Transactional information systems: theory, algorithms, and the practice of concurrency control and recovery
H-store: a high-performance, distributed main memory transaction processing system
Proceedings of the VLDB Endowment
The mixed workload CH-benCHmark
Proceedings of the Fourth International Workshop on Testing Database Systems
HyPer: A hybrid OLTP&OLAP main memory database system based on virtual memory snapshots
ICDE '11 Proceedings of the 2011 IEEE 27th International Conference on Data Engineering
In-Memory Data Management: An Inflection Point for Enterprise Applications
In-Memory Data Management: An Inflection Point for Enterprise Applications
| Hi-index | 0.00 |
Ever increasing main memory sizes and the advent of multi-core parallel processing have fostered the development of in-core databases. Even the transactional data of large enterprises can be retained in-memory on a single server. Modern in-core databases like our HyPer system achieve best-of-breed OLTP throughput that is sufficient for the lion's share of applications. Remaining server resources are used for OLAP query processing on the latest transactional data, i.e., real-time business analytics. While OLTP performance of a single server is sufficient, an increasing demand for OLAP throughput can only be satisfied economically by a scale-out. In this work we present ScyPer, a Scale-out of our HyPer main memory database system that horizontally scales out on shared-nothing hardware. With ScyPer we aim at (i) sustaining the superior OLTP throughput of a single HyPer server, and (ii) providing elastic OLAP throughput by provisioning additional servers on-demand, e.g., in the Cloud.