PosDB is a new disk-based distributed column-store relational engine aimed for research purposes. It uses the Volcano pull-based model and late materialization for query processing, and join indexes for internal data representation. In its current state PosDB is capable of both local and distributed processing of all SSB (Star Schema Benchmark) queries. Data, as well as query plans, can be distributed among network nodes in our system. Data distribution is performed by horizontal partitioning. In this paper we experimentally evaluate the performance of our system in a distributed environment. We analyze system performance and report a number of metrics, such as speedup and scaleup. For our evaluation we use the standard benchmark - the SSB.
|Journal||CEUR Workshop Proceedings|
|State||Published - 1 Sep 2017|
|Event||2nd Conference on Software Engineering and Information Management, SEIM 2017 - Saint Petersburg, Russian Federation|
Duration: 21 Apr 2017 → …
Scopus subject areas
- Computer Science(all)