READ THIS ENTIRE DOCUMENT
Instructor: Akshitha Sriraman
Office Location: CIC 4114
Email Address: akshitha@cmu.edu
Office Hours: Mondays, 2:30 - 3:30 pm ET
Teaching Assistant: Melissa Pan
Email Address: melissapan@cmu.edu
Office Hours: Email to schedule a Zoom appointment
Course Support: Academic Services Center
Office Location: HH 1113
Website: https://www.ece.cmu.edu/academics/academic-services-center.html
Course Description: Modern data center systems support a wide range of hyperscale web services such as cloud computing, social networks, video streaming, online messaging, web search, and online banking. This course will focus on studying the systems software, hardware, and distributed systems technology that compose modern data center computing systems. The course will also expose students to cross-cutting data center problems related to service level objectives, latency unpredictability, total cost of ownership, energy efficiency, scalability, sustainability, and equity.
The course is a combination of paper reading and reviewing, in-class presentations, paper disucssions, and a semester-long project. Students will read up to two seminal papers per topic and submit brief summaries via hotcrp. In the classroom, we will have a student presentation of the papers accompanied by an interactive discussion on the papers in terms of design innovation and challenges. From time to time, we will have guest speakers who are typically the authors of the paper being discussed. Students will work in groups of three on a semester-long, open-ended research project on a cutting-edge data center topic.
This course is appropriate for graduate and advanced undergraduate students from ECE and CS who are interested in a vertical study of modern cloud computing and advanced data center systems. It is also appropriate for ECE and CS students who want to gain some experience with a semester-long research project on a cutting-edge modern computer systems topic. This course is especially suited for students who are more interested in open-ended discussions, are interested in learning new concepts on-the-fly, and have a curiosity to explore the unknown.
Students are expected to attend class meetings and actively participate in the discussions. Paper summaries, presentations, and class participation are a significant part of the grade.
Number of Units: 12
Pre-requisites: There is no formal prerequisite for this course. However, students tend to do well in this course when they have a good understanding of systems and architecture concepts and are comfortable with C/C++ and/or Python programming (e.g., students who have taken 18-613: Foundations of Computer Systems or an introductory cloud computing course).
Graduate Area: Computer Systems
Class lecture: MW 3:30 - 4:50 pm ET at WEH 5302. Class lectures will not be recorded since we have moved to in-person teaching. However, if you are unable to attend class for some reason, please let the professor know in advance; we are more than happy to accommodate your needs.
Required Textbook: None
Suggested Reading: The Datacenter as a Computer
Other Supplemental Materials: Papers that will be shared through the course
Course Canvas: We will not use canvas for this course.
Course Website: You should check this website regularly for changes in schedule and other announcements.
Discussion Platform: Students enrolled in the course should have received a Slack invitation. If you are on the waiting list for this course, you will not receive a Slack invitation. If you have recently enrolled in the course, you will be added to the Slack channel 24 - 48 hours after enrollment. Please contact the instructors via email if you still have not received a Slack invitation.
- (30%) Course Project: Students will work in teams of three on a research project on improving data center systems. Deliverables will include a proposal document, deliverables document, milestone one document, milestone two document, final presentation, and final project report. This assignment will provide students with the experience of performing research in the field of modern data center computing.
- (30%) Paper Reviews: We will be writing reviews for the papers that we read in this course. These reviews will help practice critical examination of research papers related to data center computing. Each student can skip a total of three reviews across the semester (choose wisely!). Each miss beyond three will result in 25% decrease in grade for this portion of the course. Reviews will be submitted through hotCRP. We will discuss the guidelines for writing reviews during the first couple of lectures.
- (15%) Paper Presentations: Each student will be responsible for being a part of approximately three to five paper presentations during the semester. Each student will mostly be able to choose the papers that they present. We will discuss guidelines for presentations during the first couple of lectures.
- (25%) Participation: Much of the discovery and learning in this course will be driven by in-class discussion. Participation will be assessed by engagement during in-class discussions. Therefore, it is critical that students attend lectures and actively participate.
