In this paper we present BatchDB, an in-memory database engine designed for hybrid OLTP and OLAP workloads. BatchDB achieves good performance, provides a high level of data freshness, minimizes load interaction between the transactional analytical engines, thus enabling real time analysis over fresh under tight SLAs both

As a result of increases in both the query load and data managed, as well changes hardware architecture (multicore), last years have seen shift from query-at-a-time approaches towards shared work (SW) systems where queries are executed groups. Such groups share operators like scans joins, leading to that process hundreds thousands one go. SW range storage engines use in-memory co-operative more complex processing joins over analytical star schema queries. In all cases, they rely on either...

The release of hardware transactional memory (HTM) in commodity CPUs has major implications on the design and implementation main-memory databases, especially architecture high-performance lock-free indexing methods at core several these systems. This paper studies interplay HTM methods. First, we evaluate whether will obviate need for crafty index designs by integrating it a traditional B-tree architecture. performs well simple data sets with small fixed-length keys payloads, but its...

Database architectures typically process queries one-at-a-time, executing concurrent in independent execution contexts. Often, such a design leads to unpredictable performance and poor scalability. One approach circumvent the problem is take advantage of sharing opportunities across concurrently running queries. In this paper we propose Many-Query Join (MQJoin), novel method for join that can efficiently deal with hundreds This achieved by minimizing redundant work making efficient use...

The release of hardware transactional memory (HTM) in commodity CPUs has major implications on the design and implementation main-memory databases, especially architecture high-performance lock-free indexing methods at core several these systems. This paper studies interplay HTM methods. First, we evaluate whether will obviate need for crafty index designs by integrating it a traditional B-tree architecture. performs well simple data sets with small fixed-length keys payloads, but its...

This demonstration presents SharedDB, an implementation of a relational database system capable executing all SQL operators by sharing computation and resources across running queries. SharedDB sidesteps the traditional query-at-a-time approach executes queries in batches. Unlike proposed multi-query optimization ideas, do not have to contain common subexpressions order be part same batch, which allows for higher degree sharing. By as much possible, avoids repeating parts that is The goal...

Cloud-based data analysis is nowadays common practice because of the lower system management overhead as well pay-as-you-go pricing model. The model, however, not always suitable for query processing heavy use results in high costs. For example, query-as-a-service systems, where users are charged per processed byte, collections queries accessing same frequently can become expensive. problem compounded by limited options user to optimize execution when using declarative interfaces such SQL....