Warehouse-scale datacenters power the online services many people use every day, including web search, social networking, online shopping, video streaming, and cloud computing. The course will conduct a vertical study of modern datacenters covering the entire system stack from hardware architecture to systems software to application programming frameworks. The course will also examine cross-cutting concerns such as total cost of ownership, service level objectives, energy efficiency, reliability and availability, and privacy and security. Through lab exercises, students get practical experience with modern datacenter technologies and reinforce the basic concepts covered in the lectures.

Lectures: Tuesday and Friday, 1:30 pm - 3:00 pm, Room FST 01 202
Recitations: , 6:00 pm - 7:00 pm, FST 01 146
Labs: Tuesday, 3:00 pm - 4:30 pm, FST 01 B103

Instructor: Haris Volos

Email: hvolos01@ucy.ac.cy
Office hours: Tuesday, Friday 12:30 - 1:30 pm or by appointment, FST 01 115

schedule

Date	Lecture	Readings	Logistics
Tue 06/09	Lecture #1 : Introduction [ slides ]	BHR Textbook, Ch1
	Lab #1: CloudLab [ slides \| notes ]
Fri 09/09	Lecture #2 : Datacenter Workloads [ slides ]	BHR Textbook, Ch2 Web search for a planet: the Google cluster architecture
Tue 13/09	Lecture #3 : Datacenter Programming: Go Language [ slides ]	Cox et al. The Go Programming Language and Environment Doxsey An introduction to programming in Go
	Lab #2: Essential Tools [ slides \| notes ]
Fri 16/09	Lecture #4 : Datacenter Basics: Server Hardware [ slides ]	BHR Textbook, Ch3.1, Ch3.2
Tue 20/09	Lecture #5 : Datacenter Basics: Server Hardware [ slides ]
	Lab #3: Go Language [ slides \| notes ]
Fri 23/09	Lecture #6 : Datacenter Basics: Power, Energy, and Cost [ slides ]	BHR Textbook, Ch4, Ch5	HW1 out
Tue 27/09	Lecture #7 : Datacenter Basics: Reliability and Availability [ slides ]	BHR Textbook, Ch6, Ch7
	Lab #4: Web Search [ slides \| notes ]
Fri 30/09	Lecture #8 : Virtualization [ slides ]	Arpaci-Dusseau Virtual Machine Monitors
Tue 04/10	Lecture #9 : Virtualization [ slides ]
	Lab #5: Go Web Apps [ slides \| notes ]
Fri 07/10	Lecture #10 : Containers [ slides ]	Understanding and hardening Linux containers	HW1 due 04/10 11:59pm HW2 out
Tue 11/10	Lecture #11 : Containers [ slides ]
	Lab #6: Docker Containers [ slides \| notes ]
Fri 14/10	Lecture #12 : Microservices [ slides ]	Introduction to microservices Lewis and Fowler Microservices Gan et al. An Open-source benchmark suite for microservices and their hardware-software implications for cloud & edge systems
Tue 18/10	Lecture #13 : Midterm Exam [ slides ]
	Lab #7: Docker Containers (continued) [ slides \| notes ]
Fri 21/10	Lecture #14 : Microservices [ slides ]
Tue 25/10	Lecture #15 : Microservices - RPC [ slides ]
	Lab #8: Microservices [ slides \| notes ]
Fri 28/10	Ohi Day - No Class
Tue 01/11	Service Day - No Class
Fri 04/11	Lecture #16 : Container Orchestration [ slides ]
Tue 08/11	Lecture #17 : Serverless Computing [ slides ]	Roberts Serverless architecture Hendrickson et al. Serverless computation with OpenLambda	HW2 due 25/10 11:59pm HW3 out
	Lab #9: Kubernetes [ slides \| notes ]
Fri 11/11	Lecture #18 : Datacenter Storage [ slides ]	BHR Textbook, Ch3.4 Ghemawat et al. The Google File System Chang et al. Bigtable: A distributed storage system for structured data DeCandia et al. Dynamo: Amazon's Highly Available Key-value Store
Tue 15/11	Lecture #19 : Datacenter Storage [ slides ]		HW3 due 15/11 11:59pm
	Lab #10: Datacenter Storage [ slides \| notes ]
Fri 18/11	Lecture #20 : Datacenter Storage [ slides ]
Tue 22/11	Lecture #21 : Memcached [ slides ]
	Lab #11: Datacenter Storage [ slides \| notes ]
Fri 25/11	Lecture #22 : Flash Storage [ slides ]	Arpaci-Dusseau Flash-based SSDs
Tue 29/11	Lecture #23 : Security and Privacy [ slides ]	Ristenpart et al. Hey, you, get off of my cloud: exploring information leakage in third-party compute clouds
Fri 02/12	Lecture #24 : Potpourri, Wrap Up [ slides ]

About

Objectives

The objective of the course is to study the key principles or/and concepts that underlie a modern data center. The course will conduct a vertical study of datacenter technology covering the entire system stack, including hardware architectures, systems software, and application programming frameworks. The course will also explore cross-cutting concerns such as total cost of ownership, service level objectives, reliability and availability, energy efficiency, and privacy and security.

Content

Topics include server design, heterogeneous hardware and accelerators, total-cost of ownership (TCO) analysis, performance analysis, data-center storage, data-center networking, cloud computing, virtualization and containers, resource management, microservices, serverless computing, side channels, trusted execution.

Target Audience

The course is aimed at senior undergraduate students and early-stage postgraduate students of computer science. The language of instruction is greek but most of the educational material (book, slides, web pages) will be in english.

Prerequisites

The consent of the instructor. Basic understanding of systems and architecture concepts through undergraduate course equivalent to CS221 (Computer Organization) and undergraduate course equivalent to CS222 (Operating Systems).

Labs

Through lab exercises, students get practical experience with modern datacenter technologies and reinforce the basic concepts covered in the lectures.

Homework Assignments

Homeworks are assignments that are designed to help students develop an in-depth understanding of both the theoretical and practical aspects of ideas presented in lecture.

Bibliography

Luiz André Barroso, Urs Hölzle, and Parthasarathy Ranganatha. The Datacenter as a Computer: Designing Warehouse-Scale Machines. Third Edition Morgan & Claypool (2019).

Grading

The evaluation will be based on homework assignments and exams. The grading rubric will be as follows:

Assignments: 40%
Midterm exam: 15%
Final exam: 45%