MonetDB/X100: Hyper-Pipelining Query Execution

By Peter A. Boncz, Marcin Zukowski, Niels J. Nes. Conference on Innovative Data Systems Research, 2005

718 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/72c78448fd8630545a3582c8e24cd7b88e008ab6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Hexastore: sextuple indexing for semantic web data management (2008, methodology): “In [5], this approach is coupled with a column-oriented DBMS [14] [15] [42] (i.e., a DBMS designed especially for the vertically partitioned case, as opposed to a row-oriented DBMS, gaining benefits of compressibility [4] and performance [7]).”
Scalable Semantic Web Data Management Using Vertical Partitioning (2007, methodology): “One way to do this flattening, as suggested in [22] and [31], is to find sets of properties that tend to be defined together; i.e., clusters of subjects tend to have these properties defined.”
Cache locality is not enough: High-Performance Nearest Neighbor Search with Product Quantization Fast Scan (2015, unknown): “Besides, the use of SIMD to speed up query processing in databases [6] has been widely studied.”
Database Cracking (2007, result): “We also include results obtained with the MonetDB/X100 prototype [15, 16].”
The Snowflake Elastic Data Warehouse (2016, methodology): “This approach, pioneered by Vector-Wise (originally MonetDB/X100 [15]), saves I/O and greatly improves cache eﬃciency.”
Morsel-driven parallelism: a NUMA-aware query evaluation framework for the many-core age (2014, result): “We therefore concentrate on Vectorwise (version 2.5) in further experiments, as it was much faster than the other systems.”
Integrating compression and execution in column-oriented database systems (2006, methodology): “We then evaluate a set of compression schemes and show that the best scheme depends not only on the properties of the data but also on the nature of the query workload.”
DuckDB: an Embeddable Analytical Database (2019, methodology): “com/cwida/duckdb API C/C++/SQLite SQL Parser libpg_query [2] Optimizer Cost-Based [7, 9] Execution Engine Vectorized [1] Concurrency Control Serializable MVCC [10] Storage DataBlocks [5]”
Optimizing data-intensive computations in existing libraries with split annotations (2018, methodology): “The optimizations that SAs enable have been studied before: Vectorwise [68] and other vectorized databases [31, 35, 47] apply the same pipelining and parallelization techniques as SAs for improved cache locality.”
Efficiently Compiling Efficient Query Plans for Modern Hardware (2011, background): “This materialization has other advantages like allowing for vectorized operations [2], but in general the lack of pipelining is very unfortunate as it consumes more memory bandwidth.”
Super-Scalar RAM-CPU Cache Compression (2006, methodology): “We implemented PDICT, PFOR and PFOR-DELTA in the new ColumnBM storage manager of the Mon-etDB/X100 system [3], developed at the CWI for OLAP, data mining and multimedia/information retrieval applications.”