By Topic

Mass Storage Systems, 1991. Digest of Papers., Eleventh IEEE Symposium on

Date 7-10 Oct. 1991

Filter Results

Displaying Results 1 - 25 of 30
  • Digest of Papers. Eleventh IEEE Symposium on Mass Storage Systems (Cat. No.91CH3039-5)

    Publication Year: 1991
    Request permission for commercial reuse | PDF file iconPDF (18 KB)
    Freely Available from IEEE
  • Renaissance: managing the network computer and its storage requirements

    Publication Year: 1991, Page(s):3 - 10
    Cited by:  Papers (2)  |  Patents (9)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (666 KB)

    The authors describe a set of solutions developed to address the storage management problem for networks of workstations and fileservers. Specifically, they describe the implementation of the Renaissance Infinite Storage Manager (ISM). Based on the bitfile client and bitfile server components of the IEEE Mass Storage Reference Model, the network ISM uses a mass storage server as backing store for ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Local area gigabit networking

    Publication Year: 1991, Page(s):11 - 16
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (475 KB)

    It is noted that networks using gigabit speeds are just starting to become available and offer a whole new set of problems and potentials. The author addresses what the higher speeds are being used for, the 'standards' efforts specifying the higher-speed channels, the network architectures being proposed, and some of the open problems requiring extensive further work. It is noted that HIPPI (High ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • NASA Earth Observing System Data Information System (EOSDIS)

    Publication Year: 1991, Page(s):18 - 19
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (133 KB)

    EOSDIS is being developed by the National Aeronautics and Space Administration (NASA) to be a comprehensive data and information system providing the Earth science research community with easy, affordable, and reliable access to EOS and other appropriate Earth science data. EOSDIS will archive approximately 2 TBytes of data per day over a 15-year period. The system will be geographically distribut... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • National Supercomputing Center mass storage system requirements

    Publication Year: 1991, Page(s):20 - 21
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (147 KB)

    As part of the National Supercomputing Center effort to define mass storage requirements and directions, six national centers were asked to fill out a questionnaire on the topic. Mass storage system requirements deemed significant by the respondents are presented.<> View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Client requirements for a UNIX-based network file-transport utility

    Publication Year: 1991, Page(s):22 - 24
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (199 KB)

    A file-transport utility is being developed at Lawrence Livermore National Laboratory to meet the requirements specified by users. It is intended to be useful for general UNIX environments using TCP/IP. It will communicate with FTP demons, a centralized persistence server, and the UniTree storage system. To provide true transparency, a global machine-independent name space, an integrated storage h... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High-performance data systems

    Publication Year: 1991, Page(s):25 - 26
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (95 KB)

    It is noted that advances in massively parallel, large-memory computers and cooperative processing networks allow researchers to execute large-scale codes that generate massive amounts of data. Recent developments in high-speed networks allow large amounts of data to be transferred through a HiPPI (High Performance Parallel Interface) network in reasonable time frames. To support this high-perform... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Mass storage requirements in the intelligence community

    Publication Year: 1991, Page(s):27 - 28
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (174 KB)

    The major requirements established among a significant portion of mass storage system workloads in the intelligence community are large numbers (10/sup 7/-10/sup 9/) of objects and a range of small (10 bits) to large-sized (10/sup 8/-10/sup 10/ bits) data objects. Two very different but representative workloads are presented, one centralized and the other distributed.<> View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • MSS requirements for data acquisition systems

    Publication Year: 1991, Page(s):29 - 30
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (111 KB)

    High-performance data acquisition systems differ in their general requirements in that they accept an input data stream that is essentially uninterruptible, purposefully collect multiple petabytes of data per year, and require near-real-time, high-performance processing on the input stream: the data object creation, storage, and usage domains. The characteristics of these three domains are describ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Factors affecting the performance of a DOS-based WORM file server

    Publication Year: 1991, Page(s):33 - 37
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (367 KB)

    Commercial software was selected to support archiving and retrieving document image files to and from WORM media over a LAN. As this software was integrated into the system, a performance study was initiated to determine the quantitative effect of several factors on the speed and reliability of image file transfer. These factors include network interface hardware, LAN organization, CPU clock speed... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Benchmarking a network storage service

    Publication Year: 1991, Page(s):38 - 44
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (555 KB)

    Benchmarking a network file server introduces some unique considerations over traditional benchmarking scenarios. Since the user is executing on a client system interconnected to the file server, the client and network must be provided for during benchmarking. During a recent procurement action, Sandia National Laboratories (SNL) was challenged to develop a benchmark suite that would accurately te... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance of balanced disk array schemes

    Publication Year: 1991, Page(s):45 - 50
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (354 KB)

    The balanced information dispersal algorithm (BIDA) provides attractive reliability enhancements as well as modest performance gains for the storage of data at a number of storage sites. A balanced disk array (BDA) is an application of BIDA. Although a BDA appears as a single, very reliable, fast direct-access device, it actually consists of a number of independent disks that store data according ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analyzing the I/O behavior of supercomputer applications

    Publication Year: 1991, Page(s):51 - 55
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (435 KB)

    The authors describe the collection and analysis of supercomputer I/O (input/output) traces on a Cray Y-MP. Analyzing these traces, which came primarily from programs with high I/O requirements, shows the file system I/O patterns that these applications exhibit. The authors classify application I/Os into three categories, required, checkpoint, and data staging, and show how memory size and CPU spe... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance study of client-server storage systems

