1 / 13

Enhancing Distributed Scientific Computing through Datacenter Networking and 100GigE Wide-Area Networking

This integrated study focuses on improving end-to-end application-level performance in scientific computing with new technologies. The project addresses low utilization of high-speed connections and aims to enhance data analysis, networking solutions evaluation, and technology development. Current work involves data analysis, networking benchmarking, and conducting preliminary tests on various resources. Future plans include integrating infrastructure with GridFTP, implementing per-process monitoring solutions, and publishing results.

rclausen
Download Presentation

Enhancing Distributed Scientific Computing through Datacenter Networking and 100GigE Wide-Area Networking

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. An integrated study of datacenter networking and 100GigE wide-area networking in support of distributed scientific computing • Zhengyang Liu • University of Virginia • zl4ef@virginia.edu • Sept. 30, 2011 This work was carried out as part of a sponsored research project from the NSF OCI-1127340

  2. Problem Statement • Poor end-to-end application-level performance • 200MB/s over 10Gb/s connection -> less than 20% utilization

  3. Problem Statement • Data Analysis • Networking Solutions Evaluation • New Technology Development

  4. Data Analysis • Applications • Benchmark • Profiling • Network • Active Measurements • Passive Measurements

  5. Current Work • Learn! • Revisit Networking: A Statistical Approach • Datacenter Networking: InfiniBand, RoCE, iWARP • Iperf • Netflow

  6. Current Work • Preliminary Tests on Our Own Resources • PC / VM / EC2 • Getting familiar with tools: GridFTP, Ganglia, etc. • Preliminary Tests on ANI Testbed

  7. Demo

  8. Future Work • Immediate ToDo • Continue Learning • GridFTP tests on ANI • Per-process Monitoring Solution (DTrace?)

  9. A (nascent) Analyzing Framework Iperf What could happen? Netflow What happened inside the network? Ganglia What happened inside the host? DTrace What happened inside the application?

  10. Future Work • Within 1st Year • Further Analysis • Integrate IDC with GridFTP • Publish Results

  11. Future Work • Finish the Project!

More Related