- All paper reviews will be submitted via hotcrp (Create an account ASAP!)
- We will read and review up to two research papers before each class. Each student will be a part of approximately three to five such paper presentations during the course of the semester.
- To submit a review after logging in, click on the paper you wish to review and follow the prompts to enter your review in terms of "strengths," "weaknesses," etc. Here is the review template we will use.
- The paper reviews allotted for each class are due on the day of the class by 11 am ET.
- Please submit your review exactly once.
- Please read other students' reviews after you submitting your own review. Please click on the "Good review" button for the reviews you like. Please refrain from clicking on the "Needs work" button for other students' reviews (you will be penalized for this).
- For additional information, please check the FAQ section below.
- Each student will be a part of about three to five paper presentations during the semester.
- For each presentation, one student will present a 5 minute summary, another student will advocate for the paper's strengths (5 minutes), and the third student will present critiques (5 minutes). Each student will present for no more than 5 minutes; please refrain from exceeding this time limit - you will be timed.
- When presenting the paper summary: The summary of each paper should be brief (i.e., 5 minutes). For example, an effective summary might include a slide for each of the following: (1) research problem motivation, (2) limitations of the state-of-the-art, (3) paper's main contribution, (4) high-level design description, (5) summary of one main result. Please include your slides under the "summary" section for the appropriate paper here.
- When advocating for a paper: Create a few slides (5 minute presentation) to (1) summarize the strengths that others in the class pointed out (please include the student's name when talking about their review), (2) point out the strengths that you particularly liked and show why these strengths are important, (3) identify how these strengths can be extended beyond the context of the paper (e.g., can aspects of the design be applied to a different problem?) - this third point is the most important part of your presentation. Please include your slides under the "advocate" section for the appropriate paper here.
- When critiquing a paper: Create a few slides (5 minute presentation) to (1) summarize the opportunities for improvement that others in the class pointed out (please include the student's name when talking about their review), (2) point out opportunities for improvement and show how/why these may (or may not) be deal breakers, (3) detail how you might have executed the work differently to overcome these weaknesses - this third point is the most important part of your presentation. Please include your slides under the "critic" section for the appropriate paper here. Please be respectful when critiquing a paper.
- When advocating for or critiquing against a paper, please try to make substantial points that are technical (e.g., should the experiments have included a different baseline? Is the proposed system falling short in some way, such as causing an increased power consumption?) - avoid basing your argument on non-technical aspects such as the paper's readability, graph layouts, organization, etc.
- All presentations will be accessible via a single google slides link.
Students will work in groups of three on a semester-long, open-ended research project on a cutting-edge data center topic (larger/smaller groups will not be approved). Please note that it is critical to work on the project every single week to make significant progress on an open-ended research problem by the end of the semester. If you are stuck at any point in the project, please attend OH or schedule a separate conversation with the instructor/TA to unblock yourself. If you happen to be working with a PhD student on your project, please refrain from contacting the PhD student one week before the deadline expecting an immediate response.
The goal of each project is to (1) explore an open-ended research problem, (2) conduct a brief literature survey, (3) become comfortable with setting up and running data center applications/infrastructure, (4) become comfortable with tooling and infrastructure, (5) conduct meaningful experiments to show at least one new result that has not been shown before (this could be a positive/negative result), (6) communicate your work via speaking and writing. You may use Overleaf to create your project document. Please find the project milestones below.
- (Aug 30/ASAP) Find project partners: Create a group of three like-minded students (e.g., does everyone in your group want to work on a software solution?) Please feel free to use the "project-groups" Slack channel to find your project peeps.
- (Sep 8, 11:59 pm ET) Submit project proposal: Email the instructors with your group's project proposal document describing your potential project topic (1 - 2 paragraphs). The expectation for this document is quite minimal. You are expected to describe the research project that you are potentially interested in pursuing (nothing is set in stone yet). List your group members in this document. During the first week, we will share a list of potential project topics that you can choose from, via Slack. You are also welcome to pick your own topic as long as you run it by us first.
