ChronoStream: Elastic stateful stream computation in the cloud

Author

Yingjun Wu ; Kian-Lee Tan

Author_Institution

Sch. of Comput., Nat. Univ. of Singapore, Singapore, Singapore

fYear

2015

fDate

13-17 April 2015

Firstpage

723

Lastpage

734

Abstract

We introduce ChronoStream, a distributed system specifically designed for elastic stateful stream computation in the cloud. ChronoStream treats internal state as a first-class citizen and aims at providing flexible elastic support in both vertical and horizontal dimensions to cope with workload fluctuation and dynamic resource reclamation. With a clear separation between application-level computation parallelism and OS-level execution concurrency, ChronoStream enables transparent dynamic scaling and failure recovery by eliminating any network I/O and state-synchronization overhead. Our evaluation on dozens of computing nodes shows that ChronoStream can scale linearly and achieve transparent elasticity and high availability without sacrificing system performance or affecting collocated tenants.

Keywords

cloud computing; concurrency (computers); operating systems (computers); parallel processing; resource allocation; system recovery; ChronoStream; OS-level execution concurrency; application-level computation parallelism; distributed system; dynamic resource reclamation; elastic stateful stream computation; failure recovery; first-class citizen; flexible elastic support; network I/O; state-synchronization overhead; transparent dynamic scaling; Checkpointing; Computational modeling; Containers; Elasticity; Instruction sets; Peer-to-peer computing; Runtime;

fLanguage

English

Publisher

ieee

Conference_Titel

Data Engineering (ICDE), 2015 IEEE 31st International Conference on

Conference_Location

Seoul

Type

conf

DOI

10.1109/ICDE.2015.7113328

Filename

7113328