Computer Systems Laboratory Colloquium (2007-2008)

Video Lectures

Displaying all 22 video lectures.
Lecture 1
The PeakStream Platform for Many-Core Computing
Play Video
The PeakStream Platform for Many-Core Computing


September 26, 2007

Lecture by Matthew Papakipos for the Stanford University Computer Systems Colloquium (EE 380). The history of PeakStream is told - from startup to Google acquisition - and many-core processors are introduced; challenges, interesting design choices and solutions in the development of the company's software are also discussed.

Lecture 2
Programmable Microfluidics
Play Video
Programmable Microfluidics


October 3, 2007

Lecture by Bill Thies for the Stanford University Computer Systems Colloquium (EE 380). Bill Thies provides an overview of microfluidic technologies from a computer science perspective, highlight areas in the which computer science researchers can contribute to this field; he will also describe recent work in developing new architectures, programming languages, and CAD tools for the microfluidic domain.

Lecture 3
Open Source Study: Analytics, Economics & Best Practices
Play Video
Open Source Study: Analytics, Economics & Best Practices


October 10, 2007

Lecture by Dirk Riehle for the Stanford University Computer Systems Colloquium (EE 380). This talk discusses the economics driving stakeholder behavior in the open source ecosystem, presents selected analytical results of how open source works (or doesn't) and takes a look at how companies can benefit from employing open source best practices internally.

Lecture 4
Self-Improving Artificial Intelligence
Play Video
Self-Improving Artificial Intelligence


October 24, 2007

Lecture by Steve Omohundro for the Stanford University Computer Systems Colloquium (EE 380). Steve presents fundamental principles that underlie the operation of "self-improving systems," i.e., computer software and hardware that improve themselves by learning from their own operations.

Lecture 5
The Challenges of Implementing Matlab
Play Video
The Challenges of Implementing Matlab


October 31, 2007

Lecture by Randy Allen for the Stanford University Computer Systems Colloquium (EE 380). Some of the compiler optimization challenges presented by a naturally-interpreted, dynamically-typed, vector-centric language such as MATLAB, and how compiler theory can be both extended or stretched to accommodate these overly-hyphenated challenges are discussed.

Lecture 6
Parallel Programming 2.0
Play Video
Parallel Programming 2.0


November 7, 2007

Lecture by Renee James and Wei Li for the Stanford University Computer Systems Colloquium (EE 380). Renee and her co-presenter touch upon Intel's processor direction for multi-core and then dive into its impact to software; parallel programming in the mainstream has different characteristics than in the previous era when it was restricted to high performance computing - the two cover this new paradigm, its challenges and some of the solutions being developed by the software industry.

Lecture 7
On the Road to Computer Literacy
Play Video
On the Road to Computer Literacy


November 14, 2007

Lecture by for the Stanford University Computer Systems Colloquium (EE 380). In order to identify some of the technological gaps that hinder the implementation of universal computer literacy, which Robert defines as society's ability to read and write computer programs, this talk looks back at the development of literacy and the associated technologies from before Plato through Aldus Manutius and beyond.

Lecture 8
Android: Building a Mobile Platform to Change the Industry
Play Video
Android: Building a Mobile Platform to Change the Industry


November 28, 2007

Lecture by Richard Miner for the Stanford University Computer Systems Colloquium (EE 380). This talk provides technical and commercial background on Android, a new mobile platform that Google and a large set of partners (known as the open handset alliance) have recently announced.

Lecture 9
Science Communication, Science Literacy and Public Support
Play Video
Science Communication, Science Literacy and Public Support


December 5, 2007

Lecture by Rob Semper for the Stanford University Computer Systems Colloquium (EE 380). Rob discusses the new models of public science literacy that are beginning to change the paradigm from a one-way transmission approach of lecture, publishing and journalism to one of public engagement and even debate: from public understanding to public dialog to the role of science centers as intermediary institutions between the world of science and the public, new thinking is emerging for this critical arena.

Lecture 10
Tracking the Internet into the 21st Century
Play Video
Tracking the Internet into the 21st Century


January 9, 2008

Lecture by Vint Cerf for the Stanford University Computer Systems Colloquium (EE 380).



Vint Cerf, the Chief Internet Evangelist at Google, talks to an audience about the past and future of the internet and its continuing development. His talk includes information about current needs for internet research including internet security, improved networks and IPv6 addresses.

Lecture 11
Computing in Transition
Play Video
Computing in Transition


February 20, 2008

Lecture by Nick Tredennick for the Stanford University Computer Systems Colloquium (EE 380).



Nick Tredennick talks about the semiconductor industry and its impact in the world. He takes the audience through the history of semiconductors and where he believes they are heading in the future as well as the economics of its development and the shrinking market for cutting edge technological growth.

Lecture 12
Scalable Parallel Programming with CUDA on Manycore GPUs
Play Video
Scalable Parallel Programming with CUDA on Manycore GPUs


February 27, 2008

Lecture by John Nickolls for the Stanford University Computer Systems Colloquium (EE 380).



John Nickolls from NVIDIA talks about scalable parallel programming with a new language developed by NVIDIA, CUDA. NVIDIA's programming of their graphics processing unit in parallel allows for the dissection of large data sets into smaller sets, each to be handled by separate processors. This significantly increases the performance and handling of processing intensive application.

Lecture 13
Building a Safer Web
Play Video
Building a Safer Web


March 12, 2008

Lecture by Charles Reis for the Stanford University Computer Systems Colloquium (EE 380).



