|Charlotte Tang||Yunxiang Chi|
|Kaiyuan Wang||Yuke Liu|
|Fengyuan Wu||Xiyan Shao|
|Remzi H. Arpaci-Dusseau and Andrea C. Arpaci-Dusseau|
Operating Systems: Three Easy Pieces
Version 1.00 (Available free online!)
This course covers the principles of operating systems. It emphasizes the basic concepts of OS kernel organization and structure, processes and threads, concurrency and synchronization, memory management, file systems, and communication. It is also a project course, providing essential experience in programming with concurrency, implementing and unmasking abstractions, working within an existing complex system, and collaborating with other students in a group effort.
The following table outlines the schedule for the course. We will update it as the quarter progresses.
The optional readings include primary sources and in-depth supplements for concepts in the class. Supplemental reading is for your own interest — the readings are not required, nor will you be tested on the material. Note that some of the links to the documents point to the ACM Digital Library. UCSD has a subscription to the ACM Digital Library, so you can access these links from a web browser on campus.
|Tu 4/4||Course Intro||CH. 1, CH. 2||The UNIX Time-Sharing System||HW 1: Out||PR 0: Out|
|Th 4/6||Interactions with Apps and Hardware||CH. 6||
Linux context switch
A Case Against (Most) Context Switches
|Tu 4/11||Processes||CH. 3, CH. 4, CH. 5||A fork() in the road||PR 0: Due
PR 1: Out
|Th 4/13||Threads||CH. 26, CH. 27||Illustrated Tales of Go Runtime Scheduler
C++ Coroutine Tutorial
|Tu 4/18||Synchronization||CH. 28, CH. 29||RCU Usage In the Linux Kernel: One Decade Later||HW 1: Due
HW 2: Out
|Th 4/20||Semaphores||CH. 31||The Structure of the 'THE'-Multiprogramming System|
|Tu 4/25||Condition Variables and Deadlock||CH. 30, CH. 32||Understanding Real-World Concurrency Bugs in Go|
|Th 4/27||CPU Scheduling||CH. 7, CH. 8||Lottery Scheduling: Flexible Proportional-Share Resource Management
Shenango: Achieving High CPU Efficiency for Latency-sensitive Datacenter Workloads
|PR 2: Out|
|Tu 5/2||Midterm Review||HW 2: Due
HW 3: Out
|PR 1: Due|
|Th 5/4||Midterm Exam|
|Tu 5/9||Memory Management Overview||CH. 15, CH. 16||A Study of Virtual Memory Usage and Implications for Large Memory|
|Th 5/11||Paging||CH. 18, CH. 19, CH. 20||Mitosis: Transparently Self-Replicating Page-Tables for Large-Memory Machines|
|Tu 5/16||TLBs, Swapping||CH. 21||Coordinated and Efficient Huge Page Management with Ingens||PR 3: Out|
|Th 5/18||Page Replacement and Memory Allocation||CH. 17, CH. 22||Learning-based Memory Allocation for C++ Server Workloads|
|Fri 5/19||n/a||PR 2: Due|
|Tu 5/23||Storage Devices and File System API||CH. 37, CH. 39||A case for redundant arrays of inexpensive disks (RAID)||HW 3: Due
HW 4: Out
|Th 5/25||File System Disk Layout||CH. 40||The Google File System|
|Tu 5/30||File Caching and Reliability||CH. 41, CH. 42||IRON File Systems|
|Th 6/1||Virtual Machines||Appendix B||Virtual Machine Monitors: Current Technology and Future Trends|
|Tu 6/6||Protection||CH. 53, CH. 55||RedLeaf: Isolation and Communication in a Safe Operating System||HW 4: Due|
|Th 6/8||Final Review|
|Sat 6/10||n/a||PR 3: Due|
|M 6/12||Final Exam, 3-6 pm|
The course is organized as a series of lectures by the instructor, discussion sections by the TAs, reading, homework, and project assignments, and exams:
The course will have four homeworks, and I will post them as the quarter progresses. You will submit all your homework electronically via Canvas. We will reduce homework grades by 20% for each day that they are late.
Due to extensive copying on homeworks in the past, we have changed how homeworks are graded. As long as you submit a technical answer related to the question, you will get full credit for the question. The goal of the homeworks is to give you practice learning the material. The homework questions both supplement and complement the material from lecture and in the project, and you will also find the homework questions to be useful for practicing for the exams. We will post solutions to all homeworks after they are submitted, and you can use them for studying as well. But, even with the solutions, the amount you learn from the homeworks will be directly correlated with your effort working on them.
I encourage you to collaborate on the homeworks: you can learn a lot from your fellow students. Collaboration consists of discussing problems with other students and independently writing your own answers to the problems based upon those discussions. As a rule of thumb, you should be able to discuss a homework problem in the hall with others, go home, and then write up your answer to the problem on your own.
Your grade for the course will be based on your performance on the homeworks, midterm exam, final exam, and projects, as follows:
The academic honesty guidelines outlined by Charles Elkan apply to this course. I urge you to resist any temptation to cheat, no matter how desperate the situation may seem. If you are in circumstances that you feel compel you to cheat, come to me first before you do so.