- (Sep 18, 11:59 pm ET) Finalize deliverables: Finalize project deliverables after regular back-and-forth discussions with the instructors (you should have these conversations during OH, via slack, or schedule a separate time). You should have received approval from the instructors before submitting your deliverables document (so plan accordingly). Email the instructors with your group's project deliverables document, describing what you will tentatively achieve for milestone 1, milestone 2, and the final report (<=1 page). Your document will answer the following questions: (1) What is the specific goal of your project? (2) What will you (tentatively) achieve in each of the milestones?
- (Oct 9, 11:59 pm ET) First milestone: Submit your first milestone document that defines and motivates the problem, surveys related work, forms an initial hypothesis and idea, and includes some preliminary result to show that the idea is promising (email instructors). This document should be less than 5 pages (not including references).
- (Nov 8, 11:59 pm ET) Second milestone: Submit your second milestone document with a brief description of the idea, your design and implementation details, and a portion of your evaluation (email instructors). This document should be less than 10 pages (not including references).
- (Dec 6, 11:59 pm ET) Final presentation: Prepare a brief presentation (~10 minutes) of your project work and present it with your group. Divide your presentation such that each person presents for an equal amount of time. Each member must participate in answering questions on the project.
- (Dec 11, 11:59 pm ET) Final report: Submit a final paper on your work that is similar to the papers that we read and discussed in class. This document should be less than 10 pages (not including references).
- The goal of this class is for all of us to learn and grow together via open-ended discussions.
- Please feel free to raise questions/comments at any point during the lecture .
- During the class, we will have several discussions in small groups. Please actively engage in these small group discussions.
- Please do not engage with the same small group all the time. Move to a different group from time-to-time.
- The ECE Numbers cluster: You will likely have to share these resources with other folks. If your project involves fine-grained performance measurements on real hardware, this might not be the best option since you might encounter interference effects from co-runners.
- The ECE Data Science cloud: You will likely have to share these resources with other folks. If your project involves fine-grained performance measurements on real hardware, this might not be the best option since you might encounter interference effects from co-runners.
- CloudLab: Please contact the instructor with your email addresses. The instructor will create a project on CloudLab for you and add you to the project. You will get dedicated access to machines using CloudLab and will therefore not encounter interference effects.
- Chameleon Cloud: Please contact the instructor with your email addresses. The instructor will create a project on Chameleon Cloud for you and add you to the project. You will get dedicated access to machines using Chameleon Cloud and will therefore not encounter interference effects.
- Pittsburgh Supercomputing center: If you need substantial computing, we can do an education allocation at the Pittsburg Supercomputing Center. Here, we can get access to the Bridges and Bridges-DL machine with 60,000 CPU cores and about 80+ A100 GPUs. Of course you wouldn't use the entire machine, but you have flexibility.
- Google Cloud Platform: Contact the instructor if you need access to GCP educational credits. If your project involves fine-grained performance measurements on real hardware, this might not be the best option since it might be harder to measure hardware performance counters, you might face interference effects from co-runners, might face overheads from virtualization, etc.
- Please contact the instructor if none of the above computing options work for you. We will work with you to meet your needs as best as we can.
Please submit a review for each paper on hotcrp by 11 am ET on the day of the class. Here is the review template we will use.
