Events

COMPASS: Computing Platforms Seminar Series

CAB E 72

Speaker: Eric Sedlar, Oracle Labs

Title: Why Systems Research Needs Social Science Added to the Computer Science

Abstract:

Computer scientists are very good at improving metrics that can be quantified: performance per-core/per-server/per-Watt, scalability, reliability, and are even getting better at a bit fuzzier metrics like accuracy of ML systems. However, it is many of the fuzziest metrics that are driving trends in computing: programmer productivity, usability, cognitive load and/or degree of security provided by a particular system. The biggest trend in computing for the past few decades is the explosion in the use of open-source software in the bulk of computing tasks. This is true even in environments as security-conscious as defense applications, as the stack that needs to execute in an application becomes too complicated for one programmer or one software vendor to comprehend or master. This move to open source software makes most system metrics worse as much of the code run is not optimized for CPU efficiency and may be understood by nobody working for the firm operating the software or its vendors. What is a systems researcher to do in the face of their inconsequential metrics?

Shortbio:

As VP & Technical Director of Oracle Labs, Eric manages a team of close to 200 systems researchers and engineers worldwide. In his tenure in the Labs, Eric has started a number of long-term system research projects that have led to technology transfer into products, including the GraalVM programming language runtime, PGX Parallel Graph Analytics, and the Parfait tool for Program Analysis. His personal research interests have been in the field of data processing and the intersection with compiler technologies. Eric was the co-author of the SIGMOD Best Paper in 2009 and has been an inventor on 85 granted patents.

COMPASS Talks