Web content has shifted from simple documents to active programs, but web protocols and browsers have not evolved adequately to support them. As a result, safety problems in web sites and web browsers now regularly make headlines, from browser exploits to ISPs that modify web pages. In this talk, Charles Reis discusses his research into improving the security and reliability of web content and browsers.

Lecture 14
CMOS Process Variations: A Critical Operation Point Hypothesis
Play Video
CMOS Process Variations: A Critical Operation Point Hypothesis


April 2, 2008

Lecture by Janak H. Patel for the Stanford University Computer Systems Colloquium (EE380).



Prevailing understanding of a chip's behavior under large process variations with statistical delay assumptions leads one to conclude that a small number of errors are likely as we progress further down on Moore's Law. This understanding is challenged by a new hypothesis which states that in every large CMOS chip, there exist critical operations points (frequency, voltage, temperature) such that it divides the 3-D space in to two distinct spaces: Error-free operation and Massive errors.

Lecture 15
Demonstration of Brain Computer Interface Using the Emotive Epoc
Play Video
Demonstration of Brain Computer Interface Using the Emotive Epoc


April 9, 2008

Lecture by Randy Breen for the Stanford University Computer Systems Colloquium (EE380).



The Emotiv EPOC (www.emotiv.com) now makes it possible for games to be controlled and influenced by the player's mind. Engaging, immersive, and nuanced, Emotiv-inspired game-play will be like nothing ever seen before. Based on the latest developments in neuro-technology, Emotiv has developed a new personal interface for human computer interaction.

Lecture 16
A Head in the Cloud: The Power of Infrastructure as a Service
Play Video
A Head in the Cloud: The Power of Infrastructure as a Service


April 23, 2008

Lecture by Dr. Werner Vogels for the Stanford University Computer Systems Colloquium (EE380).



This presentation reviews some of the lessons learned from building one of the world's largest distributed systems; Amazon.com. Dr. Vogels focuses on state management which is one of the dominating factors in the scalability, reliability, performance and cost-effectiveness of the overall system.

Lecture 17
Distributed Systems: Computation With a Million Friends
Play Video
Distributed Systems: Computation With a Million Friends


April 30, 2008

Lecture by Adam L. Beberg for the Stanford University Computer Systems Colloquium (EE380).



Distributed systems provide powerful capabilities to those willing to embrace the uncertainty involved. This talk explores the current methods for constructing these systems, the 35 years of history they draw upon, and active work integrating massive storage and on-demand post-processing into a volunteer-powered system dubbed [email protected]

Lecture 18
Dynamic Languages Strike Back
Play Video
Dynamic Languages Strike Back


May 7, 2008

Lecture by Steve Yegge for the Stanford University Computer Systems Colloquium (EE380).



Dynamically typed programming languages such as Perl, Python and Ruby have been gradually gaining popularity and momentum for the past fifteen years. However, dynamic languages are also arguably the biggest source of controversy in the industry. In this talk, Steve Yegge debunks some of the issues considered central to the debate, and then shares some novel techniques people are using to produce static-quality tools and performance in dynamic languages.

Lecture 19
The Challenge of Small Form Factor: The ASUS Eee PC
Play Video
The Challenge of Small Form Factor: The ASUS Eee PC


May 14, 2008

Lecture by Jerry Dien for the Stanford University Computer Systems Colloquium (EE380).



The ASUS Eee PC, which stands for "Easy to Learn, Easy to Work and Easy to Play," has been a tremendous success for ASUS since its initial launch in October of 2007. The 7-inch, 2 pound mobile internet gadget has also set a new trend for the PC industry and defined a whole new category of portable computers.

Lecture 20
Spookytechnology and Society
Play Video
Spookytechnology and Society


May 21, 2008

Lecture by Dr. Charles Tahan for the Stanford University Computer Systems Colloquium (EE380).



In this lecture, Dr. Charles Tahan gives a non-physics perspective on quantum information science and technology, both via some of his own scientific work, but mostly from the vantage of science and technology studies. He touches on why the new quantum revolution is for real, what "new quantum revolution" actually means, why it's more interesting than nanotechnology, and how we should prepare for it.

Lecture 21
The Search for Jim Gray
Play Video
The Search for Jim Gray


May 28, 2008

Lecture for the Stanford University Computer Systems Colloquium (EE380).



When Jim Gray, a noted computer scientist and Turing award winner, disappeared at sea, a massive effort was mounted to find him. A panel of speakers touches on both the technical approaches employed in the search and on the social side where the community organized a massive effort quickly and efficiently.



Panel Members:

Captain David J. Swatland (Deputy Sector Commander for the Coast Guard Sector San Francisco)

Donna Carnes (Jim Gray's Wife)

Mike Olson (Long time friend of Jim Gray; founder of Sleepcat Software, former Oracle VP)


Lecture 22
The Role of Accelerated Computing in the Multi-Core Era
Play Video
The Role of Accelerated Computing in the Multi-Core Era


June 4, 2008

Lecture by Charles Moore for the Stanford University Computer Systems Colloquium (EE380).



In recent years, we have seen a decline in the rate of improvement on several traditional drivers of value in computer systems, namely transistor performance, wire delays, the return on deep pipelining, and techniques for extracting high numbers of instructions per cycle. In this talk, Charles Moore highlights the implications of some of these shifts and makes some observations about the emergence of a new framework for future innovation.