. Our initial evaluations using the TPC-C workload show they are capable of utilizing more processing power for scalability. We believe our contributions to be an important first step toward our goal of a scalable, cloud aware and full-featured RDBMS architecture.