Study on network architecture of big data center for the efficient control of huge data traffic
Author(s) -
Hyung Woo Park,
Il Yeo,
Jongsuk Lee,
Haengjin Jang
Publication year - 2014
Publication title -
computer science and information systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.244
H-Index - 24
eISSN - 2406-1018
pISSN - 1820-0214
DOI - 10.2298/csis130916067p
Subject(s) - computer science , big data , architecture , floating car data , computer network , network architecture , data center , traffic generation model , data architecture , service (business) , distributed computing , traffic congestion , reference architecture , data mining , software architecture , operating system , transport engineering , art , engineering , visual arts , economy , software , economics
The network architecture of typical data centersis characterized bytiered networks and aggregation-based traffic controls. The emergence of big data makes it difficult for these data centers to incorporate big data service. The tier and aggregation based traffic management systems can magnify the seriousness of the traffic congestion and extend the congested region when big data moves around in the data center. As a consequence, big data has been forcing data centers to change their architecture dramatically. In this paper, we first address the important paradigm shifts of network architecture caused by big data traffic. We then show the new network architecture which resulted from our experience of the CERN LHC data service. Finally, we illustrate the effect of the throughput improvements of the proposed network architecture using a NS2 simulation.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom