Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 1Database Management SystemsSession 1 (CSCI-585)Instructor: Cyrus [email protected] Management Systems 3ed, R. Ramakrishnan and J. Gehrke 2What Is a DBMS? A very large, integrated collection of data. Models real-world enterprise. Entities (e.g., students, courses) Relationships (e.g., Madonna is taking CS564) A Database Management System (DBMS) is a software package designed to store and manage databases.Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 3Files vs. DBMS Application must stage large datasets between main memory and secondary storage (e.g., buffering, page-oriented access, 32-bit addressing, etc.) Special code for different queries Must protect data from inconsistency due to multiple concurrent users Crash recovery Security and access controlDatabase Management Systems 3ed, R. Ramakrishnan and J. Gehrke 4Why Use a DBMS? Data independence and efficient access. Reduced application development time. Data integrity and security. Uniform data administration. Concurrent access, recovery from crashes.Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 5Why Study Databases?? Shift from computation to information at the “low end”: scramble to webspace (a mess!) at the “high end”: scientific applications Datasets increasing in diversity and volume.  Digital libraries, interactive video, Human Genome project, EOS project  ... need for DBMS exploding DBMS encompasses most of CS OS, languages, theory, AI, multimedia, logic?Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 6Data Models A data model is a collection of concepts for describing data. A schema is a description of a particular collection of data, using the a given data model. The relational model of data is the most widely used model today. Main concept: relation, basically a table with rows and columns. Every relation has a schema, which describes the columns, or fields.Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 7Levels of Abstraction Many views, single conceptual (logical) schemaand physical schema. Views describe how users see the data.  Conceptual schema defines logical structure Physical schema describes the files and indexes used.* Schemas are defined using DDL; data is modified/queried using DML.Physical SchemaConceptual SchemaView 1 View 2 View 3Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 8Example: University Database Conceptual schema:  Students(sid: string, name: string, login: string, age: integer, gpa:real) Courses(cid: string, cname:string, credits:integer)  Enrolled(sid:string, cid:string, grade:string) Physical schema: Relations stored as unordered files.  Index on first column of Students. External Schema (View):  Course_info(cid:string,enrollment:integer)Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 9Data Independence * Applications insulated from how data is structured and stored. Logical data independence: Protection from changes in logical structure of data. Physical data independence: Protection from changes in physical structure of data.* One of the most important benefits of using a DBMS!Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 10Concurrency Control Concurrent execution of user programs is essential for good DBMS performance. Because disk accesses are frequent, and relatively slow, it is important to keep the cpu humming by working on several user programs concurrently. Interleaving actions of different user programs can lead to inconsistency: e.g., check is cleared while account balance is being computed. DBMS ensures such problems don’t arise: users can pretend they are using a single-user system.Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 11Transaction: An Execution of a DB Program Key concept is transaction, which is an atomicsequence of database actions (reads/writes). Each transaction, executed completely, must leave the DB in a consistent stateif DB is consistent when the transaction begins. Users can specify some simple integrity constraints on the data, and the DBMS will enforce these constraints. Beyond this, the DBMS does not really understand the semantics of the data. (e.g., it does not understand how the interest on a bank account is computed). Thus, ensuring that a transaction (run alone) preserves consistency is ultimately the user’s responsibility!Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 12Scheduling Concurrent Transactions DBMS ensures that execution of {T1, ... , Tn} is equivalent to some serialexecution T1’ ... Tn’. Before reading/writing an object, a transaction requests a lock on the object, and waits till the DBMS gives it the lock. All locks are released at the end of the transaction. (Strict 2PLlocking protocol.) Idea: If an action of Ti (say, writing X) affects Tj (which perhaps reads X), one of them, say Ti, will obtain the lock on X first and Tj is forced to wait until Ti completes; this effectively orders the transactions. What if Tj already has a lock on Y and Ti later requests a lock on Y? (Deadlock!) Ti or Tj is aborted and restarted!Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 13Ensuring Atomicity DBMS ensures atomicity (all-or-nothing property) even if system crashes in the middle of a Xact. Idea: Keep a log (history) of all actions carried out by the DBMS while executing a set of Xacts: Before a change is made to the database, the corresponding log entry is forced to a safe location. (WAL protocol; OS support for this is often inadequate.) After a crash, the effects of partially executed transactions are undone using the log. (Thanks to WAL, if log entry wasn’t saved before the crash, corresponding change was not applied to database!)Database Management Systems 3ed, R. Ramakrishnan and J. Gehrke 14The Log The following actions are recorded in the log: Ti writes an object: The old value and the new value.• Log record must go to disk before the changed page! Ti commits/aborts: A log record indicating this action. Log records chained together by Xact id, so it’s easy to undo a specific Xact (e.g., to resolve a