Queueing Analysis of the IBM 2314 Disk Storage Facility

Abate, Joseph; Dubner, Harvey; Weinberg, Sheldon B.

doi:10.1145/321479.321484

Cited by 45 publications

(14 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Abate, Dubner, and Welnberg [1], Kritzmger [8], and Gotheb and MacEwen [6] modeled the IBM 2314 as an FCFS disk. Abate et al assumed that the FCFS disk could be described by two independent queueing systems m tandem, as shown in Figure 5(a), the first accounting for the spindle queues and the seek servers, and the other accountmg for the preparation and channel service phases.…”

Section: The Fcfs Diskmentioning

confidence: 99%

“…Abate et al assumed that the FCFS disk could be described by two independent queueing systems m tandem, as shown in Figure 5(a), the first accounting for the spindle queues and the seek servers, and the other accountmg for the preparation and channel service phases. The use of mdependent queueing systems has two consequences: (1) seek commands can be issued without use of the channel; (2) the seek servers become free immediately after seek completions, so that another seek operation may be begun on a spindle before the record associated with the previous seek on that spindle has been transmitted. Kntzinger's model ( Figure 5(b)) is more hke ours, with the seek servers being blocked until the corresponding requests have left the system; he does, however, allow seek commands to be issued without their passing through the channel His model and that of Gotheb and MacEwen are essentially the same.…”

Section: The Fcfs Diskmentioning

confidence: 99%

“…We specialize our general model to the case of the FCFS disk, arriving at a more precise, but still approximate, description than either [1], [8], or [6]. We test the formulas we derive against simulations to determine their accuracy.…”

Section: The Fcfs Diskmentioning

confidence: 99%

“…Another characteristic we associate with the FCFS disk is that the process of locating a record requires the channel, so an access request will use the channel for both its latency and transfer times. Abate, Dubner, and Wemberg [1], Kntzinger [8], and Gotlieb and MacEwen [6] used the FCFS disk as a model of the IBM 2314. One could also order transfers by the completion of latencies instead of the completion of seeks.…”

mentioning

confidence: 99%

See 3 more Smart Citations

A General Model for the Performance of Disk Systems

Wilhelm

1977

J. ACM

View full text Add to dashboard Cite

The performance of a disk system IS often measured in terms of the length of the waiting hne or queue of requests for each of the system's spindles Thus 1t is natural to formulate and analyze queuemg models of disk systems While most disk systems have certain characteristics, such as channel interference and concurrent seeks, in common, previous analyses have always been begun from scratch, without exploiting this commonahty We introduce a general queuelng model for disk systems, which incorporates the characteristics common to most disk systems, and use 1t in the approximate analyses of models of the IBM 2314 and 3330 disk systems Comparisons with simulation statistics show that the approximations made are very good over a wide range of arrival rates and system parameters We also show how to use the analytic results to investigate performance differences between devices KEY WORDS AND PHRASES disk storage, queuemg model, queuelng analysis, scheduling, secondary storage CR CATEGORIES 4 3, 4 41, 6 2, 6 35 lntroducttonMass storage devices are important elements of ttme shared and multiprogrammed computer systems because they provide the storage for both the permanent system and user files and the temporary files necessary for compilations, assembhes, etc Since the performance of a computer system is dependent on the speed with which files can be accessed, the mass storage devices of a system can have a significant effect on its overall performance. Thus it is important to study the behavior of mass storage devices and to develop mathematical models of thetr behavior from which performance predictions can be derwed. We formulate a general model for mass storage devices composed of several independent, moving head disk units attached to a single input/output channel. The model includes only the salient features of disk operation and is therefore not dependent on details of device architecture. We then perform a queueing analysis of the general model, and show how the results can be applied to two cases of practical interest, the IBM 2314 and 3330.Accessing a record on a moving head disk reqmres three sequential operations. First is the seek operation, in which the read/write heads are mechanically positioned to the cylinder containing the record that was requested and the read/write head for the proper surface is electromcally selected. Next the records passing under the read/write head are examined to find the one to be transferred; the disk is said to be latent during this searching operation, and the time taken Is the latency time. Some disk systems cannot store the identifier of the record which is to be located and transmitted. Consequently, each time a record appears under the read/write head, the identifier of the requested record must be transmitted through the channel to the disk so that a comparison can be made with the identifier of the record under the read/write head. Other systems dwlde the disk surfaces into sectors and reqmre that records begin on sector boundaries, so that

show abstract

Section: The Fcfs Diskmentioning

confidence: 99%

Section: The Fcfs Diskmentioning

confidence: 99%

Section: The Fcfs Diskmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

A General Model for the Performance of Disk Systems

Wilhelm

1977

J. ACM

View full text Add to dashboard Cite

show abstract

“…'Fulle'r and Baskett [20] analyzed two approximate models for thIs type of channel and compared the perfonmance of these to that of Abate and Dubner [19]. As mentioned earlier, Gorenstein [4] analyzed this channel under paging operatIon using the results of Wang and Ghosh " [5] j data transfers were taken to be a sector traversal time (I.e., one page of In,format ion) • In contras t. references [19] and [20] a' low for arbitrary data trimsmiss,lol) time.…”

Section: Queueing Models For Channel Operationmentioning

confidence: 99%

Estimating response time for auxiliary memory configurations with multiple movable-head disk modules

Omahen

1975

Proceedings of the 1st International Conference on Very Large Data Bases - VLDB '75

View full text Add to dashboard Cite

Abst ractThe hardware architecture for a large data base application often involves the use of movable-head disk modules for auxiliary memory. This paper considers design calculations for such systems and is divided into two main sections:(1) A survey is given of the literature dealing with queueing models for multiple-module movable head disk configurations. Results for these models allow a system designer to estimate the performance of a specified auxiliary memory configuration; for example, the average file response time is one result typically given.References are also provided to papers related to slngl.= movablehead disks which Lescr be techniques for estimating seek time distributiuns, queueing models for channel operations. etc. which ara useful for the multiple-module case.(2) A simple method is provided for estimating the average response time for a multi-module configuration of movable-head disk units attached to a single block multiplexer channel. The technique is a synthesis of a method described by Seaman, Lind, and Wilson for analyzing a similar configuration having a selector channel and variations of a method for treating a block multiplexer channel described by Fuller and Baskett. Seaman~~. find the mean response time by viewing the operation of each disk module as an M/G/l queueing system using the FCFS discipline, where the service time is the sum of the seek time, channel waiting time, and channel service time (rotational delay plus data transfer time).The channel operation in turn is analyzed using the l'machineinterference" model (i .e-., finite-Polsson-source. single-server system with exponential service time distribution). Fuller and Baskett treat the operation of a channel wlt~rotational position sensing by means of a queueing model with Persson arrivals (infinltesource) and service process consIsting of (a) two exponential stages corresponding to rotational delay and data transmission time, respectively, where the first stage has variable service (2) rate which is a function of the number of requests at the channel, or (b) one exponential stage with variable service rate dependent on number of requests at the channel system. The proposed technique involves the use of the method of Seaman~~. but replaces the machine-interference model for channel operation with either of two finite-source queueing models similar to those of Fuller and Baskett. IntroductionThe architecture of the auxtliary memary subsystem plays a critical role in determining the overall performance of a large data base system. For systems of this type, economic considerations often require that movabJehead disk units be employed because these devices offer lower cost per bit of secondary storage than that for drums and fixed-head disks yet are capable of achieving reasonable performance levels. This paper is concerned with certain design calculations of interest to a systems analyst, and the material covered is very relevant to the problem of hardware architecture for large data bases. The body of this paper is divided...

show abstract