| Date | Topic | Readings | Notes | Summary | Advocate | Critic |
|---|---|---|---|---|---|---|
| Aug 28 (M) Lecture 1 | Introduction to Data center Computing | Introduction | - Find project partners - Form due |
|||
| Aug 30 (W) Lecture 2 |
Introduction to Data center Computing | Introduction | - Finalize project partners |
Akshitha Sriraman | ||
| Sep 4 (M) | Holiday: No Class | |||||
| Sep 6 (W) | An Open-Source Benchmark Suite for Microservices and Their Hardware-Software Implications for Cloud and Edge Systems | Web application paradigms | - Submit project proposal document describing project topic (<=1 page) on Sep 8 -Supplemental reading |
Sara Mahdizadeh Shahri | Shruti Mittal | Tianyun Zhang |
| Sep 11 (M) | Profiling a Warehouse-Scale Computer | Microarchitecture analysis | Simon Spivey | Simon Men | Veronica | |
| Sep 13 (W) | SoftSKU: Optimizing Server Architectures for Microservice Diversity @Scale | Microarchitecture analysis | -Supplemental reading | Minh Truong | Simon Spivey | Sara Mahdizadeh Shahri |
| Sep 18 (M) | A Reconfigurable Fabric for Accelerating Large-Scale Datacenter Services | Hardware acceleration | - Supplemental reading | Jeff Chen | Xuesi Chen | Rachana Murali Narayanan |
| Sep 20 (W) | No Class | - Submit project deliverables document (<=1 page) |
||||
| Sep 25 (M) | A hardware accelerator for protocol buffers | Hardware acceleration | Guest: Sagar Karandikar (UC Berkeley) - Supplemental reading 1 Supplemental reading 2 |
Lucas Castanheira | Jeff Chen | Xuesi Chen |
| Sep 27 (W) | AsmDB: Understanding and Mitigating Front-End Stalls in Warehouse-Scale Computers | Compiler/PGO | - Supplemental reading | Xuesi Chen | Edwin Lim | Simon Men |
| Oct 2 (M) | -The Tail at Scale -Attack of the Killer Microseconds |
Tail latency, Killer microseconds | Supplemental reading | Veronica | Sara Mahdizadeh Shahri | Lucas Castanheira |
| Oct 4 (W) | The Demikernel Datapath OS Architecture for Microsecond-scale Datacenter Systems | OS | - Guest: Thomas Wenisch (Google) - Supplemental reading |
Tianyun Zhang | Minh Truong | Edwin Lim |
| Oct 9 (M) | Dynamo: facebook's data center-wide power management system | Power/energy | - Guest: Justin Meza - First project milestone due (<5 pages) |
Shruti Mittal | Rachana Murali Narayanan | Jeff Chen |
| Oct 11 (W) | Resource Central: Understanding and Predicting Workloads for Improved Resource Management in Large Cloud Platforms | Resource management/scheduling | - Supplemental reading | Minh Truong | Simon Spivey | Simon Men |
| Oct 16 (M) | Fall break: No Class | |||||
| Oct 18 (W) | Fall break: No Class | |||||
| Oct 23 (M) | Serverless in the Wild: Characterizing and Optimizing the Serverless Workload at a Large Cloud Provider | Serverless computing | - Guest: Mohammad Shahrad (UBC) - Supplemental reading |
Lucas Castanheira | Lucas Castanheira | Tianyun Zhang |
| Oct 25 (W) | Cost-Efficient Overclocking in Immersion-Cooled Datacenters | DC cooling | Jeff Chen | Veronica | Shruti Mittal | |
| Oct 30 (M) | Chasing Carbon: The Elusive Environmental Footprint of Computing | Sustainability | Rachana Murali Narayanan | Simon Men | Simon Men | |
| Nov 1 (W) | ACT: Designing Sustainable Computer Systems With An Architectural Carbon Modeling Tool | Sustainability | - Supplemental reading |
Xuesi Chen | Tianyun Zhang | Edwin Lim |
| Nov 6 (M) | - Nines are not enough | DC business model | Sara Mahdizadeh Shahri | Minh Truong | Veronica | |
| Nov 8 (W) | No paper review | Ethics | - Second project milestone due (<10 pages) |
|||
| Nov 13 (M) | Software-defined far memory in warehouse-scale computers | ML for datacenter systems | - Supplemental reading | Edwin Lim | Rachana Murali Narayanan | Simon Spivey |
| Nov 15 (W) | Bolt: I Know What You Did Last Summer... in the Cloud | Security | Supplemental reading | Veronica | Shruti Mittal | Rachana Murali Narayanan |
| Nov 20 (M) | No Class | |||||
| Nov 22 (W) | Thanksgiving: No Class | |||||
| Nov 27 (M) | Cores that don't count | Reliability | Edwin Lim | Jeff Chen | Minh Truong | |
| Nov 29 (W) | No Class | |||||
