11.07.2015 Views

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

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

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

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

272 Chap. 8 File Processing <strong>and</strong> External SortingIntersector GapSectorHeaderSector<strong>D<strong>at</strong>a</strong>SectorHeaderSector<strong>D<strong>at</strong>a</strong>Intrasector GapFigure 8.4 An illustr<strong>at</strong>ion of sector gaps within a track. Each sector begins witha sector header containing the sector address <strong>and</strong> an error detection code for thecontents of th<strong>at</strong> sector. The sector header is followed by a small intra-sector gap,followed in turn by the sector d<strong>at</strong>a. Each sector is separ<strong>at</strong>ed from the next sectorby a larger inter-sector gap.8.2.2 Disk Access CostsWhen a seek is required, it is usually the primary cost when accessing inform<strong>at</strong>ionon disk. This assumes of course th<strong>at</strong> a seek is necessary. When reading a file insequential order (if the sectors comprising the file are contiguous on disk), littleseeking is necessary. However, when accessing a r<strong>and</strong>om disk sector, seek timebecomes the dominant cost for the d<strong>at</strong>a access. While the actual seek time is highlyvariable, depending on the distance between the track where the I/O head currentlyis <strong>and</strong> the track where the head is moving to, we will consider only two numbers.One is the track-to-track cost, or the minimum time necessary to move from <strong>at</strong>rack to an adjacent track. This is appropri<strong>at</strong>e when you want to analyze accesstimes for files th<strong>at</strong> are well placed on the disk. The second number is the averageseek time for a r<strong>and</strong>om access. These two numbers are often provided by diskmanufacturers. A typical example is the Western Digital Caviar serial ATA drive.The manufacturer’s specific<strong>at</strong>ions indic<strong>at</strong>e th<strong>at</strong> the track-to-track time is 2.0 ms <strong>and</strong>the average seek time is 9.0 ms. In 2008 a typical drive in this line might be 120GBin size. In 2011, th<strong>at</strong> same line of drives had sizes of up to 2 or 3TB. In both years,the advertised track-to-track <strong>and</strong> average seek times were identical.For many years, typical rot<strong>at</strong>ion speed for disk drives was 3600 rpm, or onerot<strong>at</strong>ion every 16.7 ms. Most disk drives in 2011 had a rot<strong>at</strong>ion speed of 7200 rpm,or 8.3 ms per rot<strong>at</strong>ion. When reading a sector <strong>at</strong> r<strong>and</strong>om, you can expect th<strong>at</strong> thedisk will need to rot<strong>at</strong>e halfway around to bring the desired sector under the I/Ohead, or 4.2 ms for a 7200-rpm disk drive.Once under the I/O head, a sector of d<strong>at</strong>a can be transferred as fast as th<strong>at</strong>sector rot<strong>at</strong>es under the head. If an entire track is to be read, then it will require onerot<strong>at</strong>ion (8.3 ms <strong>at</strong> 7200 rpm) to move the full track under the head. If only part ofthe track is to be read, then proportion<strong>at</strong>ely less time will be required. For example,if there are 16,000 sectors on the track <strong>and</strong> one sector is to be read, this will requirea trivial amount of time (1/16,000 of a rot<strong>at</strong>ion).Example 8.1 Assume th<strong>at</strong> an older disk drive has a total (nominal) capacityof 16.8GB spread among 10 pl<strong>at</strong>ters, yielding 1.68GB/pl<strong>at</strong>ter. Each

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!