Streams - Outline Introduction Distributed DBMS...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Distributed & Parallel DBMS M. Tamer Özsu Page 11.1 Outline ± Introduction ± Distributed DBMS Architecture ± Distributed Database Design ± Distributed Query Processing ± Distributed Transaction Management ± Data Replication ± Parallel Database Systems ± Data Integration Systems ± Web Search/Querying ± Peer-to-Peer Data Management ² Data Stream Management ² Models, languages ² Query processing & optimization
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Distributed & Parallel DBMS M. Tamer Özsu Page 11.2 Inputs & Outputs ± Inputs: One or more sources generate data continuously, in real time, and in fixed order z Sensor networks – weather monitoring, road traffic monitoring, motion detection z Web data – financial trading, news/sports tickers z Scientific data – experiments in particle physics z Transaction logs – telecom, point-of-sale purchases z Network traffic analysis (IP packet headers) – bandwidth usage, routing decisions, security ± Outputs: Want to collect and process the data on- line z Environment monitoring z Location monitoring z Correlations across stock prices z Denial-of-service attack detection ± Up-to-date answers generated continuously or periodically
Background image of page 2
Distributed & Parallel DBMS M. Tamer Özsu Page 11.3 Processing Model ± Stream-in-stream-out ± Problem: z Streams have unbounded length ( system point of view ) z New data are more accurate/interesting ( user point of view ) ± Solution: z Windows Data Stream System
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Distributed & Parallel DBMS M. Tamer Özsu Page 11.4 Streams and Windows ± Sliding windows z Keep the last C items ( count-based windows ), or items which have arrived in the last T time units ( time-based windows ) z Old items expire as time goes on z Different from non-overlapping windows
Background image of page 4
Distributed & Parallel DBMS M. Tamer Özsu Page 11.5 Traditional Database Management System (DBMS) Persistent data - stored until deleted by user or application Transient queries - issued once, then forgotten
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Distributed & Parallel DBMS M. Tamer Özsu Page 11.6 Data Stream Management System (DSMS) Persistent queries - generate up-to-date answers as time goes on Transient data - deleted as window slides forward
Background image of page 6
Distributed & Parallel DBMS M. Tamer Özsu Page 11.7 DSMSs – Novel Problems ± Push-based (data-driven), rather than pull- based (query-driven) computation model z New data arrive continuously and must be processed z Query plans require buffers, queues, and scheduling mechanisms z Query operators must be non-blocking z Must adapt to changing system conditions throughout the lifetime of a query z Load shedding may be required if the system can’t keep up with the stream arrival rates
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Distributed & Parallel DBMS M. Tamer Özsu Page 11.8 DSMS Implementation Choices ± Application on top of a relational DBMS z Application simulates data-driven processing z Inefficient due to the semantic gap between the DBMS and the DSMS-like application ± Use advanced features of the DBMS engine z Triggers, materialized views, temporal/sequence data models z Still based upon query-driven model, triggers don’t
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 12/23/2009 for the course DBST 663 taught by Professor Tba during the Spring '09 term at MD University College.

Page1 / 32

Streams - Outline Introduction Distributed DBMS...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online