Data Management1 why Data Management ? We have already considered hardware and software in some In this chapter， we turn our attention to a third basic computer resource， Many computer applications require that data be stored for subsequent [1] Simply storing the data is not enough， A typical computer system， even a small one， can have dozens of disks and tapes， each holding data for dozens of different For any given application， one and only one set of data will We must be able to store， Locate， and retrieve the specific data needed by a given That is the concern of data 2 Accessing DataImagine a single diskette containing several For a particular application， only one of those programs will How is a given program selected， loaded， and executed? In Chapter 6， we learned that the operating system， responding to a use’s command， reds the disk’s index， searches it for the requested program name， extracts the program’s track and sector address， and issues primitive commands to read it into main Later， following a RUN command， the program is given control of the Accessing data presents a similar A single diskette can hold data for several different For a given application， one and only one set of data will do， and finding the right data is much like finding the right There are differences between accessing programs and accessing data， When a program is need， all its instructions must be loaded into Data， on the other hand， are typically processed selectively， a few elements at Thus， it is not enough merely to locate the data; we must be able to distinguish the individual data elements， 1 Data Structures The key to retrieving data is remembering where they are stored， If the data elements are stored according to a consistent and well understood structure， it is possible to retrieve by remembering that The simplest data structure is a For example， data for a program that computes an average might be stored as a series of numbers separated by commas (F1) The commas distinguish the individual data Most programming languages support amore complex data structure called an array (Fig 2) Each array element can hold one data Each element is assigned a unique identifying number or numbers， and individual data elements can be inserted， extracted， or manipulated by referencing those For example， in the array pictured in F 2 elements are identified by a row number and a column number， and row 1， column3 (element1，3) contains the value Once an array has been filled， it can be written to disk， tape， or any other secondary medium， and back into memory for Consider a program that generates name and address For each label， we need a name， a street address， a city， a state， and a zip If we needed only a few labels， we might store the data in a list， but separating the elements would soon become An option is to set up an array of names and addresses， with each row holding the data for a single The only problem is that the entire array must be in main memory before the individual elements can be accessed， and main memory space is Thus， even with an array， we could generate relatively few F 1 The simplest data structure is a Separators， such as these commas， serve to distinguish individual Often， a “sentinel” value， such as a negative number， marks the end of the 4410， 843， 184， 31， 905， 6357， 44， 7702， 228， 59， -1 F 2 Most programming languages support a more complex data structure called an array Individual cells are assigned a number or numbers， and data values are inserted manipulated， and extracted by referencing those A better solution is to organize the data as a file (F3) All computer data begin as patterns of bits， On a file， the bits are grouped to form Groups of characters， in turn， form meaningful data elements called A group of related fields is a record the file is a set of related For example， in a name and address file， an individual’s name is a Each record holds a complete set of data for data for a single individual ( a name， a street address， and so on) he file consists of all the Fig 3 Characters are grouped to form Fields are prouped to form A file is a group of related The data in a file are processed record by Normally， the file is stored on a secondary medium such as Programs are written to read a process its fields， generate the appropriate output， and then read and process another Because only one record is in main memory at a time， very little memory is Because many records can be stored on a single disk， a great deal of data can be processed in this limited 2 Locating Files Imagine a file stored on The first step in accessing its data is finding the The task is much like finding a program， but there are Following a command such as LOAD or RUN， programs are loaded by the application Data， on the other hand， are processed by application programs， in the context of a program’s Typically， just before the data are required， the program asks the operating system to open the Each file has a name; the open logic ( 4 ) reads the index， searches it by name， and finds the address of the first record in the F4 when a file is opened， the disk index is read into main memory and searched for the desired file’s If the file name is found， the file’s start address is extracted from the 3 Locating Records Once a file has located， the process of accessing its records can When a program needs input data， it reads a record; when it is ready to output results， it writes a Note that these instructions deal with selected records， not with the entire We open We read and write Let’s examine the data accessing process more A programmer views data logically， requesting the next record， or the name and address for a particular The data are stored on a secondary medium such as To access a record physically， the disk drive must be give a set of primitive commands: seeks， reads， and The programmer thinks in terms of logical I/O The external device stores and retrieves physical sectors; it “thinks” in terms of physical I/O There must be a mechanism for translating the programmer’s logical requests to the appropriate physical commands (Fig 5) On small computers， much of the logic is found in the operating system’s input/output control system; on larger machines， access methods are Increasingly， the programmer’s logical data request is translated to physical form by a database management 4 The Relative Record Concept How does software， be it operating system， access method， or database software， find specific records in a file? [2] The key to many storage and retrieval techniques is the relative record Imagine a string of 100 Numbers indicate a given record’s position relative to the first record in the The file’s first record ( relative record 0 ) is at “ start of file plus 0 ”;Its second record is at “ start of file plus 1”， and so Access Methods Imagine preparing meeting announcements for a You need a set of mailing labels， and each member’s name and address is recorded on an index Probably the easiest way to generate the labels is to copy the data from the first card， turn to the second card and copy it， and so on， processing the records