This preview shows page 1. Sign up to view the full content.
Unformatted text preview: ECS 165B: Database System Implementa6on UC Davis, Spring 2010 Acknowledgements: design of course project for this class borrowed from CS 346 @ Stanford's RedBase project, developed by Jennifer Widom, and used with permission. Slides based on earlier ones by Raghu Ramakrishnan, Johannes Gehrke, Jennifer Widom, Bertram Ludaescher, and Michael Gertz. Welcome to ECS 165B! Agenda for today's class: Logis6cs and course overview Introduc6on to the DavisDB project Technical material: pages, files, buffers, records (Chapter 9 of textbook) Course Logis6cs Instructor: Prof. Todd J. Green (email@example.com) Office hours: Tuesdays, 11:00-11:50am, 3055 Kemper Hall Teaching assistant: Mingmin Chen (firstname.lastname@example.org) Office hours: Wednesdays, 11:00-11:50am, 053 Kemper Hall Mee4ng 4mes: MWF 4:10-5:00pm (1 Wellman Hall) Discussion sec6on Fridays 11:00-11:50am (1 Wellman Hall) More Logis6cs Course webpage: hcp://www.cs.ucdavis.edu/~green/courses/ecs165b Class mailing list: email@example.com Anyone in the class can post! Don't be shy! Textbook (op4onal): Database Management Systems, 3rd Edi4on, Ramakrishnan and Gehrke, McGraw Hill, 2003 What's This Course About? ECS 165A (last quarter): how to use a DBMS ECS 165B (this quarter): how to build a DBMS What's This Course About? Primary focus (new this year!): quarter-long implementa6on project - You will build major components of a (simplified) rela6onal database system, DavisDB, in C++ - In teams of 2, delivered in 5 stages Secondary focus: a sampler of further topics in databases - XML and semistructured data, data warehousing, ... - A taste of database theory Meta-focus: large-scale solware engineering (debugging, revision control systems, best coding prac6ces, ...) How Will This Course Be Graded? Basic formula: project 80% (in 5 parts), closed-book quizzes 20% (2 of them) No midterm, no final... ...but this will be a difficult, 6me-consuming class! Code graded for correctness, efficiency, and style Extra credit for winners of DavisDB I/O efficiency contest, as well as the DavisDB code beauty contest Should I Take This Class? Pre-requisites: DBMS fundamentals (ECS 165A) C/C++ programming and data structures (ECS 60) Ability to work independently Time, ingenuity, and a sense of humor (This class is a beta- version!) What you'll get out of the class: Deeper and broader knowledge of DBMS Solware engineering experience that will pay off once you enter the real world Images of the CSIF lab's soul-crushing mountain scenery posters forever burned into your re6nas Forming Teams The project will be done in teams of 2.* Choose your partner carefully! Your grades for the project will be iden4cal. It's up to you to figure out how to share the work and get along. No marriage counseling provided.** Send an email to the TA by Wednesday with your preference for a project partner (or "no preference" if you have none). *If you prefer to work alone, you may do so, but you will s6ll be responsible for the same work as the teams, and no special allowance will be made in grading. **Divorces may be granted on a case-by-case basis. Some Project Logis6cs Team members will coordinate their efforts, and submit their code, via subversion (a standard revision control system) A short (1-2 page), high-level writeup will be part of the submiced work Standard plaLorm: the CSIF Linux machines Automated tes6ng for correctness (~80% of score), manual grading of writeup, design, and code style (~20% of score) We'll emphasize fundamental skills, such as the proper use of a debugger. (prins won't cut it in this class, just as it doesn't in the real world.) More on the logis6cs next 6me... Review: BArchitecture Architecture asic DBMS (Fig. 1.3, p. 20) Basic Database
Web Forms Application Front Ends SQL Interface SQL Commands DBMS Plan Executor Operator Evaluator Parser Optimizer Query Evaluation Engine% Transaction Manager Lock Manager Concurrency Control% File and Access Methods Recovery Manager Buffer Manager Disk Space Manager Index Files Data Files
()*+!,'-%./0!$% System Catalog
!'% Basic Database Architecture (Fig. 1.3, p. 20)
Web Forms Application Front Ends SQL Interface DavisDB Architecture (What's Lel Out) SQL Commands (select-project-join SQL fragment!) Parser Optimizer DBMS Plan Executor Operator Evaluator (simplified!) Query Evaluation Engine% Transaction Manager Lock Manager Concurrency Control% File and Access Methods Recovery Manager Buffer Manager Disk Space Manager Index Files Data Files System Catalog
!'% + user-defined extension ()*+!,'-%./0!$% Major Components of DavisDB User results commands Command Parser (given) es ueri q Query Engine (4) index scans get metadata System Manager (3) read/write/scan records indic es create files, read/ write pages Indexing (2) Record Manager (1) data, metadata Disk Space Manager (given) Buffer Manager (given) OS File System + User Extension (5) Important Dates Project due dates, subject to change: Part 1 (record manager): 4/11 Part 2 (indexing): 4/25 Part 3 (system manager): 5/2 Part 4 (query engine): 5/23 Part 5 (user extension): proposal due 5/16, code due 6/6 Quizzes: Quiz #1: 5/5 Quiz #2: 6/2 Also: Mid-quarter course review: 5/7 File and Buffer Management Review File and Buffer Management Review User results commands Command Parser (given) es ueri q Query Engine (4) index scans get metadata System Manager (3) read/write/scan records indic es create files, read/ write pages Indexing (2) Record Manager (1) data, metadata Disk Space Manager (given) Buffer Manager (given) OS File System + User Extension (5) Disks and Files (Tradi6onal) DBMS stores informa6on on hard disks This has major implica6ons for DBMS design! READ: transfer data from disk to memory (RAM) WRITE: transfer data from RAM to disk Both are high-cost opera6ons, rela6ve to in-memory opera6ons, so must be planned carefully! DavisDB I/O efficiency contest: minimize total READS and WRITES Why Not Store Everything in Main Memory? Tradi6onal arguments: It costs too much. In 1995, $1000 would buy you either 128MB of RAM or 7.5GB of disk. Main memory is vola4le. We want data to be saved between runs. (Obviously!) Tradi6onal storage hierarchy: Main memory (RAM) for currently-used data Disk for the main database (secondary storage) Tapes for archiving older versions of the data (ter6ary storage) DavisDB follows tradi6onal model (minus the tapes Discussion: do the tradi6onal arguments s6ll hold water? Disks and Paged Files Secondary storage device of choice Main advantage over tapes: random access versus sequen4al Data on hard disks is stored and retrieved in units called disk blocks or (as we'll term them in DavisDB) pages Unlike RAM, 6me to retrieve a disk page varies depending upon loca6on on disk... ...therefore, rela6ve placement of pages on disk has major impact on DBMS performance! For simplicity, we'll overlook this in DavisDB File is organized as a sequence of pages Buffer Management Main memory is limited Pages of disk files move in/out of in-memory buffer pool DavisDB # pages in buffer pool: 40 Total buffer size (@4K pages): 160K -- 6ny! ...
View Full Document
This note was uploaded on 04/29/2010 for the course ECS 152 taught by Professor Mr. during the Spring '10 term at University of Great Falls.
- Spring '10