Data Structures and Algorithm Analysis - Computer Science at ...

282 Chap. 8 File Processing and External Sortingbecome important. Note however that the in-memory copy time will always be farless than the time required to write the contents of a buffer to disk. For applicationswhere disk I/O is the bottleneck for the program, even the time to copy lots ofinformation between the buffer pool user and the buffer might be inconsequential.Another advantage to buffer passing is the reduction in unnecessary read operationsfor data that will be overwritten anyway.You should note that these implementations for the buffer pool ADT do not usegenerics. Instead, the space parameter and the buffer pointer are declared to bebyte[]. When a class uses a generic, that means that the record type is arbitrary,but that the class knows what the record type is. In contrast, using byte[] for thespace means that not only is the record type arbitrary, but also the buffer pool doesnot even know what the user’s record type is. In fact, a given buffer pool mighthave many users who store many types of records.In a buffer pool, the user decides where a given record will be stored but hasno control over the precise mechanism by which data are transferred to the backingstorage. This is in contrast to the memory manager described in Section 12.3 inwhich the user passes a record to the manager and has no control at all over wherethe record is stored.8.4 The Programmer’s View of FilesThe Java programmer’s logical view of a random access file is a single streamof bytes. Interaction with a file can be viewed as a communications channel forissuing one of three instructions: read bytes from the current position in the file,write bytes to the current position in the file, and move the current position withinthe file. You do not normally see how the bytes are stored in sectors, clusters, andso forth. The mapping from logical to physical addresses is done by the file system,and sector-level buffering is done automatically by the disk controller.When processing records in a disk file, the order of access can have a greateffect on I/O time. A random access procedure processes records in an orderindependent of their logical order within the file. Sequential access processesrecords in order of their logical appearance within the file. Sequential processingrequires less seek time if the physical layout of the disk file matches its logicallayout, as would be expected if the file were created on a disk with a high percentageof free space.Java provides several mechanisms for manipulating disk files. One of the mostcommonly used is the RandomAccessFile class. The following methods canbe used to manipulate information in the file.• RandomAccessFile(String name, String mode): Class constructor,opens a disk file for processing.