| Dec 4 (M) | How to excel as a systems engineer? | |||||
| Dec 6 (W) | Final project presentations | |||||
| Dec 11 (M) | Final project report due |
The ECE department is accredited by ABET to ensure the quality of your education. ABET defines 7 Educational Objectives that are fulfilled by the sum total of all the courses you take. The following list describes which objectives are fulfilled by this course and in what manner they are fulfilled. The objectives are numbered from “1” through “7” in the standard ABET parlance. Those objectives not fulfilled by this course have been omitted from the following list:
- an ability to identify, formulate, and solve complex engineering problems by applying principles of engineering, science, and mathematics
- an ability to apply engineering design to produce solutions that meet specified needs with consideration of public health, safety, and welfare, as well as global, cultural, social, environmental, and economic factors
- an ability to communicate effectively with a range of audiences
- an ability to recognize ethical and professional responsibilities in engineering situations and make informed judgments, which must consider the impact of engineering solutions in global, economic, environmental, and societal contexts
- an ability to function effectively on a team whose members together provide leadership, create a collaborative and inclusive environment, establish goals, plan tasks, and meet objectives
- an ability to develop and conduct appropriate experimentation, analyze and interpret data, and use engineering judgment to draw conclusions
- an ability to acquire and apply new knowledge as needed, using appropriate learning strategies.
(http://www.ece.cmu.edu/programs-admissions/masters/academic-integrity.html):
The Department of Electrical and Computer Engineering adheres to the academic integrity policies set forth by Carnegie Mellon University and by the College of Engineering. ECE students should review fully and carefully Carnegie Mellon University's policies regarding Cheating and Plagiarism; Undergraduate Academic Discipline; and Graduate Academic Discipline. ECE graduate student should further review the Penalties for Graduate Student Academic Integrity Violations in CIT outlined in the CIT Policy on Graduate Student Academic Integrity Violations. In addition to the above university and college-level policies, it is ECE's policy that an ECE graduate student may not drop a course in which a disciplinary action is assessed or pending without the course instructor's explicit approval. Further, an ECE course instructor may set his/her own course-specific academic integrity policies that do not conflict with university and college-level policies; course-specific policies should be made available to the students in writing in the first week of class. This policy applies, in all respects, to this course.
CMU Academic Integrity Policy (http://www.cmu.edu/academic-integrity/index.html):
In the midst of self-exploration, the high demands of a challenging academic environment can create situations where some students have difficulty exercising good judgment. Academic challenges can provide many opportunities for high standards to evolve if students actively reflect on these challenges and if the community supports discussions to aid in this process. It is the responsibility of the entire community to establish and maintain the integrity of our university.
This site is offered as a comprehensive and accessible resource compiling and organizing the multitude of information pertaining to academic integrity that is available from across the university. These pages include practical information concerning policies, protocols and best practices as well as articulations of the institutional values from which the policies and protocols grew. The Carnegie Mellon Code, while not formally an honor code, serves as the foundation of these values and frames the expectations of our community with regard to personal integrity.
This policy applies, in all respects, to this course.
