CSE538 Online Wiki:Course overview & logistics

CSE 538 (Modeling and Performance Evaluation of Computer Systems)

COURSE WELCOME

Welcome to CSE 538T. My name is Roch Guerin and I am the instructor for the class. As part of this course, we will be communicating synchronously during zoom sessions or possibly in person when feasible (M-W 4:00-5:20pm), and asynchronously using Piazza that we will be using for all Q&A sessions. However, if you absolutely need to reach me personally, my email is guerin@wustl.edu. You can also find a bit more about myself either on my webpage or in the Instructor Background below.

Coming back to the course, note the "T" suffix of the course, which is meant to indicate that this is a Theory course. Hopefully, this is not a surprise (in its first year, the course had an "S" suffix and the students were then justifiably surprised. Not so anymore!), as we will be seeing quite a bit of it even if it all has some very practical applications. Now, the instructions below are meant to facilitate you getting started in the class, and so please read them carefully.

• First and foremost, make sure you bookmark somewhere the URL for the course wiki: https://classes.engineering.wustl.edu/cse538
  • The wiki sidebar shows you the different sections of the course wiki. Some are devoted to the course structure and approach, while others are meant to give you quick access to most of the material used in the course, e.g., lecture slides, problems we will use to practice material from the slides, assignments, etc.
  • The course “Syllabus & (tentative) schedule” offers a birds-eye view of the course, its content, what will be covered in what order, and due dates for assignments and quizzes, together with links to the corresponding material.
  • Review sessions/office hours have been scheduled 4 times a week in 45mins slots. The schedule and associated zoom links are here. They will be used to either go over problems to put the material in practice, or to answer specific student questions.
• Go over the rest of this page and the Course structure section and make sure you familiarize yourself with their content, including noting the due dates for the different types of assignments and the exam dates, making sure you have a copy of the course textbook, and carefully reading the course’s Academic Integrity Policy.
• Make sure you register on the course Piazza page, as we will use it for Q&A
• Familiarize yourself with the course Canvas site as all grades will be posted there and access to course videos will be provided there as well.
• Understand the purpose of the Preview problems and Reading Assignments
• Secure a copy of our textbook as most of the problems we will be working on come from there.
• And finally take a look at your first preview problem. Unlike subsequent Preview problems, it does not need to be turned in, but it will give you a sense of what to expect.

Course Description

Modern computing systems consist of multiple interconnected components, which all influence performance. The focus of this course is on developing modeling tools aimed at understanding how to design and provision such systems to meet certain performance or efficiency targets, and the trade-offs involved. The course covers Markov chains and their applications to simple queues, and proceeds to explore more complex systems including server farms and how to optimize their performance through scheduling and task assignment policies. The course includes a brief review of the necessary probability and mathematical concepts. Prerequisite: ESE 326.
Note: This is a graduate course, and so while we will go through a brief refresher on probability and transform methods, neither is meant to replace having taken a course on those topics.

Why Should I Take This Course? (or What Should I Expect to Learn from It?)

The course is about creating and solving models that allow you to evaluate the performance of different computer systems (though the techniques are general and applicable to a wide range of other systems).

A model has usually three components: (1) How work arrives to the system, (2) the structure of the system (can jobs wait for service and how many, in which order are they served, etc.), and (3) a characterization of the amount of work each job brings. Performance metrics of interest are commonly in the form of order statistics of quantities such as waiting time, system time (waiting + service), throughput (number of jobs processed per unit of time), loss probability (odds that a job is turned away because of a full system), etc.

The course will equip you with a set of tools for choosing the proper representations for each of those three components, and will show you how to combine and apply them in a number of standard system configurations associated with common computer systems. Through those examples, you will not only learn how to apply those tools, but also how to interpret the results they produce.

Instructor's Background

The bulk of my career has been spent developing and applying algorithmic and performance modeling solutions to problems in computer networking.

I spent about 12 years at IBM Research T.J. Watson developing what at the time were viewed as "next generation" networking solutions. Some of that work led to various quality-of-service enhancement to routers made by IBM, and to the development of what was supposed to be IBM's advanced packet switching network architecture called Networking BroadBand Services (NBBS). That work did not go anywhere as it was caught by the Internet tsunami (to which I also contributed to some extent by co-authoring a few RFCs such as RFC 2212, RFC 2676, RFC 2697, RFC 2698, RFC 2753), but on the modeling front it gave rise to the concept or equivalent or effective bandwidth that predicts how much bandwidth a variable rate connection needs to meet a certain performance level. I left IBM to join the University of Pennsylvania where I continued working on various networking problems, extending into the area of routing, traffic engineering, multimedia applications, etc., anything that had to do with improving performance over packet networks. I also became interested in questions of network economics and technology adoption in systems with positive and negative externalities, and this led to a separate branch of investigations that still continues to this day. Finally, I joined WashU in 2013 where besides continuing investigations related to network economics, I have explored performance questions in cloud systems with a focus on improving efficiency and enabling real-time guarantees, e.g., how to make the cloud useful to systems with tight latency requirements.

For more information see the Background section of my webpage.

