Software and Hardware Systems for Emerging Graph Algorithms

Speaker: Xuhao Chen , MIT CSAIL

Date: Wednesday, April 28, 2021

Time: 3:00 PM to 4:00 PM Note: all times are in the Eastern Time Zone

Public: Yes

Location: https://mit.zoom.us/meeting/register/tJUrdOqopj8uHdO4gUyVMnfglOFEqIye_Je0 (Registration required, only if you haven't registered for this series before; please read IMPORTANT NOTE below)

Event Type: Seminar

Room Description:

Host: Julian Shun, MIT CSAIL

Contact: Linda Lynch, lindalynch@csail.mit.edu

Relevant URL:

Speaker URL: https://chenxuhao.github.io/

Speaker Photo:
None

Reminders to: fast-code-seminar@lists.csail.mit.edu, seminars@csail.mit.edu, pl@csail.mit.edu, commit@lists.csail.mit.edu

Reminder Subject: TALK: Software and Hardware Systems for Emerging Graph Algorithms (Please read the IMPORTANT NOTE regarding registration)

******************IMPORTANT NOTE ABOUT REGISTRATION******************
- If you have already registered for any previous Fast Code Seminar on Zoom, please use the Zoom link that you have received in the past. This link will stay the same for all subsequent Fast Code seminars on Zoom. Please save it to your calendar!
- Zoom does not recognize a second registration, and will not send out the link a second time. The organizers will not be notified of any second registration.
- If you have any problems with registration, please contact lindalynch@csail.mit.edu by 2:30pm on the day of the seminar, so that we can try to resolve it before the seminar begins.
*********************************************************************

Abstract: Numerous real-world applications in social networks, e-commerce, biomedicine and security, are driven by graph algorithms. The massive irregular data processed in these algorithms poses challenges on computing system design, and requires rethinking the tradeoff between programmability and efficiency with a full system stack view.

In this talk, I will give an overview of my research that explores system design methodologies for efficient and productive processing of graph mining algorithms.

I will describe experiences creating new systems and abstractions for graph pattern mining (GPM) across different layers, including a software GPM programming system and a dedicated hardware accelerator for GPM. As I will demonstrate, cross-layer system design can bridge the parallel performance and programmability gap between the domain algorithms and the underlying hardware. I will then briefly discuss how these system design experiences can be leveraged for the emerging graph learning algorithms, and what key challenges remain.

Bio: Xuhao Chen is a postdoctoral researcher in the Computation Structures Group at MIT CSAIL, working with Prof. Arvind. Dr. Chen is broadly interested in computer systems, focusing on domain specific systems and architectures for graph algorithms. He has built various systems for data mining and machine learning on graphs. Before joining MIT, Dr. Chen was a research fellow in the Intelligent Software Systems group at University of Texas at Austin. His work has been published in ISCA, MICRO, VLDB, ICS and DAC. His current focus is on software & hardware codesign for graph machine learning.

Research Areas:
Computer Architecture, Programming Languages & Software, Systems & Networking

Impact Areas:
Big Data

See other events that are part of the Fast Code 2020 - 2021.

Created by Julian J. Shun at Thursday, April 22, 2021 at 11:26 AM.