Efficient updates for a shared nothing analytics platform

Doka, K; Tsoumakos, D; Koziris, N

dc.contributor.author	Doka, K	en
dc.contributor.author	Tsoumakos, D	en
dc.contributor.author	Koziris, N	en
dc.date.accessioned	2014-03-01T02:46:46Z
dc.date.available	2014-03-01T02:46:46Z
dc.date.issued	2010	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32839
dc.subject	data cube	en
dc.subject	data warehousing	en
dc.subject	distributed systems	en
dc.subject	time series	en
dc.subject	updates	en
dc.subject.other	Customizable	en
dc.subject.other	Data cube	en
dc.subject.other	Data warehousing	en
dc.subject.other	Distributed storage	en
dc.subject.other	Distributed systems	en
dc.subject.other	Ordered data	en
dc.subject.other	Prototype implementations	en
dc.subject.other	Query response	en
dc.subject.other	Shared nothing	en
dc.subject.other	Target application	en
dc.subject.other	Time-series data	en
dc.subject.other	Software prototyping	en
dc.subject.other	Time series	en
dc.subject.other	Warehouses	en
dc.subject.other	World Wide Web	en
dc.subject.other	Data warehouses	en
dc.title	Efficient updates for a shared nothing analytics platform	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1145/1779599.1779606	en
heal.identifier.secondary	http://dx.doi.org/10.1145/1779599.1779606	en
heal.identifier.secondary	1779606	en
heal.publicationDate	2010	en
heal.abstract	In this paper we describe a cloud-based data-warehouselike system especially targeted to time series data. Apart from the benefits that a distributed storage built on top of a shared-nothing architecture offers, our system is designed to efficiently cope with continuous, on-line updates of temporally ordered data without compromising the query throughput. Through a totally customizable process performing asynchronous aggregation of past records, we achieve significant gains in storage and update times compared to traditional methods, maintaining a high accuracy in query responses for our target application. Experiments using our prototype implementation over an actual testbed prove that our scheme considerably accelerates (by a factor above 3) the update procedure and reduces required storage by at least 30%. We also show how these gains are related to the level and rate of aggregation performed. © 2010 ACM.	en
heal.journalName	ACM International Conference Proceeding Series	en
dc.identifier.doi	10.1145/1779599.1779606	en