Students at Carnegie Mellon, because they are members of an academic community dedicated to the achievement of excellence, are expected to meet the highest standards of personal, ethical and moral conduct possible. These standards require personal integrity, a commitment to honesty without compromise, as well as truth without equivocation and a willingness to place the good of the community above the good of the self. Obligations once undertaken must be met, commitments kept. As members of the Carnegie Mellon community, individuals are expected to uphold the standards of the community in addition to holding others accountable for said standards. It is rare that the life of a student in an academic community can be so private that it will not affect the community as a whole or that the above standards do not apply. The discovery, advancement and communication of knowledge are not possible without a commitment to these standards. Creativity cannot exist without acknowledgment of the creativity of others. New knowledge cannot be developed without credit for prior knowledge. Without the ability to trust that these principles will be observed, an academic community cannot exist. The commitment of its faculty, staff and students to these standards contributes to the high respect in which the Carnegie Mellon degree is held. Students must not destroy that respect by their failure to meet these standards. Students who cannot meet them should voluntarily withdraw from the university.
This policy applies, in all respects, to this course.
(http://www.cmu.edu/academic-integrity/cheating/index.html) states the following: According to the University Policy on Academic Integrity, cheating "occurs when a student avails her/himself of an unfair or disallowed advantage which includes but is not limited to:
- Theft of or unauthorized access to an exam, answer key or other graded work from previous course offerings.
- Use of an alternate, stand-in or proxy during an examination.
- Copying from the examination or work of another person or source.
- Submission or use of falsified data.
- Using false statements to obtain additional time or other accommodation.
- Falsification of academic credentials.”
- This policy applies, in all respects, to this course.
(http://www.cmu.edu/academic-integrity/plagiarism/index.html) states the following: According to the University Policy on Academic Integrity, plagiarism "is defined as the use of work or concepts contributed by other individuals without proper attribution or citation. Unique ideas or materials taken from another source for either written or oral use must be fully acknowledged in academic work to be graded. Examples of sources expected to be referenced include but are not limited to:
- Text, either written or spoken, quoted directly or paraphrased.
- Graphic elements.
- Passages of music, existing either as sound or as notation.
- Mathematical proofs.
- Scientific data.
- Concepts or material derived from the work, published or unpublished, of another person."
- This policy applies, in all respects, to this course.
(http://www.cmu.edu/academic-integrity/collaboration/index.html) states the following:
According to the University Policy on Academic Integrity, unauthorized assistance "refers to the use of sources of support that have not been specifically authorized in this policy statement or by the course instructor(s) in the completion of academic work to be graded. Such sources of support may include but are not limited to advice or help provided by another individual, published or unpublished written sources, and electronic sources. Examples of unauthorized assistance include but are not limited to: Collaboration on any assignment beyond the standards authorized by this policy statement and the course instructor(s). Submission of work completed or edited in whole or in part by another person. Supplying or communicating unauthorized information or materials, including graded work and answer keys from previous course offerings, in any way to another student. Use of unauthorized information or materials, including graded work and answer keys from previous course offerings. Use of unauthorized devices. Submission for credit of previously completed graded work in a second course without first obtaining permission from the instructor(s) of the second course. In the case of concurrent courses, permission to submit the same work for credit in two courses must be obtained from the instructors of both courses." This policy applies, in all respects, to this course.
(http://www.cmu.edu/academic-integrity/research/index.html) states the following:
According to the University Policy for Handling Alleged Misconduct in Research, “Carnegie Mellon University is responsible for the integrity of research conducted at the university. As a community of scholars, in which truth and integrity are fundamental, the university must establish procedures for the investigation of allegations of misconduct of research with due care to protect the rights of those accused, those making the allegations, and the university. Furthermore, federal regulations require the university to have explicit procedures for addressing incidents in which there are allegations of misconduct in research.”
The policy goes on to note that “misconduct means:
-
fabrication, falsification, plagiarism, or other serious deviation from accepted practices in proposing, carrying out, or reporting results from research;
-
material failure to comply with Federal requirements for the protection of researchers, human subjects, or the public or for ensuring the welfare of laboratory animals; or
-
failure to meet other material legal requirements governing research.”
“To be deemed misconduct for the purposes of this policy, a ‘material failure to comply with Federal requirements’ or a ‘failure to meet other material legal requirements’ must be intentional or grossly negligent.”
