달력

92019  이전 다음

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  •  
  •  
  •  
  •  
  •  
LINK : http://portal.acm.org/citation.cfm?id=1555384
ACM
Proceedings of the eleventh international joint conference on Measurement and modeling of computer systems 2009, Thomas Sandholm, Kevin Lai Hewlett-Packard Laboratories, Palo Alto, CA, USA

Abstract

We present a system for allocating resources in shared data and compute clusters that improves MapReduce job scheduling in three ways. First, the system uses regulated and user-assigned priorities to offer different service levels to jobs and users over time. Second, the system dynamically adjusts resource allocations to fit the requirements of different job stages. Finally, the system automatically detects and eliminates bottlenecks within a job. We show experimentally using real applications that users can optimize not only job execution time but also the cost-benefit ratio or prioritization efficiency of a job using these three strategies. Our approach relies on a proportional share mechanism that continuously allocates virtual machine resources. Our experimental results show a 11-31% improvement in completion time and 4-187% improvement in prioritization efficiency for different classes of MapReduce jobs. We further show that delay intolerant users gain even more from our system.

[측정 방법 설명]
To measure this effect, we introduce a total system efficiency metric that is based on th average ratio of actual application performance in a shared system to the application performance in a dedicated system.

[시나리오 설명]
2. Usage scenario
This section describes the usage scenario for the system described in this paper.

[질문을 제시하고 설명]
(1)How much do I want to spend?
.
(2) How do I want to spend?
.
(3)Should I spend more or less?

[수식 설명]
A rprivider allocates resource share qi to user i at time t as follows:[수식]

Posted by Teshi