Theodore Johnson and Ethan L. Miller. Benchmarking tape system performance. In Proc. of Joint NASA/IEEE Mass Storage Systems Symposium, March 1998.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....be extended to include other type of storage media, drives and jukebox hardware. A good starting point to include magnetic tapes is to implement the model to es7 timate the locate time on serpentine tapes provided by Hillyer et al. 8] and the benchmark methodology presented by Johnson et al. [9]. 6 Generator Tools We have developed multiple tools to generate different workloads automatically. The data generated by these tools is stored in a database and can be used for multiple simulations. Thus, we can guarantee that different simulations are running with exactly the same input. The ....

T. Johnson and E. L. Miller. Benchmarking tape system performance. In Proc. of Joint NASA/IEEE Mass Storage Systems Symposium, March 1998.

....of tapes between robotic libraries for load balancing purposes. Robotic tape libraries are also modeled in [42, 43] The performance characteristics of a few particular tape drives are modeled in detail in [44, 45, 46] and the performance of a wider variety of tape drives is described in [47, 48]. 3.3 Simulation Simulation has long been a tool for detailed performance analysis on storage devices. Pantheon and DiskSim are two well known detailed disk simulators [49, 50] These simulators and others like them have been used to support research into various aspects of storage subsystems, ....

T. Johnson and E. L. Miller, \Benchmarking tape system performance," in Sixth Goddard Conference on Mass Storage Systems and Technologies in cooperation with the Fifteenth IEEE Symposium on Mass Storage Systems, pp. 95-112, Mar. 1998.

....the wear tapes suffer from random access is considerable. We do not present a profound analysis of tape technology. Through the years much has been written about magnetic tape and their performance. Hillyer et al. 8] and Chervenak [3] provide a good overview of tape technology. Johnson et al. [11] present a benchmark methodology for tapes. The PC technical guide [15] gives an on line overview of current technology and products in the markets. Magnetic tapes store data as small magnetized regions. The tape is composed of magnetic material deposited on a thin flexible substrate. The tape is ....

....in bad predictions. Therefore, there is not a unique locate time model for a given tape drive technology, but there must be an individual model for each tape in the jukebox. Estimating the locate time for linear tapes is easier, because the seek time is linear in the distance the tape must travel [11]. 3 Model We now present an analytical model to compute the time to perform each of the jukebox operations. Figure 5 shows the structure of the model. At the higher level we have the jukebox model that provides the higher level functions that compute the time to load and unload a disc and to ....

Theodore Johnson and Ethan L. Miller. Benchmarking tape system performance. In Proc. of Joint NASA/IEEE Mass Storage Systems Symposium, March 1998.

.... of work in addressing the flow of data to and from secondary storage devices (e.g. magnetic disks) 1, 2, 3, 4, 5, 6, 7, 8, 9] There has also been a significant amount of work on the management of large scale data in a storage hierarchy involving tertiary storage devices (e.g. tapes devices) [10, 11, 12, 13, 14]. Striping has been studied to improve the response time of tertiary storage devices [15, 16] The Department of Energy s ASCI plan draws an outline of the expected storage requirements for large scale computational challenges. According to this plan, a large scientific application today is ....

T. Johnson and E. L. Miller. Benchmarking Tape System Performance. In Proc. of the 15th IEEE Mass Storage Systems Symposium, 1996.

....of video tape library acceses, nor an experimental investigation of the performance issues in video tape library environments under efficient scheduling algorithms. 3 Related work has mostly concentrated on modeling the performance characteristics of tape drive and tape library products ([1, 9, 13, 14]) on comparative studies of the use of tertiary storage for multimedia ( 1, 2, 7] on storing and elevating video blocks from tertiary for playback ( 5, 18] on caching digital library documents in secondary storage ( 15] on striping and analytical modeling of tape libraries under FCFS ....

T. Johnson and E.L. Miller, "Benchmarking tape system performance", Proc. of joint IEEE/NASA Symp. on Mass Storage Systems, March 98.

....concentrated on techniques for cleverly exploiting the PS and SS resources of the system in order to increase its performance. With respect to tertiary storage, there have been efforts at modelling the performance characteristics of individual tape drives and robotic tape libraries [HS96b,HS96a,JM98a,JM98b] and [Joh96,JM98a,JM98b] respectively, to predict future references and prefetch documents in the SS [KW98] for digital library applications, to derive intelligent placement strategies of data in the library [CTZ97] as well as efforts to derive intelligent scheduling algorithms for ....

....for cleverly exploiting the PS and SS resources of the system in order to increase its performance. With respect to tertiary storage, there have been efforts at modelling the performance characteristics of individual tape drives and robotic tape libraries [HS96b,HS96a,JM98a,JM98b] and [Joh96,JM98a,JM98b] respectively, to predict future references and prefetch documents in the SS [KW98] for digital library applications, to derive intelligent placement strategies of data in the library [CTZ97] as well as efforts to derive intelligent scheduling algorithms for multiplexed video streams over ....

T. Johnson and E. Miller. Benchmarking tape system performance. In Proceedings of the Joint NASA and IEEE Mass Storage Conference, March 1998.

.... in addressing the flow of data to and from secondary storage devices (e.g. magnetic disks) 6, 9, 11, 34, 16, 29, 31, 32, 5] There has also been a significant amount of work on the management of large scale data in a storage hierarchy involving tertiary storage devices (e.g. tapes 8 9 devices) [37, 22, 24, 25, 30]. Striping has been studied to improve the response time of tertiary storage devices [15, 19] Also, using libraries on top of hierarchical storage systems is not a new concept. Watson et al. [46] implemented a MPI IO version in the HPSS. 3. APRIL In this section, we give an overview of the ....

