anagement Systems Chapter 11 Distributed Databases Jerry Post Copyright © 2001Distributed Databases ~> Definition «> Advantages / Uses <> Problems / Complications “> Client-Server / SQL Server <> Microsoft Access SELECT Sales FROM Britain.Sales UNION ‘SELECT Sales FROM France.Sales UNION SELECT Sales FROM lItaly.Sales Germany FranceDistributed Database Definition <> Multiple independent databases + Each DBMS is a complete DBMS (engine, queries, locking, transactinos, etc.) + Usually on different machines. + Usually in different locations. ~ Connected by a network. «> Might be different environments + Hardware + Operating System + DBMS Software Database Zeus England as “here United StatesDistributed Database Rules > C.J. Date + Distributed query processing. ~ Rule 0: Transparency: the + Distributed transaction user should not know or care management. that the database is distributed. + Hardware independence. + Local autonomy. + Operating system independence. + No reliance on a central site. + Network independence. + Continuous operation. + DBMS independence. + Location independence. + Fragmentation independence (physical storage). + Replication independence.Distributed Features «> Each database can continue to run even if portion fails. «> Data and hardware can be moved without affecting operations or users. + Expanding operations. + Performance issues. <> System expansion and upgrades. + Add new section without affecting others. + Upgrade hardware, network and DBMS.Advantages and Applications ,,.. «> Business operations are often distributed + Work and data are segmented by department. + Work and data are segmented by geographical location. <> Improved performance + Most updates and queries are performed locally. + Maintain local control and responsibility over data. ~» Can still combine data across the system. 1 “> Scalability and expansion + Add on, not replacement. transactions ’ future Nexpansion = 6Creating a Distributed Database ~ Design administration plan. ~ Choose hardware and DBMS vendor, and network. ~ Set up network and DBMS connections. ~ Choose locations for data. ~ Choose replication strategy. ~> Create backup plan and strategy. ~ Create local views and synonyms. ~ Perform stress test: loads and failures.Distributed Query Processing «> Networks are slow + Drives: 10 - 20 MB per sec. + LANs: 1 - 10 MB per sec. + WANs: 0.01 - 5 MB per sec. + Faster is possible but expensive! «> Goal is to minimize transmissions. + Each system must be capable of evaluating queries--preferably SQL. + Results depend heavily on how the system joins tables. Disk driveDistributed Query Processing ~ Example + NY: Customers: 1 M rows a Customers(C#, ...) + LA: Production: 10 M rows 1,000,000 + Chicago: Sales: 20 M rows C# list from + Query: List customers who ino aie: 6 bought blue products on 1-Mar-01 Sales(SH OF, Sdate) | Customer + Bad idea #1 20,000,000 data 4 Transfer all rows to Chicago Saleltem(S#, Pit...) 4 Then JOIN and select. SUBD 000 + Better idea #2 (probably) P# sold on 4 Transfer blue products from LA 1-Mar-01 to Chicago Blue P# + Better idea #3 sold on 1-Mar-01 4 Get sale items on 1-Mar-01 4 Get blue products from LA _ | Products(P#, Color...) 4 Send C# to NY 2,000,000Data Replication ~> Goals + Minimize transmissions + Improve performance + Support heavy multiuser access. «> Problems + Updating copies 4 Bulk transmissions 4 Site unavailable + Concurrency 4 Easier for two people to change the same data at the same time. ~ Decision support systems. ~ Data warehouse. France: Customers| & Sales Spain: Customers & Sales Britain: Customers & Sales France: Customers) & Sales Spain: Customers. & Sales Market research & data corrections. Periodic updates 2 a Update data.Concurrency and Locks <> Each DBMS must maintain lock facility. «> To update, each DBMS must utilize and recognize other lock mechanisms and return codes. «> Each DBMS must have a deadlock resolution protocol that recognizes the distributed databases. + Random wait. DBMS #1 Accounts 8898 + Optimistic updates. + Two-phase commit. Jones Transadtion B Waiting 3561 ‘LockedTransactions & Two-Phase Commit <> Two (or more) separate lock managers. «> DBMS initiating update serves as the coordinator. <> Two phases + Coordinator sends message and data to all machines to “get ready.” + Local machines save data in logs, verify update status and return message. + Ifall locals report OK, then coordinator writes log and instructs others to proceed. If any fail, it sends Rollback message. Database 1 Initiate Transaction 1, Prepare to commit. 2. Commit Database 2 Lock tables. Save log." Update all tables!Distributed Design Questions juestion Concurrent Replication What level of data consistency is needed? High Low = Medium ‘How expensive is storage? Medium-High | Low +4 What are the shared access requirements? Global Local How often are the tables updated? Often Seldom ‘Required speed of updates (transactions)? Fast. Slow =| How important are predictable transaction times? _| Hig Low DBMS support for concurrency and locking? ‘Good — Excellent_| Poor Can shared access be avoided? No YesDistributed Databases In Oracle ~ Database Links Schema.Table@Location + Full database names. [email protected] + CONNECT command. Server 2 ~ Linking through synonyms. database \ + CREATE SYNONYM ... ae 1 + Central control over permissions. y' Procedure: «> Linking through Views/queries. Employee + CREATE VIEWAS ... WHERE ... + Can assign local permissions. / «> Linking through stored procedures. + DELETE ... View MJ User can only + Strong control over actions. Ji ; ~7 run procedure. user No other access. permissionsClient-Server Server Clients Clients Shared Database Front-end User InterfaceLAN File Server File Server «> Not a distributed database. + Data file stored on server. + Server is passive, appears as giant disk drive to PC. + PC processes all data. All data from all tables are + Retrieves all needed data | read by PC, which performs across the network. JOIN and WHERE test. If : available, reads index first. <> Performance improvements. + Indexes are crucial. ! + Store some data on each a PC (replication). SELECT Name, SaleDate + Store applications on PC _| FROM Customer INNER JOIN Sales (graphics & forms). ON Customer.C# = Sales.C# WHERE SaleDate BETWEEN #1-Mar-97# + Convert to SQL-Server AND #9-Mar-97#; 16LAN File Server: Slow File Server MyFile.mdb CustID Name ... 15 Jenkins... Forms||425 Juarez ... Application and query DBMS = software One row at a time transferred. transferred, until SELECT™ all rows are examined. FROM Customer WHEREClient-Server Databases File Server “> One machine machine is dominant (server) and handles data for many clients. ~» Client machines handle front-end tasks and small data tables that are not shared. Send SQL statement. | applicationMicrosoft Access «> Access is a client-side database. + In file server environment. + Asaclient to a database server. ~> Attach or Link to other databases. + Link to an Access database is file server. + Link through ODBC is database server. ~» ODBC: Open DataBase Connectivity «> Pass-Through QueriesOpen Database Connectivity: ODBC Server Computer ~ Microsoft connectivity standard. poe + Most DBMS companies provide drivers. i + SQL Server, Oracle, Ingres, ODBC driver etc. i —- z + Driver is installed on PC and 2 > Server. & & ~ ODBC handles: i + Login to database. ODBC driver + Send query. | Link Tables + Interpret result codes. - + Exchange data. Access Client ComputerODBC Basics ~ Set the connection string. + Datatype: odbc + DSN: data source name holds linkage data--built with Windows Control Panel or RegisterDatabase. + uid: login Username + pwd: login Password + database: full database name on remote system. 21ADO and Direct Connections Server Computer The Database vendor peas provides its own data : transport (e.g,. Oracle or SQL Server) installed on \ + Poems transport the server and the client. ee oO 4 ADO provides a driver that w@ 3 connects your application to a & the transport services. DBMS transport ODBC can serve as the data transport if nothing ADO else is available vistaEasig application Client Computer 22Three-Tier Client-Server ; ~> Server Databases ~ Client front-end «> Middle + Locate databases + Business rules + Program code Legacy applications. Database links. Business rules. Program code. Front-end. y) Database Middleware Application. 2p User Interface. Client S 25The Internet as Client-Server information Internet Router lent) (oes Browser Web Server http://server.location/page Pintpaces Forms GraphicsHTML Limited Clients

