The End of an Architectural Era (It's Time for a Complete Rewrite)

By M. Stonebraker, Samuel Madden, Daniel J. Abadi, Stavros Harizopoulos, N. Hachem, Pat Helland. Very Large Data Bases Conference, 2007

777 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/136eefe33796c388a15d25ca03cb8d5077d14f37

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

Cache craftiness for fast multicore key-value storage (2012, background): “H-Store [25, 35] and VoltDB, its commercial version, are in-memory relational databases designed to be orders of magnitude faster than previous systems.”
When things matter: A survey on data-centric internet of things (2016, background): “Furthermore, superior performance of column store based systems has been shown over the major RDBMS (relational DBMS) system (Stonebraker et al., 2007).”
In-Memory Big Data Management and Processing: A Survey (2015, background): “H-Store [36], [229] or its commercial version VoltDB [150] is a distributed row-based in-memory relational database targeted for high-performance OLTP processing.”
HyPer: A hybrid OLTP&OLAP main memory database system based on virtual memory snapshots (2011, background): “Based on an early study for banking transactions [16] the authors of H-Store [17], [6], [5] deserve the credit for analyzing the overhead imposed by various traditional database management features (buffer management, logging, locking, etc.”
Design and Implementation of a General Purpose Parallel Programming System (1996, unknown): “There is by now a widespread belief that SQL is too limited for many applications [15, 26, 31, 34, 35].”
Skew-aware automatic database partitioning in shared-nothing, parallel OLTP systems (2012, methodology): “These are usually the “look-up” tables in OLTP applications [43], and thus we want to avoid exploring neighborhoods where they are not replicated.”
Data Management in the Cloud: Limitations and Opportunities (2009, background): “Furthermore the main beneﬁt of a shared-nothing architecture is its scalability [24]; however this advantage is less relevant for transactional data processing for which the overwhelming majority of deployments are less than 1 TB in size [33].”
Spanner: Google's Globally-Distributed Database (2012, unknown): “HStore [39] and Granola [11] each supported their own classification of transaction types, some of which could avoid locking.”
H-store: a high-performance, distributed main memory transaction processing system (2008, methodology): “We begin in Section 2 by ﬁrst providing an overview of the internals of the H-Store system.”
Untangling Blockchain: A Data Processing View of Blockchain Systems (2017, background): “Although protocols are costly in both time and space[15].”
Speedy transactions in multicore in-memory databases (2013, background): “LLAMA [22] adds support for transactional logging, but its logger is centralized.”
Staring into the Abyss: An Evaluation of Concurrency Control with One Thousand Cores (2014, methodology): “Finally, we use the same timestamp allocation optimizations from above to avoid the mandatory wait time to account for clock skew [38].”
Schism (2010, methodology): “The partitioning produced by the tool distributes tuples based on warehouse id, and replicates the item table—the same strategy found by human experts [21].”