T. Johnson and E. L. Miller. Benchmarking Tape System Performance. In Proc. of the 15th IEEE Mass Storage Systems Symposium, 1996.

.... work in addressing the flow of data to and from secondary storage devices (e.g. magnetic disks) 3, 4, 5, 10, 11, 14, 15, 16, 2] There has also been a significant amount of work on the management of large scale data in a storage hierarchy involving tertiary storage devices (e.g. tapes devices) [19, 12, 17, 18, 7]. Striping has been studied to improve the response time of tertiary storage devices [13, 6] The Department of Energy s ASCI plan draws an outline of the expected storage requirements for large scale computational challenges. According to this plan, a large scientific application today is ....

T. Johnson and E. L. Miller. Benchmarking Tape System Performance. In Proc. of the 15th IEEE Mass Storage Systems Symposium, 1996.

.... of work in addressing the flow of data to and from secondary storage devices (e.g. magnetic disks) 1, 2, 3, 4, 5, 6, 7, 8, 9] There has also been a significant amount of work on the management of large scale data in a storage hierarchy involving tertiary storage devices (e.g. tapes devices) [10, 11, 12, 13, 14]. Striping has been studied to improve the response time of tertiary storage devices [15, 16] The Department of Energy s ASCI plan draws an outline of the expected storage requirements for large scale computational challenges. According to this plan, a large scientific 1 application today is ....

T. Johnson and E. L. Miller. Benchmarking Tape System Performance. In Proc. of the 15th IEEE Mass Storage Systems Symposium, 1996.

....as good as a vertical layout with full replication. Finally, the space given to replicas at the ends of the tapes can be recaptured by overwriting the replicas with base data. 5 Related Work Several recent articles describing tape technology and schemes to improve tape performance have appeared [10, 11, 9, 7, 8, 5, 12, 16, 2, 1, 13, 14, 15], but none of them study replication to increase tape performance, and only [14, 15] address the scheduling of requests over multiple tapes in a jukebox. In [16] query pre execution is used to determine the set of tape blocks to be retrieved from a single tape so that scheduling can minimize ....

T. Johnson and E. L. Miller. Benchmarking tape system performance. In Sixth Goddard Conference on Mass Storage Systems and Technologies in cooperation with the Fifteenth IEEE Symposium on Mass Storage Systems, pages 95--112, Mar. 1998.

....and Applications, Ambleside, U. K. July 1993; the IEEE International Symposium on Information Theory, Trondheim, Norway, June 1994; and the IEEE International Symposium on Information Theory, Whistler, Canada, September 1995. 2 and digital libraries are anticipated over the next few years [2, 20, 34]. The information retrieval problem arises when the quantity of information in the database overwhelms the ability of the storage system to deliver requested information quickly, and thus in which there are nontrivial lower limits on access times. The task in the design of the information ....

....with respect to perturbations of this model, as discussed in Section VIII. We have a given physical structure for the storage system, with a scale of costs associated with transitions within the system. Reviews of the physical structures of current storage systems are given by Johnson and Miller [20], Katz [22] and McKusick [30] Data storage is typically organized in a storage hierarchy, represented as a pyramid, with a small amount of data in fast caches (at the top of the pyramid) ranging to large amounts of data stored in lower cost, higher retrieval time memory (at the bottom of the ....

T. Johnson and E. L. Miller, "Benchmarking tape system performance," in Sixth NASA GSFC Conf. on Mass Storage Systems and Technologies / Fifteenth IEEE Symp. on Mass Storage Systems (College Park, Maryland, Mar. 1998). 31

....enormous storage requirements of modern applications, such as continuous media services, digital libraries, and scientific computing systems. Hence, research on design and evaluation of configurations and resource management techniques at various levels of the storage hierarchy is gaining momentum [23, 18, 22, 29, 1, 17]. All these works illustrate two important points: 1) it is difficult to evaluate and predict performance of storage devices and (2) the required performance characteristics are often achieved at the cost of complex solutions (be they data layout, scheduling, or other algorithms) given the ....

....of application related workload data) and facilities for collection of measurements for user defined (as well as predefined) performance metrics. Work has already been done on some aspects of these requirements, for instance on benchmarking of disks [33] and to a lesser extent tape systems [22, 19]. All of these are needed for a useful design tool. Thus, we incorporate existing techniques into ViPEr HiSS, as appropriate, and develop our own where the need exists. 2 Note also that, the goals of ViPEr HiSS are different from those of designing a multimedia file system mainly, in that it is ....

T. Johnson and E. L. Miller. Benchmarking tape system performance. In Joint Sixth Goddard Conference on Mass Storage Systems and Techn ologies and Fifteenth IEEE Symposium on Mass Storage Systems, pages 95--112, College Park, March 1998.

.... part of the total transfer time (the remainder of the time being spent loading the cartridge and fastforwarding to the file) 21, 22, 23] However, with smaller files, the read write time is often negligible, and the time required to mount the tape and find the file dominates the user wait time [11, 14, 15, 16, 20]. An algorithm that ignores file size may migrate many small files to tertiary storage, instead of one large file with a slightly different migration value. The result is fewer files on disk, and normally a higher on disk miss rate and longer user wait times. Despite this drawback, some commercial ....

Theodore Johnson and Ethan L. Miller, "Benchmarking Tape System Performance," Fifteenth IEEE Symposium on Mass Storage Systems, Greenbelt, MD, March 1998, pages 355--372.

No context found.

Theodore Johnson and Ethan L. Miller. Benchmarking tape system performance. In Proc. of Joint NASA/IEEE Mass Storage Systems Symposium, March 1998.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC