QuickQuery: GPU-Based Approximate Query Processing for Sub-Second Exploration at Scale

Speaker: Larry Rudolph (Two Sigma Investments, LP and CSAIL Affiliate) and Steven Martin (Two Sigma Investments, LP)

Date: Monday, July 27, 2020

Time: 2:00 PM to 3:00 PM Note: all times are in the Eastern Time Zone

Public: Yes

Location: https://mit.zoom.us/meeting/register/tJUrdOqopj8uHdO4gUyVMnfglOFEqIye_Je0 (Registration required)

Event Type: Seminar

Room Description:

Host: Julian Shun, MIT CSAIL

Contact: Julian Shun, jshun@mit.edu, lindalynch@csail.mit.edu

Relevant URL: http://fast-code.csail.mit.edu/

Speaker URL: https://www.linkedin.com/in/larryrudolph/, http://www.cubicsky.com/~steven/

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: QuickQuery: GPU-Based Approximate Query Processing for Sub-Second Exploration at Scale

Abstract:
Exploring data for signals can be viewed as a series of questions, queries, and hypotheses, each of which leads to either deeper exploration or a new line of inquiry. QuickQuery employs sophisticated sampling, exploits memory hierarchy, and leverages the power of GPUs to provide sub-second response times no matter the size of the dataset. Even with trillion-row datasets, researchers can execute many such queries at the speed of thought, all while providing confidence bounds on the results. QuickQuery lets the researcher specify constraints on both accuracy and the response time. It is highly tuned to be very performant; our proof-of-concept implementation on a single GPU server was more than 500 times faster than a 400-core Spark implementation for simple queries, and more than 1,000 times faster for more complex ones.

Bio: Larry Rudolph received his PhD from Courant Institute, NYU (Combining Fetch-and-Add), Postdoc at University of Toronto (The Balanced Sorting Network), and was on the faculty at Carnegie Mellon University (Snoopy Caching & The Skiing Analogy), The Hebrew University (Free Space Optical Interconnection Networks), New England Complex Systems Institute, and the Massachusetts Institute of Technology (Column Caching & Commit-Reconcile-Fence). He worked at Bell Labs, Digital Equipment Corp, Thinking Machines, IBM TJ Watson, and VMWare. He co-founded and was CTO of ReDigi (an on-line used music store). He is currently a member of the Research Labs at Two Sigma Investments, , a visiting scholar at NYU (recommender-verifiers), affiliate at M.I.T and Boston University, a member of the Artificial Intelligence Finance Institute, and a board member of the Massachusetts Open Cloud, and member of RxCovia Research Group.

Steven is a Quantitative Software Engineer at Two Sigma Investments. His primary focus is on applying modern hardware and systematic performance engineering to scalable data analysis. Supporting interests for this are high-performance computing, uncertainty quantification, computer architecture, and visualization. He has a PhD in Computer Science from The Ohio State University, and conducted research in high performance scientific visualization at Los Alamos National Lab, Siemens, and Nvidia.

Research Areas:
Algorithms & Theory, Computer Architecture, Programming Languages & Software

Impact Areas:
Big Data

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

Created by Julian J. Shun at Wednesday, July 22, 2020 at 11:47 PM.