    Publication Year: 1991, Page(s):67 - 72
    Cited by:  Papers (2)  |  Patents (9)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (470 KB)

    The authors study the performance of a system in which mass storage is provided by a file server node on a network of client processors. The server controls a hierarchy of storage devices and provides all necessary file management functions. The clients are large mainframe UNIX systems that have their own local storage devices. The movement of files among the various storage levels is managed by a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • E-systems Modular Automated Storage System (EMASS) software functionality

    Publication Year: 1991, Page(s):73 - 76
    Cited by:  Papers (6)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (338 KB)

    The software functionality of the E-Systems Modular Automated Storage System (EMASS) is described. EMASS is a complete data storage and retrieval system that operates as a large data storage node on a network, servicing client requests over a number of different interfaces. The EMASS server, a CONVEX supercomputer, stores files in an extended UNIX File System. These extensions provide for transpar... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design for a transparent, distributed file system

    Publication Year: 1991, Page(s):77 - 84
    Cited by:  Papers (2)  |  Patents (16)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (668 KB)

    At the Lawrence Livermore National Laboratory (LLNL), caching and migration protocols have been designed to integrate distributed UniTree File Management System servers running on separate machines to create a single file system. These protocols allow files to migrate between levels of a storage hierarchy to create a unified distributed storage system. The design provides clients with a single met... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dynamic hierarchies and optimization in distributed storage systems

    Publication Year: 1991, Page(s):85 - 91
    Cited by:  Papers (12)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (545 KB)

    The authors propose techniques to allow controlled penetration of the location transparencies of a distributed storage system by generalizing the simple disk-tape hierarchy paradigm of current storage systems to include multiple co-existing dynamic storage hierarchies. An orthogonal, semantics-free optimization primitive is proposed to manipulate the relation between an object and its hierarchy. B... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Volume management by the book: the NAStore Volume Manager

    Publication Year: 1991, Page(s):95 - 99
    Cited by:  Patents (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (325 KB)

    The NAStore Volume Manager provides controlled access to removable media under UNIX. It is compatible with the latest informal consensus of the IEEE Storage System Standards Working Group. The modular design allows it to manage robotic and manual drives of various types in many locations: a UNIX daemon fields requests and passes them to the appropriate 'Repository Controller' based on the medium a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The NCSC mass storage project

    Publication Year: 1991, Page(s):100 - 103
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (351 KB)

    The authors describe the development of the NCSC (North Carolina Supercomputing Center) Mass Storage System Project, which includes implementing a data storage system based on UniTree running on an IBM 3090 connected to a Cray Y-MP via a high-speed network. They review project planning and execution, giving up-to-date results of testing and user experiences. Integration and extension of the UniTre... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A high-performance UNIX network storage server

    Publication Year: 1991, Page(s):104 - 108
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (404 KB)

    The author describes the design and prototype testing of a large UNIX-based file server, the File Manager. The primary design aim for the File Manager was to provide network file service at an attractive cost/performance ratio. Economical storage as well as fast, concurrent file transfers to and from the server were essential. It was felt that this goal was best achievable through the selection of... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distributed storage management in high energy physics

    Publication Year: 1991, Page(s):109 - 112
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (348 KB)

    To cope with the large quantities of data produced in high energy physics, CERN has developed a system for the management of, and access to, data in a fully distributed environment. The principal user interface is via a package known as FATMEN (file and tape management: experimental needs), which provides a worldwide distributed file catalog and offers system- and medium-independent access to data... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The HADES file server

    Publication Year: 1991, Page(s):113 - 116
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (334 KB)

    HADES (Heidelberg Automatic Data management and Editor System), a file server for the IBM/370 world under the operating systems VM or MVS, is described. It may be used from CMS, MVS batch, and TSO directly and from UNIX systems via FTP and NFS. HADES has its own data management system independent of the host operating system. It uses automatic data migration onto tape to save disk space and keeps ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An AFS-based mass storage system at the Pittsburgh Supercomputing Center

    Publication Year: 1991, Page(s):117 - 122
    Cited by:  Papers (12)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (495 KB)

    The AFS (Andrew File System) with extensions forms the basis of the distributed mass storage system being implemented at the Pittsburgh Supercomputing Center (PSC). AFS offers transparent, locally cached, secure access to a universal file name space from a wide variety of UNIX platforms. The PSC is extending the standard AFS offering to provide a number of features that are useful in a supercomput... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Swift: a storage architecture for large objects

    Publication Year: 1991, Page(s):123 - 128
    Cited by:  Papers (6)  |  Patents (35)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (527 KB)

    The authors describe an input/output architecture called Swift that addresses the problem of storing and retrieving very large data objects from slow secondary storage at very high data rates. Swift addresses the problem of providing data rates required by digital video by exploiting the available interconnection capacity and by using several slower storage devices in parallel. Two studies have be... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • DataMesh-parallel storage systems for the 1990s

    Publication Year: 1991, Page(s):131 - 136
    Cited by:  Papers (2)  |  Patents (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (510 KB)

    A description is given of DataMesh, a research project with the goal of developing software to maximize the performance of mass storage I/O, while providing high availability, ease of use, and scalability. The DataMesh hardware architecture is that of an array of disk nodes, with each disk having a dedicated 20-MIPS single-chip processor and 8-32 MBytes RAM. The nodes are linked by a fast, reliabl... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.