An introduction to disk drivemodelingChris Ruemmler and John WilkesHewlett-Packard Laboratories, Palo Alto, CAMuch research in I/O systems is based on disk drive simulation models, but howgood are they? An accurate simulation model should emphasize the performance-critical areas.This paper has been published in IEEE Computer 27(3):17–29, March 1994. Itsupersedes HP Labs technical reports HPL–93–68 rev 1 and HPL–OSR–93–29.Copyright © 1994 IEEE.Internal or personal use of this material is permitted. However, permission toreprint/republish this material for advertising or promotional purposes or forcreating new collective works for resale or redistribution must be obtained from theIEEE. To receive more information on obtaining permission, send a blank emailmessage to [email protected]: this file was obtained by scanning and performing OCR on the IEEEpublished copy. As a result, it may contain typographic or other errors that are notin the published version. Minor clarifications and updates have been made to thebibliography.1Modern microprocessor technology is advancing at an incredible rate, and speedups of 40 to 60 percentcompounded annually have become the norm. Although disk storage densities are also improvingimpressively (60 to 80 percent compounded annually), performance improvements have been occurring atonly about 7 to 10 percent compounded annually over the last decade. As a result, disk system performanceis fast becoming a dominant factor in overall system behavior.Naturally, researchers want to improve overall I/O performance, of which a large component is theperformance of the disk drive itself. This research often involves using analytical or simulation models tocompare alternative approaches, and the quality of these models determines the quality of the conclusions;indeed, the wrong modeling assumptions can lead to erroneous conclusions. Nevertheless, little work hasbeen done to develop or describe accurate disk drive models. This may explain the commonplace use ofsimple, relatively inaccurate models.We believe there is much room for improvement. This article demonstrates and describes a calibrated, high-quality disk drive model in which the overall error factor is 14 times smaller than that of a simple first-ordermodel. We describe the various disk drive performance components separately, then show how theirinclusion improves the simulation model. This enables an informed trade-off between effort and accuracy.In addition, we provide detailed characteristics for two disk drives, as well as a brief description of asimulation environment that uses the disk drive model.Characteristics of modern disk drivesTo model disk drives, we must understand how they behave. Thus, we begin with an overview of the currentstate of the art in nonremovable magnetic disk drives with embedded SCSI (Small Computer SystemsInterconnect) controllers, since these are widely available.Disk drives contain a mechanism and a controller. The mechanism is made up of the recording components(the rotating disks and the heads that access them) and the positioning components (an arm assembly thatmoves the heads into the correct position together with a track-following system that keeps it in place). Thedisk controller contains a microprocessor, some buffer memory, and an interface to the SCSI bus. Thecontroller manages the storage and retrieval of data to and from the mechanism and performs mappingsbetween incoming logical addresses and the physical disk sectors that store the information.Below, we look more closely at each of these elements, emphasizing features that need to be consideredwhen creating a disk drive model. It will become clear that not all these features are equally important to amodel’s accuracy.The recording components. Modern disks range in size from 1.3 to 8 inches in diameter; 2.5, 3.5, and 5.25inches are the most common sizes today. Smaller disks have less surface area and thus store less data thantheir larger counterparts; however, they consume less power, can spin faster, and have smaller seekdistances. Historically, as storage densities have increased to where 2–3 gigabytes can fit on a single disk,the next-smaller diameter in the series has become the most cost-effective and hence the preferred storagedevice.2Increased storage density results from two improvements. The first is better linear recording density, whichis determined by the maximum rate of flux changes that can be recorded and read back; current values arearound 50,000 bits per inch and will approximately double by the end of the decade. The second comes frompacking the separate tracks of data more closely together, which is how most of the improvements areoccurring. Current values are about 2,500 tracks per inch, rising to perhaps 20,000 TPI by the end of thedecade. The product of these two factors will probably sustain a growth rate above 60 percent per year tothe end of the decade.A single disk contains one, two, or as many as a dozen platters, as shown in Figure 1. The stack of plattersrotates in lockstep on a central spindle. Although 3,600 rpm was a de facto standard for many years, spindlerotation speed has increased recently to as much as 7,200 rpm. The median rotation speed is increasing at acompound rate of about 12 percent per year. A higher spin speed increases transfer rates and shortensrotation latencies (the time for data to rotate under the head), but power consumption increases and betterbearings are required for the spindle. The spin speed is typically quoted as accurate within 0.5 to 1 percent;in practice, the disk speeds vary slowly around the nominal rate. Although this is perfectly reasonable forthe disk’s operation, it makes it nearly impossible to model the disk’s rotational position some 100-200revolutions after the last known operation. Fortunately, many I/O operations occur in bursts, so theuncertainty applies only to the first request in the burst.Each platter surface has an associated disk head responsible for recording (writing) and later sensing(reading) the magnetic flux variations on the platter’s surface. The disk drive has a single read-write datachannel that can be switched between the heads. This channel is responsible for encoding and decoding thedata stream into or from a series of magnetic phase changes stored on the disk. Significant fractions of the encoded data stream are dedicated to error correction. The application of digitalsignal
View Full Document
Unlocking...