Take note, there is a fun, new, four credit CSE course being offered this fall. Last minute addition. This is a graded 400 level course so it will apply towards CSE senior electives. Registration is now open.
Course Number: CSE 490m
Title: Introduction to Really Distributed Systems
SLN: 23153 (add code has been removed)
Instructors: Professor John Zahorjan and guest instructor Russel Power
Enrollment limit: 12
Meeting times: MW 3:30-4:50
Prereq: Substantial programming experience, ideally finish required 300’s and likely some 400’s completed
Email advising if you are uncertain or show up on day one to ask questions.
Course announcement:
This project course provides experience with today’s most common
computing platform: mobile devices accessing a proprietary cloud. The
course imposes a coarse software architecture, and suggests an example
application domain, but most of the decisions about what is built are
left to the students. http://homes.cs.washington. edu/~zahorjan/cse490r/583165. pdf
presents a survey of applications and implementations similar in spirit
to those we envision for this course.
The software architecture is composed of three pieces: the front end,
the back end, and applications. The front end views the mobile
devices as display and sensing hardware and the cloud as a data
repository. Sensor values are uploaded to that repository.
The back end operates on the stored sensor values. It may provide
access to inferred values – values that are not direct readouts
of sensors, but may be (approximately) distilled from them.
The emphasis on the back end is on processing, in the style of data
mining. Applications make use of the back end to provide useful
functionality.
Our motivating example applications imagine using the infrastructure
to provide airplane black box functionality — recording and
persisting sensor data from the mobile device on a regular basis. We
are interested in both an individual’s use of his/her own data (“how
much time do I spend on average waiting for my transfer?”) and
“community sensing,” in which data from a large number of devices are
fused to answer questions about broad effects (“How does the Fremont
bridge going up affect travel times in the U district?”).
Students will be expected to contribute substantially to the
specification, design, and implementation of this system. There will
be regular meetings of the class as a whole, as well as smaller
meetings between the instructors and individual teams.