UCSCBaskin School of Engineering  
General Information Events, News & Organizations Degrees & Departments Research Classes Admissions & Advising People & Jobs Administration
CMPS 229 - Spring 2003
April 1

CS 229: Advanced Topics in Storage Systems

Course Description

This course is a graduate level study of the issues in the design and implementation of storage systems. The readings are taken from the current research literature and articles of historical significance.

The field of storage systems is one of the fastest growing and most interesting research areas in computer science. The storage system is at the core of most computer systems, and its performance often dominates the entire system. It is also responsible for the safe-keeping of the most valuable assets of an organization – its data. The course will cover the design and implementation of storage systems and the architecture and characteristics of the components on which storage systems are built. Topics will range from the device level up to distributed systems concepts.

In terms of preparation, you are expected to have basic operating system knowledge, such as presented in a standard undergraduate course such as CS 111. You are expected to have taken CS 221, Advanced Operating Systems. If you have do not have this level of preparation, you should discuss it with me. You may be admitted if you can demonstrate the knowledge and sophistication necessary for successful completion of the course.

Instructor

Prof. Darrell Long

247 Baskin Engineering
darrell@cs.ucsc.edu
831-459-2616

 

Office hours yet to be arranged.

Course Requirements

Reading will be assigned before each class (a tentative schedule appears below). Usually this will consist of two articles from the research literature. You must read these articles carefully and prepare a short summary (including a few questions or insightful comments about the material).

These summaries must be prepared before the class where the articles are discussed. A suitable summary of an article consists of brief answers to the following seven questions:

  1. What is the problem the authors are trying to solve?
  2. What other approaches or solutions existed at the time that this work was done?
  3. What was wrong with the other approaches or solutions?
  4. What is the authors’ approach or solution?
  5. Why is it better than the other approaches or solutions?
  6. How does it perform?
  7. Why is this work important?

The summaries will be used in the discussion during class. You must turn in your summaries at the end of each class, and at the end of the quarter you will be expected to turn in a portfolio containing all of the summaries that you prepared during the quarter.

Each student will give a presentation of at least two articles during the quarter (the actual number depends on the number of students enrolled in the class). Presentations must cover the essential ideas in the article including the goals of the research, related research, essential details of the research, key results, and the significance of those results. The presenter must also be prepared to lead the class discussion of the article.

Students will be required to write a report on a topic in the area of storage systems. This report should be the results of a project, original research (strongly preferred), or a strong survey of prior art. Reporting work done for another course is completely unacceptable. You must choose a topic by second week of the quarter. Each student will give a final presentation on their project during a poster session at the end of the quarter. Faculty and students will be invited to the poster session, and evaluation will be done by the faculty of the Storage Systems Research Center (Profs. Long, Brandt and Miller).

Your grade in the course is based 25% on preparedness and class participation, 25% for presentations, and 50% for your term project and report.

Attendance

Class attendance is mandatory. Since this is a graduate class, students are expected participate actively in class, and you can’t do that if you’re not actually there. I won’t take attendance at class, but you will not pass if you miss too many classes. If you need to miss a class for a good reason, such as a conference or other research-oriented commitment, please notify me in advance if possible.

You must present the articles assigned to you and lead the discussion. If unforeseen circumstances force you to change the date of presentation, you must make arrangements to trade with another student, and notify me well in advance so the reading schedule can be changed.

Tentative Schedule

 

Date

Presenter

Topic

Comments

April 1

Long

Introduction

FAST 02 Video

April 3

Long

·        The UNIX Time Sharing System

·        A Fast File System for UNIX

·        The Design and Implementation of a Log-Structured File System

Historical File Systems

April 8

Kroeger

·        Design and Implementation of a Predictive File Prefetching Algorithm

·        The Case for Efficient File Access Pattern Modeling

Predictive Prefetching

April 10

Golding

·        Highly Concurrent Shared Storage

·        Fault-tolerant Replication Management in Large-scale Distributed Storage Systems

Palladio

April 15

Palmer

·        An Introduction to Disk Drive Modeling

Industrial Tutorial

April 17

Kallahalla

·        Plutus: Scalable Secure File Sharing on Untrusted Storage

Industrial Presentation

April 22

Tarnopolsky

·        Parametric Optimization for Terabit Perpendicular Recording

·        Theoretical Analysis of Longitudinal and Perpendicular Recording Potential

Industrial Tutorial:

High Areal Density Recording

April 24

Miller

·        Strong Security for Network-Attached Storage

·        Fast and Secure Distributed Read-only File System

Secure File Systems

April 29

Wilcke

·        Collective Intelligent Bricks

·        Percolation in Dense Storage Arrays

Industrial Presentation

May 1

Qin Xin

Qin Xin

·        ARC: A Self-tuning, Low Overhead Replacement Cache

·        Group-Based Management of Distributed File Caches

Caching

May 6

Bo Hong

Jie

·        Zone Scheduling for MEMS-based Storage Devices

·        Using MEMS-based Storage in Disk Arrays

Alternative Storage Devices

May 8

Karamanolis

·        Taming Aggressive Replication in the Pangaea Wide-area File System

Industrial Presenatation

May 13

Ge Yi

Teng Xu

·        A Low-bandwidth Network File System

·        Venti: a New Approach to Archival Storage

Redundancy Reduction

May 15

Edel

Ge Yi

·        Efficient Distributed Back-up with Delta Compression

·        In-Place Reconstruction of Delta Compressed Files

Delta Compression

May 20

Joel Wu

 

Garcia

·        Optimizing File Availability in a Secure Severless Distributed File System

·        Disk Array Storage System Reliability

Reliability

May 22

Peter Tsui

 

Joel Wu

·        GPFS: A Shared-disk File Systems for Large Computing Clusters

·        Data Sieving and Collective I/O in ROMIO

Parallel I/O

May 27

Meyer

Teng Xu

·        Measurements of a Distributed File System

·        A Comparison of File System Workloads

Workload Analysis

May 29

Edel

 

Bisson

·        Track-aligned Extents: Matching Access Patterns to Disk Drive Characteristics

·        Freeblock Scheduling Outside of Disk Firmware

Scheduling

June 3

Meyer

Bisson

·        The Evolution of Coda

·        Adaptive Disk Spin-Down for Mobile Computers

Mobile File Systems

June 5

Peter Tsui

Garcia

·        Ivy: A Read/Write Peer-to-peer File System

·        Making the Most out of Direct-Access Network Attached Storage

Distributed File Systems

 

 

Poster Session

Presentation of Final Projects

 


General info · News · Events · Degree Programs · Research · Classes · Admissions · Advising · People · Jobs · Administration
SOE Webmail · SOE SSH · SOE Wiki · Search · Sitemap · Contact us · Driving directions · Privacy · UCSC
© Baskin School of Engineering, University of California, Santa Cruz
1156 High St., Santa Cruz, CA 95064 · (831) 459-2158 · webmaster@soe.ucsc.edu