sequentially， form the beginning of the file to the Magazine publishers face the same problem with each new issue， but need mailing labels for tens of thousands of Rather than using index cards， they store customer data on disk or magnetic tape， one record per The easiest way to ensure that all labels are generated is to process the records in the order in which they are stored， proceeding sequentially from the fist record in the file to the To simplify handing， the records might be presorted by zip code or a mailing zone， but the basic idea of processing the data in physical order still How dos this relate to the relative record number concept? A relative record number indicates a record’s position on the With sequential access， processing begins with relative record 0， then moves to relative record 1， 2， and son Accessing data sequentially involves little more than For example， imagine a program has just finished processing relative record What is the next record? Obviously， relative record We’ve already seen how a relative record number can be converted to read them， or write them， in physical Processing records in sequence is not always For example， when a subscriber moves， his or her address must be changed in the fire searching for that subscriber’s record sequentially is like looking for a telephone number by starting with the first page of the telephone book and reading line and That’s no how we use a telephone Instead， knowing the record are stored in alphabetical order ，we quickly narrow our search to a portion of a single page and then begin reading the entries， ignoring the bulk of the data the way we use a telephone book is good example of direct， or random， A disk drive reads or writes one record at a To randomly access a specific record， all the programmer must do is remember its address， and ask for it the problem is remembering all those disk addresses， One solution is maintaining an index of the Again， we’ll use the name and address fire as an We want to access individual customer record by As the file is created， records are written one at a time， in relative record number Additionally， as each record is written， the customer name and the associated relative record number are recorded in an array or After the last record has been written to disk and its position recorded on the index， the index is itself Once the index has been created， it can be used to find individual Assume， for example， that Susan Smith has changed her To record her now address on the file， a program read the file index， search the index for her name， find her relative record number， compute the disk address， and read her record， change her address， rewrite the record to the same place on Note that this specific record is accessed directly， and that no other records in the file are The basic idea of direct access is assigning each record an easy to remember， logical， and then converting that key to a relative record number， Given this relative location， a physical address can be computed， and record Using an index is one technique for converting keys to physical An option is passing a numeric key to an algorithm and computing a relative record Both techniques have the same objective; converting a programmer’s logical data requests to physical Earlier in the chapter we identified the gap separating logical and physical I/O An access method is a software module that bridges this gap， converting logical keys to physical addresses， and issuing the appropriate primitive There are many variations of sequential， indexed， and direct organizations， and each one has its own access Using a variety of data access techniques can be confusing， and this is one reason for the growing popularity of database management systems4 Database Management There are problems with traditional data Many result from viewing applications For example， consider Most organizations prepare their payrolls by computer because using a machine instead of a small army of clerks saves Thus， the firm develops a payroll program to process a payroll file， Inventory， accounts receivable， accounts payable， and general ledger analysis are similar applications， so the firm develops an inventory program， and inventory file， an accounts receivable program， an accounts receivable file， and so Each program is independent， and each processes it own independent data Why is this a problem? For on thing， different application often need the same data For example， schools generate both bills and student grade View the applications The billing program reads a file of billing data， and the grade report program reads an independent file of grade The outputs of both program are mailed to each student’s home; thus， student names and addresses must be redundantly recorded on both What happens when a student moves? Unless both file are up data， one will be Redundant data are difficult to A more subtle problem is data Each access method has its own rules for storing and retrieving data， and certain “tricks of the trade” can significantly improve the efficiency of a given Bacause the motivation for using the computer is saving The programmer is often tempted to save even more by taking advantage of these Thus， the program’s logic becomes dependent upon the physical structure of the When a program’s logic is tied to its physical data structure， changing that structure will almost certainly require changing the As a result， programs using traditional access method can difficult to The solution to both problems is often organizing the data as a single， integrated The task of controlling access to all the data can then be concentrated in a centralized database management How dose the use of a centralized database solve the data redundancy problem? All data are collected and stored in a single place; consequently， there is one and only one of any given data When the value of an element ( an address， for example) changes， the single database copy is corrected， Any program requiring access to this data element gets the same value， because there is only on How dose a database help to solve the data dependency problem? Since the responsibility for accessing the physical data rests with the database management system， the programmer can ignore the physical data As a result， programs tend to be much less dependent upon their data， and generally much easier to Expect the trend toward database management to There are problems with traditional data Because different applications often require the same data， certain data elements may be stored in several different places， and such redundant data are difficult to Another problem is data If a program’s logic is too closely linked to the physical structure of its data， that program can be difficult to The solution to both problems is often collecting all the organization’s data in a centralized With a database， there is only one copy of each data element， so the data redundancy problem is Because every program must access data through a database management system， programs are insulated from the physical data structure; thus， data dependency is
qynwsp83
