Search

This paper tells the story of Vectorwise, a high-performance analytical database system, from multiple perspectives: its history from academic project to commercial product, the evolution of its technical architecture, customer reactions to the product and its future research and development roadmap. One take-away from this story is that the novelty in Vectorwise is much more than just column-storage: it boasts many query processing innovations in its vectorized execution model, and an adaptive mixed row/column data storage model with indexing support tailored to analytical workloads. Another one is that there is a long road from research prototype to commercial product, though database research continues to achieve a strong innovative influence on product development.

Today's large-scale IR systems are not implemented using general-purpose database systems, as the latter tend to be significantly less efficient than custom-built IR engines. This paper demonstrates how recent developments in hardwareconscious database architecture may however satisfy IR needs. The advantage is flexibility of experimentation, as implementing a retrieval system on top of a DBMS boils down to relational query formulation, rather than system programming. We demonstrate in the context of the TeraByte TREC efficiency task that our experimental MonetDB/X100 database system provides highly competitive results both regarding precision and speed. We analyze the two innovations in MonetDB/X100 that most contributed to this successful application of DB technology in IR, namely vectorized incache processing and the use of two new light-weight compression schemes that work between the RAM and CPU cache memory levels.

This paper analyzes the performance of concurrent (index) scan operations in both record (NSM/PAX) and column (DSM) disk storage models and shows that existing scheduling policies do not fully exploit data-sharing opportunities and therefore result in poor disk bandwidth utilization. We propose the "Cooperative Scans" framework that enhances performance in such scenarios by improving data-sharing between concurrent scans. It performs dynamic scheduling of queries and their data requests, taking into account the current system situation. We first present results on top of an NSM/PAX storage layout, showing that it achieves significant performance improvements over traditional policies in terms of both the number of I/Os and overall execution time, as well as latency of individual queries. We provide benchmarks with varying system parameters, data sizes and query loads to confirm the improvement occurs in a wide range of scenarios. Then we extend our proposal to a more complicated DSM scenario, discussing numerous problems related to the two-dimensional nature of disk scheduling in column stores.

This paper describes our participation to the TREC HARD track (High Accuracy Retrieval of Documents) and the the TREC Enterprise track. The main goal of our HARD participation is the development and evaluation of so-called query profiles: Short summaries of the retrieved results that enable the user to perform more focused search, for instance by zooming in on a particular time period. The main goal of our Enterprise track participation is to investigate the potential of the structural information for this type of retrieval task. In particular, we study the use of the thread information and the subject and header fields of the email documents. As a secondary and long standing research goal, we aim at developing an information retrieval framework that supports many diverse retrieval applications by means of one simple yet powerful query language (similar to SQL or relational algebra) that hides the implementation details of retrieval approaches from the application developer, while still giving the application developer control over the ranking process. Both the HARD system and the Enterprise system (as well as our TRECVID video retrieval system) are based on MonetDB, an open source database system developed at CWI.

X100 is a new execution engine for the MonetDB system, that improves execution speed and overcomes its main memory limitation. It introduces the concept of in-cache vectorized processing that strikes a balance between the existing column-at-a-time MIL execution primitives of MonetDB and the tuple-at-a-time Volcano pipelining model, avoiding their drawbacks: intermediate result materialization and large interpretation overhead, respectively. We explain how the new query engine makes better use of cache memories as well as parallel computation resources of modern super-scalar CPUs. MonetDB/X100 can be one to two orders of magnitude faster than commercial DBMSs and close to hand-coded C programs for computationally intensive queries on in-memory datasets. To address larger disk-based datasets with the same efficiency, a new ColumnBM storage layer is developed that boosts bandwidth using ultra lightweight compression and cooperative scans.

Database systems tend to achieve only low IPC (instructions-per-cycle) efficiency on modern CPUs in compute-intensive application areas like decision support, OLAP and multimedia retrieval. This paper starts with an in-depth investigation to the reason why this happens, focusing on the TPC-H benchmark. Our analysis of various relational systems and MonetDB leads us to a new set of guidelines for designing a query processor. The second part of the paper describes the architecture of our new X100 query engine for the MonetDB system, that follows these guidelines. On the surface, it resembles a classical Volcano-style engine, but the crucial difference to base all execution on the concept of vector processing makes it highly CPU efficient. We evaluate the power of MonetDB/X100 on the 100GB version of TPC-H, showing its raw execution power to be between one and two orders of magnitude higher than previous technology.

Recent research shows that traditional database technology does not efficiently exploit the features of modern computer hardware. This problem is especially visible in the area of computationally-intensive applications, where specialized programs achieve performance orders of magnitude higher than DBMSs. In this article we present techniques that aim at bridging this gap. Three issues are discussed: efficient query execution on superscalar CPUs, optimizing disk bandwidth on commodity hardware, and exploiting the features of emerging multi-threaded and multi-core CPUs.

High disk bandwidth in data-intensive applications is usually achieved with expensive hardware solutions consisting of a large number of disks. In this article we present our current work on software methods for improving disk bandwidth in ColumnBM, a new storage system for MonetDB/X100 query execution engine. Two novel techniques are discussed: superscalar compression for standalone queries and cooperative scans for multi-query optimization.