Contribution Details |
| |
| Name: |
 |
BoF 17: High Performance Computing Programing Techniques For Big Data Hadoop |
|
| |
| Time: |
 |
Wednesday, June 20, 2012
12:15 PM - 1:00 PM |
|
| |
| Room: |
|
Hall C2.1
CCH - Congress Center Hamburg |
|
| |
| Speakers: |
|
Eyal Gutkind, Mellanox Technologies |
|
|
|
Gilad Shainer, Mellanox Technologies |
|
| |
| Abstract: |
|
Hadoop MapReduce and High Performance Computing share many characteristics such as, large data volumes, variety of data types, distributed system architecture, required linear performance growth with scalable deployment and high CPU utilization. RDMA capable programing models enable efficient data transfers between computation nodes. In this session we will discuss a collaborative work done among several industry and academic partners on porting Hadoop MapReduce framework to RDMA, the challenges, the techniques used, the benchmarking and testing. We will review the RDMA programing examples and their benefits for Hadoop MapReduce applications, how the usage of RDMA programing increased computation productivity by over 50%, while decreasing the number of nodes by 30% or more. |
|
|
|
|
|