To become familiar with the expectations around the responsible conduct of research, please review the guidelines for Research Ethics published by the Office of Research Integrity and Compliance.
This policy applies, in all respects, to this course.
Do your best to maintain a healthy lifestyle this semester by eating well, exercising, avoiding drugs and alcohol, getting enough sleep and taking some time to relax. This will help you achieve your goals and cope with stress.
All of us benefit from support during times of struggle. You are not alone. There are many helpful resources available on campus and an important part of the college experience is learning how to ask for help. Asking for support sooner rather than later is often helpful.
If you or anyone you know experiences any academic stress, difficult life events, or feelings like anxiety or depression, we strongly encourage you to seek support. Counseling and Psychological Services (CaPS) is here to help: call 412-268-2922 and visit their website at http://www.cmu.edu/counseling/. Consider reaching out to a friend, faculty or family member you trust for help getting connected to the support that can help.
If you have questions about this or your coursework, please let me know.
The ways we are diverse are many and are critical to excellence and an inclusive community. They include but are not limited to: race, color, national origin, sex, disability, age, sexual orientation, gender identity, religion, creed, ancestry, belief, veteran status, or genetic information. We at CMU, will work to promote diversity, equity and inclusion because it is just and necessary for innovation. Therefore, while we are imperfect, we will work inside and outside of our classrooms, to increase our commitment to build and sustain a community that embraces these values.
It is the responsibility of each of us to create a safer and more inclusive environment. Bias incidents, whether intentional or unintentional in their occurrence, contribute to creating an unwelcoming environment for individuals and groups at the university. If you experience or observe unfair or hostile treatment on the basis of identity, we encourage you to speak out for justice and support in the moment and and/or share your experience anonymously using the following resources:
csdi@andrew.cmu.edu, (412) 268-2150, www.cmu.edu/student-diversity Report-It online anonymous reporting platform: www.reportit.net username: tartans password: plaid
All reports will be acknowledged, documented and a determination will be made regarding a course of action.” All experiences shared will be used to transform the campus climate.
- ABLE CMU
- Here for You
- Acute Medical Episodes: For example an accident, emergency surgery, etc: Seek a third party opinion (Diane Dawson at UHS). This offers confidential, retroactive excuses for exams/homeworks.
- Q: Why doesn't this course have more traditional lectures?
A: This course follows the innovative flipped classroom teaching pedagody. As such, students are expected to complete the readings prior to the class. Class time is primarily used for presentations and interactions with the instructor and student peers. Research has shown that such a teaching pedagogy is superior, enabling students to develop a holistic understanding of the subject matter. Moreover, since this is an advanced graduate research-focused course, its primary intention is focused on enabling students to critically think about new ideas and explore the unknown together. - Q: Why aren't there more lectures on background knowledge in data centers?
A: For a more traditional lecture-based course on data center or cloud computing, please consider taking the “Advanced Cloud Computing” course. Our course deals with state-of-the-art data center systems, with a specific goal of discussion-based education that can facilitate critical thinking when building/designing large-scale data center systems. - Q: May I submit a deliverable a couple of days late?
A: The short answer is yes. The goal of this course is to help you learn about new topics in data center computing in a relaxed environment that promotes learning. If you are in a stressful situation and need an extra day to submit a review etc., please let the instructor know about your situation; we are happy to accommodate your needs. - Q: May I form a project group of two or four?
A: Unfortunately no. We will have project groups of three only. - Q: Will I get feedback on my paper reviews?
A: The professor will provide feedback in three phases: initial, mid-point, and final. During each class, the professor will discuss paper reviews that stood out. Please read other students' reviews, especially those that the professor mentions in class, to better understand how to improve your own review. - Q: How do I start critiquing a paper?
A: Everything in engineering is about the "T" word --- the "Trade-off". A good way to start is by analyzing the trade-off the paper makes. For example, to improve performance, what did the paper compromise? Power? Security? How bad can this be? - Q: How long should my paper review be?
