The wide distribution of XML documents and the standardization of the Query languages XPath and XQuery have led to a wide variation of XML database implementations. Yet the efficient processing of really large XML documents is still supported by just a few products such as e.g. MonetDB/XQuery as open-source solution [1] or X-Hive as commercial product [2]. Following the main memory and relation...

Multi-version database systems preserve the explicit history of values taken by each database record. They do so by maintaining versions; if the value of a record needs to be updated, the system creates a new version, while keeping the previous value untouched. In contrast, a single-version database system updates records in place. As a consequence, in his seminal paper describing the “transact...

In this paper, we propose a concurrency control protocol, called the Prudent-Precedence Concurrency Control (PPCC) protocol, for high data contention main memory databases. PPCC is prudently more aggressive in permitting more serializable schedules than two-phase locking. It maintains a restricted precedence among conflicting transactions and commits the transactions according to the serializat...

Linear Hashing has been proposed as a simple and eecient technique for storing and retrieving key sets whose cardinality is not known in advance. Its main advantage is that it allows tables to grow and shrink gracefully without major reorganizations. In this paper, several modiications of the basic scheme are presented, together with performance measurements. It is also shown that the seemingly...

HYRISE is a new relational storage engine for main memory database systems. It is built on the premise that enterprise application workloads can benefit from a dedicated main-memory storage engine. The key idea behind HYRISE is that it provides dynamic vertical partitioning of the tables it stores. Since enterprise applications typically use a large number of very wide tables, we designed a nov...

The requirements for a main memory data storage model are both compactness and efficient processing for all database operations. The DBGraph storage model, proposed in this paper, achieves these goals. By representing the entire database in a unique graph-based data structure, called DBGraph, it fully exploits the directaccess capability of main memory systems. For example, Selection, Join and ...

Computer architectures are increasingly based on multi-core CPUs and large memories. Memory bandwidth, which has not kept pace with the increasing number of cores, has become the primary processing bottleneck, replacing disk I/O as the limiting factor. To address this challenge, we provide novel algorithms for increasing the throughput of Business Intelligence (BI) queries, as well as for ensur...

A possible approach to query Linked Data combines the actual evaluation of a query with the traversal of data links in order to discover and retrieve potentially relevant data. An implementation of this idea requires approaches that support an efficient and flexible management of temporary, ad hoc data collections that emerge during query execution. However, existing proposals for managing RDF ...

We present a design for multi-version concurrency control and recovery in a main memory database, and describe logical and physical versioning schemes that allow read-only transactions to execute without obtaining data item locks or system latches. These schemes enable a system to guarantee that updaters will never interfere with read-only transactions, and that read-only transactions will not ...

We perform the first study of the tradeoff space of access methods and replication to support statistical analytics using first-order methods executed in the main memory of a Non-Uniform Memory Access (NUMA) machine. Statistical analytics systems differ from conventional SQL-analytics in the amount and types of memory incoherence that they can tolerate. Our goal is to understand tradeoffs in ac...