Asynchronous view maintenance for VLSD databases

Agrawal, Parag; Silberstein, Adam; Cooper, Brian F.; Srivastava, Uma; Ramakrishnan, Raghu

doi:10.1145/1559845.1559866

Cited by 57 publications

(21 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Secondary indexes on key-value stores: Recently, a large body of academic work [10], [24], [11] and industrial projects [25], [26], [27], [28], [29], [30], [31] emerge to build secondary indexes as middleware on scalable key-value store systems. Those systems can be largely categorized by their design choices in terms of: 1) whether the index is local or global, 2) how the index is maintained, and 3) the system implementation.…”

Section: Related Workmentioning

confidence: 99%

Deferred Lightweight Indexing for Log-Structured Key-Value Stores

Tang

Iyengar²,

Tan³

et al. 2015

2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

Deferred Lightweight Indexing for Log-Structured Key-Value Stores

Tang

Iyengar²,

Tan³

et al. 2015

2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

View full text Add to dashboard Cite

show abstract

“…ITHBase [6], IHBase [9], CCIndex [11] and Asynchronous views [1] are four representative related work that use several one-dimensional indexes to accelerate multiple-fields query processing. For each column that is frequently used by user queries, a one-dimensional index was build on it.…”

Section: Query Optimization Through Indexesmentioning

confidence: 99%

Fast Multi-fields Query Processing in Bigtable Based Cloud Systems

Wang

Meng

2013

Web-Age Information Management

View full text Add to dashboard Cite

Abstract.With the rapid increase of data sizes, enterprise applications are migrating their backend data management and analytic systems into cloud based data management systems.Bigtable is among one of the major data models used by cloud storage systems as their storage layer. Such systems provide high scalability and schema flexibility, and support efficient point and range based queries based on rowkeys. However, Bigtable based systems have limited support on non-rowkey based queries and multiple-fields based queries, due to much overhead on invoking extra scanning of data. In this paper, we develop a system TNBGR(Telecom Network Browsing Gateway Records) on managing and querying large scale telecommunication data. TNBGR is built on top of HBase and MapReduce, with a focus on optimizing multi-fields query processing. TNBGR provides a novel application and system resource aware data allocation strategy to minimize data access through multi-layer region partitioning, resource parameterization, and balanced region distribution.The query composition dynamically updates application parameters based on tracked system statistics and automatically translates queries for MapReduce. Through additional query optimization by improving region locality, TNBGR achieves high efficiency on supporting multi-field queries. The experimental results show that our solution improves the performance of the queries by about 5 and 18 times respectively.

show abstract

“…As described in Section 2.3, there has been significant prior work on leveraging precomputation through automatic materialized view selection [3,51,60] and incremental maintenance [2,16,21,40,72]. These approaches, however, have focused on minimizing average cost for a given workload, rather than ensuring consistent resource requirements as the database grows.…”

Section: Scale-independent View Selectionmentioning

confidence: 99%

“…attribute 1 $ attribute 2 Add a bidirectional dependency of weight one for each attribute pair present in an equality predicate in the WHERE clause of the view definition.…”

Section: (Keyattributes)mentioning

confidence: 99%

Scale-Independent Relational Query Processing

Armbrust¹

2013

View full text Add to dashboard Cite

An increasingly common pattern is for newly-released web applications to succumb to a "Success Disaster". In this scenario, overloaded database machines and resultant high response times destroy a previously good user experience, just as a site is becoming popular. Unfortunately, the data independence provided by a traditional relational database system, while useful for agile development, only exacerbates the problem by hiding potentially expensive queries under simple declarative expressions. The disconnect between expression complexity and runtime cost often leads developers to mistrust the suitability of relational database systems for their web applications in the long term. As a result, developers of these applications are increasingly abandoning relational systems in favor of imperative code written against distributed "NoSQL" key/value stores, losing the many benefits of data independence in the process.While some claim that scalability issues are inherent in the use of the relational model, this thesis challenges that notion by extending standard data independence with the notion of scale independence. In contrast to traditional relational databases, a scale-independent system is capable of providing predictable response time for all of the queries in an application, even as the amount of data grows by orders of magnitude. This predictability is achieved by compile-time enforcement of strict upper bounds on the number of operations that will be performed for all queries. Coupled with a service level objective (SLO) compliance prediction model and a scalable storage architecture, these upper bounds make it easy for developers to write success-tolerant applications that support an arbitrarily large number of users while still providing acceptable performance.Statically bounding the amount of work required to execute a query can be easy for some queries, such as those that perform a lookup of a single record by primary key. However, such simple queries are generally insu cient for the construction of complex, real-world applications. Therefore, to enable successful development of such applications, this thesis defines four levels of scale-independent execution that greatly expand the space of queries that can be guaranteed to be scalable a priori. Each scale-independent execution level leverages increasingly sophisticated techniques, ranging from extra cardinality information in 2 the schema to incremental precomputation, to ensure that the performance of the application will not change as the amount of stored data grows. Furthermore, developers can use the levels to reason about the resource requirements of each query that is run by their application.In addition to presenting the theory of scale independence, this thesis describes PIQL, a actual implementation of a scalable relational engine. Using the PIQL system, I present an empirical analysis of scale independence that includes all the queries from the TPC-W benchmark and validates PIQL's ability to maintain nearly constant high-quantile query and upd...

show abstract

Asynchronous view maintenance for VLSD databases

Cited by 57 publications

References 19 publications

Deferred Lightweight Indexing for Log-Structured Key-Value Stores

Deferred Lightweight Indexing for Log-Structured Key-Value Stores

Fast Multi-fields Query Processing in Bigtable Based Cloud Systems

Scale-Independent Relational Query Processing

Contact Info

Product

Resources

About