CSE160 Wi16 - Home

Staff

Scott Baden, Professor (baden+160@eng.ucsd.edu)
Jingjing Xie, TA
Karthikeyan Vasuki Balasubramaniam, TA
John Hwang, Tutor
Ryan Lin, Tutor

Location and time

Lecture: Tue Thu 3:30p-4:50p (WLH 2005)
Section: Wed 4:00-4:50A (CNTR 216), Fri 12:00-12:50 (CNTR 105)
Lab hours: https://goo.gl/iXFi1i
Final: Tue 3:00-5:59A (WLH 2005) New

Office Hours

First week of classes: Thursday, Jan 7, 5pm to 6pm

Piazza and Moodle

We will use Piazza for forums and course announcements. We will use Moodle for reading quizzes and to post grades.

Textbooks

The course has 2 required textbooks. Please see these instructions for how to obtain the Williams text, as I have secured a 40% discount which is not available at the book store.

iClickers

Buy an iClicker clicker at the bookstore (it is required for the course). Register your iClicker at https://www1.iclicker.com/register-clicker so you can collect your clicker points for your grade. Use your UCSD ID (PID starts with A) when registering. You must register your iClicker by Monday at noon (January 11, 2016). If you miss this deadline, you will receive 0 credit for participation for the entire quarter.

If your your remote ID has rubbed off or is illegible, use this Remote ID Lookup Tool

What you will learn in CSE 160

Nearly all of us have a parallel computer at our fingertips, yet most of us use the parallel processors only indirectly, through software written by others. In CSE 160 you will learn how to author your own multithreaded software to realize high performance on multicore processors. The course will cover parallel architectures, programming models, algorithms and software tools, enabling you to make informed decisions about efficient parallel application design and implementation. We’ll consider applications–both ordinary and extraordinary–and develop a systematic approach for building effective parallel variants.

Syllabus

Fundamentals: execution, programming and memory models, architecture, performance, applications
Software and programming: multithreading- OpenMP and C++11 threads; SIMD; correctness
Parallel algorithm design and implementation: case studies of applications; data structures; performance tradeoffs; performance programming; memory hierarchy optimizations

CSE 160 provides a foundation for pursuing more advanced topics in parallel computation, and conducting research in this exciting field.

Our primary testbed will be the Bang cluster at located at UCSD, which is funded by the National Science Foundation’s Major Research Instrumentation program.

Prerequisites

The prerequisite for CSE 160 is CSE 100. CSE 160 entails a signfifcant programming workload and you should be comfortable programming in C or C++.