Thursday Extra: "Finding Performance Problems with ALEX"

On Thursday, October 11, Mattori Birnbaum, Linh Bui, Zoe Grubbs, Hadley Luker, and Xinya Yang will give a talk on their summer 2018 Mentored Advanced Project:

Identifying performance problems in code is crucial for software developers. Even small inefficiencies can incur large costs in complex, long-running systems. While programmers can tune their programs by using efficient algorithms and data structures, hardware resources such as caches and branch predictors can still be a major source of inefficiency. Such inefficient uses of hardware resources are rarely obvious in the program's source code.

During summer 2018, we developed a tool called AnaLysis of EXecution (ALEX) to help developers diagnose performance problems. ALEX gathers and displays performance data from unmodified programs run on GNU/Linux. Our visualization helps developers quickly find patterns of poor performance, and the accompanying analysis leads developers to the relevant source code. In our talk, we will demonstrate how ALEX can help developers, explain how it works, and discuss our summer research experience.

At 4:00 p.m., refreshments will be served in the Computer Science Commons. The talk, "Finding Performance Problems with ALEX," will begin at 4:15 p.m. in Noyce 3821. Everyone is welcome to attend!

Algorithmic arts / CS Table: Live coding

This Friday at CS Table, we will discuss “live coding,” a form of performance art that typically involves sound, image, and visible code. (It's a bit more complex than that, as the readings suggest.) Jennelle and Colin will lead the the discussion.

The readings:

Syndicate content