Abstract
MapReduce has achieved tremendous success for large-scale data processing in data centers. A key feature distinguishing MapReduce from previous parallel models is that it interleaves parallel and sequential computation. Past schemes, and especially their theoretical bounds, on general parallel models are therefore, unlikely to be applied to MapReduce directly. There are many recent studies on MapReduce job and task scheduling. These studies assume that the servers are assigned in advance. In current data centers, multiple MapReduce jobs of different importance levels run together. In this paper, we investigate a schedule problem for MapReduce taking server assignment into consideration as well. We formulate a MapReduce server-job organizer problem (MSJO) and show that it is NP-complete. We develop a 3-approximation algorithm and a fast heuristic. We evaluate our algorithms through both simulations and experiments on Amazon EC2 with an implementation in Hadoop. The results confirm the advantage of our algorithms.
Original language | English |
---|---|
Title of host publication | IEEE INFOCOM 2014 - IEEE Conference on Computer Communications |
Publisher | IEEE |
Pages | 2175-2183 |
Number of pages | 9 |
ISBN (Print) | 9781479933600 |
DOIs | |
Publication status | Published - 1 Jan 2014 |
Event | 33rd IEEE Conference on Computer Communications, IEEE INFOCOM 2014 - Toronto, ON, Canada Duration: 27 Apr 2014 → 2 May 2014 |
Conference
Conference | 33rd IEEE Conference on Computer Communications, IEEE INFOCOM 2014 |
---|---|
Country/Territory | Canada |
City | Toronto, ON |
Period | 27/04/14 → 2/05/14 |
ASJC Scopus subject areas
- Computer Science(all)
- Electrical and Electronic Engineering