a

¢ Cat Dog Fish » Category

Color

atASP Code Structure Connect to the Database ® Create the SQL Connect to the Database/Recordset Loop through the Recordset Note: Access will create Get a Field the basic structure. Display it Export a query as asp. Move to the next row Then edit the file. End Loop Set objConn = CreateObject("ADODB.Connection") Set rst = CreateObject("ADODB. Recordset”) sql =" SELECT..." tst.Open sql ... tst.MoveFirst do while Not rs.eof Response.Write Server.HTMLEncode(rst("LastName")) rst.MoveNext Loop 32ASP Query: Build SQL <% Set objConn = Server.CreateObject("ADODB.Connection") objConn.open "PetStore",” sql = "SELECT AnimallD, Name, Category, Breed, DateBorn, Color, ListPrice " sql = sql & "FROM Animal WHERE (Category =" sql = sql & Request.Form("Category") & ") AND (Color LIKE “*" sql = sql & Request.Form("Color”) &"™)" Set rst = Server.CreateObject("ADODB. Recordset") rst-Open sql, objConn, 3, 3 33ASP Query: Create Table <% On Error Resume Next tst.MoveFirst do while Not rst.eof %> erver. HTMLEncode(rst("Name"))%>
‘erver.HTMLEncode(rst("Category"))%>
erver. HTMLEncode(rst("Breed"))%>
server. HTMLEncode(rst("Color"))%>
<% rst MoveNext

ASPSearch1

AnimallD	Name	Category	Breed	DateBorn	Color	ListPrice
<%=Server.HTMLEncode(rst("AnimallD"))%>	<%=Server. HTMLEncode(rst("DateBorn"))%>	<%=Server. HTMLEncode(rst("ListPrice"))%>

aClient-Server Data Transfer Order Form Order ID 4015 Customer | Jones, Martha y Order Date 12-Aug What if there are 10,000 customers? How much time to load the combo box? How do you refresh/reload the combo box? Alternatives?XML: Transferring Data Order: OrderlD, OrderDate DTD: Document Type Definition is hierarchical. Item: ItemID, Quantity, Cost + Repeats 1 or more Item: ItemID, Quantity, Cost g Eee Carries Item: ItemID, Quantity, Cost #PCDATA: parsed _character data Ce 36XML Data Example ‘ 1 3/6/2001 $33.54 Need immediately. 30 Flea Collar-Dog- Medium 208 $4.42 27 Aquarium Filler & Pump 8 $24.65 XML: extensible markup languageXML Example in Explorer ~ - 1 3/6/2001 $33.54 Need immediately. — 30 Flea Collar-Dog-Medium 208 $4.42 27 Aquarium Filter & Pump 8 $24.65 + +