Links

• Piazza course page: Please use it for all Q&A related to the class. (Note: You need to sign-up before being able to access the site).
• Canvas Site: Used only for posting grades.

---

Announcements

ALL ANNOUNCEMENTS ARE POSTED ON THE PIAZZA COURSE PAGE

Textbook(s)

M. Harchol-Balter, "Performance Modeling and Design of Computer Systems." Cambridge University Press (2013), ISBN: 978-1-107-02750-3

Logistics

• Class times and location: Monday & Wednesday, 4:00pm-5:20pm in TBD and zoom. You will need to login to zoom to be able to access the class zoom meeting.
  • First class is on Monday September 14, 2020, and last class is on Wednesday December 16, 2020.
• Class grading: Review problems 15%; Homeworks 50%, Quizzes 35%
• Instructor: Roch Guerin, Jolley 304C, email: guerin@wustl.edu
  • Review sessions/office hours: Tuesday 8:00-8:45am & 3:00-3:45pm + Thursday 8:00-8:45am & 3:00-3:45pm over zoom.
• TAs and office hours: TBD, if any

• Late policy: Homeworks and review problems are expected to be turned by their due date by uploading them to the corresponding assignment in the course Canvas site. There will be no exception and no partial credit.

• Asking questions: You can obviously ask face-to-face questions during any of the office hours, but the other and possibly better means to get your questions answered is to use the Piazza site setup for the course (you will need to register to be able to access it). The TA (if any) and I will monitor the site and do our best to answer questions in a timely manner. Posting questions on Piazza has the advantage that questions and answers are available to all students, and it is often the case that if you have a question, other students have a similar one. Because of that, neither the TA nor I will respond to email questions. So please use the Piazza site as much as you can and don't hesitate to ask as many questions as you need to; there is no such thing as a dumb question.

• ACADEMIC INTEGRITY: Please make sure you are familiar with Washington University Academic Integrity Policy, as it will be strictly applied.

University Policies & Resources on Sexual Assault, Bias, and Mental Health

Accommodating Special Requirements

Washington University is committed to providing accommodations and/or services to students with documented disabilities. Students who are seeking support for a disability or a suspected disability should contact Disability Resources at 935-4153. Disability Resources is responsible for approving all disability-related accommodations for WU students, and students are responsible for providing faculty members with formal documentation of their approved accommodations at least two weeks prior to using those accommodations. I will accept Disability Resources VISA forms by email and personal delivery. If you have already been approved for accommodations, please provide me with a copy of your VISA within the first two weeks of the semester.

Accommodations based upon sexual assault

The University is committed to offering reasonable academic accommodations to students who are victims of sexual assault. Students are eligible for accommodation regardless of whether they seek criminal or disciplinary action. Depending on the specific nature of the allegation, such measures may include but are not limited to: implementation of a no-contact order, course/classroom assignment changes, and other academic support services and accommodations. If you need to request such accommodations, please direct your request to Kim Webb (kim_webb@wustl.edu), Director of the Relationship and Sexual Violence Prevention Center. Ms. Webb is a confidential resource; however, requests for accommodations will be shared with the appropriate University administration and faculty. The University will maintain as confidential any accommodations or protective measures provided to an individual student so long as it does not impair the ability to provide such measures.

If a student comes to me to discuss or disclose an instance of sexual assault, sex discrimination, sexual harassment, dating violence, domestic violence or stalking, or if I otherwise observe or become aware of such an allegation, I will keep the information as private as I can, but as a faculty member of Washington University, I am required to immediately report it to my Dean or directly to Ms. Jessica Kennedy, the University’s Title IX Director. If you would like to speak with directly Ms. Kennedy directly, she can be reached at (314) 935-3118, jwkennedy@wustl.edu, or by visiting the Title IX office in Umrath Hall. Additionally, you can report incidents or complaints to the Office of Student Conduct and Community Standards or by contacting WUPD at (314) 935-5555 or your local law enforcement agency. See: Title IX.

You can also speak confidentially and learn more about available resources at the Relationship and Sexual Violence Prevention Center by calling (314) 935-8761 or visiting the 4th floor of Seigle Hall. See: RSVP Center.

Bias Reporting

The University has a process through which students, faculty, staff and community members who have experienced or witnessed incidents of bias, prejudice or discrimination against a student can report their experiences to the University’s Bias Report and Support System (BRSS) team. See: brss.wustl.edu

Mental Health

Mental Health Services’ professional staff members work with students to resolve personal and interpersonal difficulties, many of which can affect the academic experience. These include conflicts with or worry about friends or family, concerns about eating or drinking patterns, and feelings of anxiety and depression. See: shs.wustl.edu/MentalHealth

Center for Diversity and Inclusion (CDI)

The Center of Diversity and Inclusion (CDI) supports and advocates for undergraduate, graduate, and professional school students from underrepresented and/or marginalized populations, creates collaborative partnerships with campus and community partners, and promotes dialogue and social change. One of the CDI's strategic priorities is to cultivate and foster a supportive campus climate for students of all backgrounds, cultures and identities. See: diversityinclusion.wustl.edu/