DocumentCode
3124255
Title
CoTS: A Scalable Framework for Parallelizing Frequency Counting over Data Streams
Author
Das, Sudipto ; Antony, Shyam ; Agrawal, Divyakant ; El Abbadi, Amr
Author_Institution
Dept. of Comput. Sci., Univ. of California, Santa Barbara, Santa Barbara, CA
fYear
2009
fDate
March 29 2009-April 2 2009
Firstpage
1323
Lastpage
1326
Abstract
Frequency counting, frequent elements and top-k queries form a class of operators that are used for a wide range of stream analysis applications. In spite of the abundance of these algorithms, all known techniques for answering data stream queries are sequential in nature. The imminent ubiquity of chip multi-processor (CMP) architectures requires algorithms that can exploit the parallelism of such architectures. In this paper, we first evaluate different naive techniques for intra-operator parallelism, and summarize the insights obtained from the naive techniques. Our experimental analysis of the naive designs shows that intra-operator parallelism is not straightforward and requires a complete redesign of the system. We then propose an efficient and scalable framework for parallelizing frequency counting, frequent elements and top-k queries over data streams. The proposed CoTS (co-operative thread scheduling) framework is based on the principle of threads co-operating rather than contending. Our experiments on a state-of-the-art quad-core chip-multiprocessor architecture and synthetic data sets demonstrate the scalability of the proposed framework, and the efficiency is demonstrated by peak processing throughput of more than 60 million elements per second.
Keywords
parallel architectures; processor scheduling; CMP; CoTS; chip multi-processor; cooperative thread scheduling; data streams; intra-operator parallelism; parallelizing frequency counting; peak processing throughput; stream analysis applications; synthetic data sets; top-k queries; Computer science; Counting circuits; Data engineering; Frequency; Monitoring; Parallel processing; Scalability; Throughput; USA Councils; Yarn;
fLanguage
English
Publisher
ieee
Conference_Titel
Data Engineering, 2009. ICDE '09. IEEE 25th International Conference on
Conference_Location
Shanghai
ISSN
1084-4627
Print_ISBN
978-1-4244-3422-0
Electronic_ISBN
1084-4627
Type
conf
DOI
10.1109/ICDE.2009.231
Filename
4812531
Link To Document