Statistical resource allocation and providing QoS guarantee for inter-DC WAN데이터센터 간 광역 통신망에서의 통계적 자원 할당 및 서비스 품질 보장

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 344
  • Download : 0
There are many data centers(DCs) that are geographically distributed in the world, for global cloud computing service. These DCs are connected by wide area network, called Inter-DC WAN. These infrastructure costs too much and affect to the quality of service(QoS). But highly utilizing these and guaranteeing QoS simultaneously are highly challenging. In this paper, we propose a system for providing bandwidth and loss rate guarantees, and achieving highly network utilization at the same time, in the Inter-DC WAN. First, we suggest a new interface for tenants to express their network demand based on statistics and QoS, guaranteed loss rate. Second, we suggest a admission control module based on this statistical interface. This admission control module checks whether a request can be accepted or not. Finally, for satisfying the SLAs(Service Level Agreements) of each request, we suggest enforcement module to limit the flow statistically. We evaluate our system in a real testbed and large-scale emulation, by using Mininet. The results show that our system can 20 percent more user and achieving highly utilization, 30% more than previous system. Furthermore, the results show our system can guarantee the request QoS of all of tenants simultaneously.
Advisors
Yi, Yungresearcher이융researcher
Description
한국과학기술원 :전기및전자공학부,
Publisher
한국과학기술원
Issue Date
2017
Identifier
325007
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2017.8,[v, 34 p. :]

Keywords

Cloud Computing▼aInter data center wide area network(Inter-DC WAN)▼aVirtual network▼aResource allocation; Bandwidth gurantee▼aQoS guarantee▼aSoftware defined networking; 클라우드 컴퓨팅▼a데이터센터 간 광역 통신망▼a가상 네트워크▼a자원 할당▼a대역폭 보장▼a서비스 품질 보장▼a소프트웨어 정의 네트워킹

URI
http://hdl.handle.net/10203/243353
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=718685&flag=dissertation
Appears in Collection
EE-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0