Course Overview

CSCI 8795 - Advanced Cloud Computing course is to introduce the underlying technologies that created the current cloud computing and infrastructure, then discuss the future of clouds by investigating the cutting-edge research that will soon be part of cloud ecosystems. This course is designed for "graduate students" and is (most importantly) focused on "research project and seminar". We will quickly cover fundamental concepts of cloud computing with lectures, then move onto (more) research-oriented components (seminar and project) of this course.

This semester's theme is "performance measurement and management of microservices and cloud functions"

For more information, please refer to the course syllabus: Syllabus


Grading

# Grading Item Percent
1 Class participation/Quiz 10%
2 Programming assignments 10%
4 Reading assignments 10%
5 Paper presentation 20%
6 Research Project
  • Project Idea: 5%
  • Proposal Doc (PPT): 5%
  • Proposal Presentation:5%
  • Mid-Phase Report & Meeting: 5%
  • Final Presentation: 10%
  • Final Report: 20%
50%


Reading Assignment (Paper Review)

All students should read each assigned paper and submit the review before the scheduled lecture (9 a.m. of the day). Please note that the reading assignment will be started in the second week of the semester. The paper review should have three paragraphs:

Paragraph #1 1-2 sentences of problem statement and summary of the paper
  • e.g., Problem Statement: what problem is being addressed in the paper?
  • e.g., Summary of Approach/Evaluation: "the authors address the problem by designing and implementing ..."
Paragraph #2 pros and cons of the paper
  • e.g., Pros: what’s compelling/effective about the paper.
  • e.g., Cons: what are the most significant flaws/limitations of the paper?
Paragraph #3 (at least) three questions for the paper discussion


Paper Presentation

The presentation should be prepared for 30–40 minutes talk and clearly explain motivation, problem statement, related work (additional literature survey is a plus), the authors’ approach, evaluation, and discussion topics (e.g., pros and cons, further improvement, etc.).

Slide for the presentation must be emailed to the instructor by 9 p.m. before the class day (if not, the student will lose 33% of the grade). Also, students can always consult with the instructor for the presentation preparation.


Programming Assignments

To be announced.


Final Research Project

An essential component of this course is a final research project. You are expected to work on a cutting-edge research problem, which is relevant to the topic of this course. You can work either individually or with other students (1 to 3 members). Project ideas have to be discussed with the instructor. Tentative schedule for this project are as follows:

Project Schedule


(Tentative) Class Schedule

Papers in the schedule are accessible anywhere on UGA campus.

Week Date Topic Presenter
1 Aug/14 Course Overview & Project Introduction Instructor
Aug/15 Introduction to Cloud Computing (I) Instructor
2 Aug/20 Introduction to Cloud Computing (II) Instructor
Aug/21 VM & Data Centers: Virtualization Instructor
Aug/22 VM & Data Centers: Data Center Fundamentals Instructor
3 Aug/27 VM & Data Centers: Migration Instructor
Aug/28 Containers: Background and Resource Isolation Instructor
Aug/29 Containers: Docker
Team Building Deadline: Aug/30 5 p.m.
Instructor
4 Sep/3 Microservices Instructor
Sep/4 Containers: Orchestration Instructor
Sep/5 Serveless and Cloud Function Instructor
5 Sep/10 (ACM TIT 16) Patterns in the Chaos: A Study of Performance Variation and Predictability in Public IaaS Clouds Pranav Pandey
Sep/11 (SoCC 15) Response Time Service Level Agreements for Cloud-Hosted Web Applications Atharva Joglekar
Sep/12 (IMC 17) Complexity vs. Performance: Empirical Analysis of Machine Learning as a Service
Project Idea Deadline: Sep/15 5 p.m.
Rohit Singh
6 Sep/17 Project Discussion -- Individual Team Meeting
Sep/18 Project Discussion -- Individual Team Meeting
Sep/19 Project Discussion -- Individual Team Meeting
Proposal Doc Deadline: Sep/23 11:00 p.m.
7 Sep/24 Proposal Presentation
Sep/25 Proposal Presentation
Sep/26 Proposal Presentation
8 Oct/1 (FSE 19) A Statistics-based Performance Testing Methodology for Cloud Applications Srujana Malisetti
Oct/2 (ICSOC 18) Microscope: Pinpoint Performance Issues with Causal Graphs in Micro-service Environments Atharva Joglekar
Oct/3 (ASPLOS 19) Seer: Leveraging Big Data to Navigate the Complexity of Performance Debugging in Cloud Microservices Ting Jiang
9 Oct/8 (Eurosys 19) GrandSLAm: Guaranteeing SLAs for Jobs in Microservices Execution Frameworks Akanksha Raina
Oct/9 (SoCC 18) Overload Control for Scaling WeChat Microservices Prajay Shetty
Oct/10 (arXiv 19) Performance Modeling of Microservice Platforms Considering the Dynamics of the underlying Cloud Infrastructure Abhilash Dorle
10 Oct/15 (IoTDI 19) An Execution Model for Serverless Functions at the Edge Prajay Shetty
Oct/16 (ICPE 19) Performance Modeling for Cloud Microservice Applications Akanksha Raina
Oct/17 (SoCC 2018) Stratus: Cost-Aware Container Scheduling in the Public Cloud
Mid-Phase Report Deadline: Oct/21 11:59 p.m.
Chinmay Parab
11 Oct/22 Mid-Phase Discussion -- Individual Team Meeting
Oct/23 (HPCA 18) Applied Machine Learning at Facebook: A Datacenter Infrastructure Perspective Rohit Singh
Oct/24 Mid-Phase Discussion -- Individual Team Meeting
12 Oct/29 (Eurosys 18) Service Fabric: A Distributed Platform for Building Microservices in the Cloud Ayushi Shukla
Oct/30 (HotCloud 19) Agile Cold Starts for Scalable Serverless Chinmay Parab
Oct/31 (SoCC 18) The Elasticity and Plasticity in Semi-Containerized Co-locating Cloud Workload: a View from Alibaba Trace Ting Jiang
13 Nov/5 (ATC 18) Putting the "Micro" Back in Microservice Ayushi Shukla
Nov/6 (CLOUD 19) Spock: Exploiting Serverless Functions for SLO and Cost aware Resource Procurement in Public Cloud Mehir Pandit
Nov/7 (ATC 18) Peeking Behind the Curtains of Serverless Platforms Srujana Malisetti
14 Nov/12 (NSDI 19) Shuffling, Fast and Slow: Scalable Analytics on Serverless Infrastructure Abhilash Dorle
Nov/13 (HotCloud 19) DynaShield: Reducing the Cost of DDoS Defense using Cloud Services Mehir Pandit
Nov/14 (HotEdge 19) Towards a Serverless Platform for Edge AI Pranav Pandey
15 Nov/19 Open Discussion for Final Project
Nov/20 Final Project Presentation
Nov/21 Final Project Presentation
16 Nov/26 Final Project Presentation
17 Dec/03 Last Day
Final Report Deadline: Dec/01 5 p.m.