A: We judge the review on its quality rather than the length. The expectation is to make a couple of solid points each for strengths, weaknesses, comments, and questions. Your answer to the open-ended final question in your review will constitute the bulk of your paper review score. - Q: How do I understand and remember what I learned from each paper in terms of technical concepts?
A: The goal of this course is not to make you remember concepts. After reading the assigned papers, even if you forget a concept (e.g., a Remote Procedure Call) at a later time, such concepts can easily be remembered by simply searching for them on the web. Rather, the goal of this course is to help you develop critical thinking skills, which is not something that you can get by performing a Google search. In this way, this course will set you up for a successful career in computer systems. - Q: I feel anxious about speaking up in class. How can I still get a good participation score?
A: Please speak to the instructor during office hours. We are happy to accommodate your specific needs. If you are uncomfortable speaking up during class, please make an effort to participate in the small group discussions. Even if you do not know much about what your small group is discussing, please ask clarification questions to stay involved in the discussion - such particiption efforts will count towards your grade. - Q: I am finding it hard to answer the questions posed during the small group discussions. How do I still participate in these discussions?
A: The small group discussion questions are intentionally a mix of easy as well as hard questions (each class is different) to accommodate the needs of diverse students. If you are finding a particular question difficult to think about, please rely on and learn from others in your group. Additionally, you can also ask the instructor to provide an example answer for the discussion question (when appropriate) to help you get started. - Q: Would it be possible to not present a paper I was assigned?
A: Unfortunately not. All paper presentation assignments are final. If you have an extenuating circumstance and have a classmate who is willing to swap a paper presentation with you, we are happy to accommodate this. - Q: I'm not sure how to pick my project topic.
A: Please look at the list of potential projects that will be shared with you via Slack. Then, attend office hours to discuss your interests with the instructor. We will work together to find the project that is the right fit for you. - Q: My teammates are not contributing to the project as much as I am. What do I do?
A: Please bring this up with the instructor during OH. We will work together to break the project into three pieces and make sure each person is assigned a separate component. - Q: I'm already working on a related research project as a part of my PhD. Could I use this for my course project?
A: Please send the instructor a Slack message describing your PhD research and why it is related. You should also clearly state the research goals that you will achieve for this project - this work cannot already be completed. You cannot work on this project alone; you will still need two other project teammates. - Q: This is my first time working on an open-ended research project and I am struggling on some aspects such as infrastructure setup.
A: Infrastructure setup is a part of the learning experience in this project. Please reach out to the instructor if you’re having trouble - please make sure you contact the instructor well in advance and not simply before deadlines. As mentioned before, you should be working on your project every single week, rather than simply before milestone deadlines. Responses can be slow before deadlines. - Q: Lectures do not always tie in perfectly for the semester-long project.
A: While the course is different from the traditional structure of lectures followed by homeworks on the concepts taught, all the meta concepts (e.g., tradeoffs) and critical thinking from the lectures will apply in the project. Moreover, your project will likely be related to at least one of the papers discussed in class. The goal of the project is to give you a taste of open-ended research in a controlled environment, for you to make a decision on whether you are inclined towards having a research-focused career. - Q: I have some thoughts on how to improve the course. How can I suggest feedback?
A: We value every student's thoughts/feedback tremendously and are happy to change aspects of the course to make it work well for you. Please contact the instructor during OH or via Slack and we can work to achieve what you need. - Q: I'm looking for a data center computing-related internship or full-time position. Could you help me with my application?
A: Certainly! Please send your CV/resume to the instructor and they can forward it along to relevant industry contacts. - Q: Will I know where I stand in terms of my grade after each paper review?
A: You will be notified about how your grade is progressing (and why) during mid-point feedback and a couple of weeks before the final presentations. - Q: Could we have more software topics rather than hardware topics?
A: This course will balance both hardware and software system design. The first half touches on hardware systems and the second half focuses on software systems. In today's era of computing, a systems engineer can excel only when they clearly understand hardware-software interactions.