{"title":"Distributed Information Storage Architecture","authors":"R. Baird, S. Karamooz, H. Vazire","doi":"10.1109/MASS.1993.289766","DOIUrl":"https://doi.org/10.1109/MASS.1993.289766","url":null,"abstract":"The authors describe the Distributed Information Storage Architecture (DISA) defined and prototyped at Hewlett-Packard by the Networked Systems Architecture organization. In the first phase of the project, an architecture framework was defined that isolated storage-management functions from database and file-management systems. The idea was to identify and encapsulate functions related to the persistent characteristics of stored data. Only functions common to a large variety of database and file-management systems were of interest. The intent was to build a new storage-management base without altering file or database systems. To validate this idea, a storage-management prototype called StoragePlus was built. The StoragePlus prototype automated several storage-management and administration functions currently done manually. It also demonstrated how one could reduce and control storage administrative costs by doing central administration of distributed storage across a network.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"195 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122141991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"High-performance data transfers using network-attached peripherals at the National Storage Laboratory","authors":"R. Hyer, R. Ruef, R. Watson","doi":"10.1109/MASS.1993.289749","DOIUrl":"https://doi.org/10.1109/MASS.1993.289749","url":null,"abstract":"One of the main goals of the National Storage Laboratory (NSL) is to advance the state-of-the-art of network-attached, high-performance storage architectures and to relate this experience back to the IEEE Storage System Standards Working Group (SSSWP). The work to date has concentrated on high-speed data transfers directly between storage devices and client processors (supercomputers, high-performance workstations, and frame buffers) using IPI-3 over HIPPI (High-Performance Parallel Interface). Data have successfully been transferred between an IBM 9570 HIPPI Disk Array and client processors using IPI-3 third-party copy protocols. The authors present work at the NSL in three areas: using IPI-3 over HIPPI and IPI-3 third-party copy to transfer data between client processors and storage devices or other processors; (2) using IPI-3 third-party copy to transfer data directly between two storage devices; and (3) relating the work at the NSL to the work of the IEEE SSSWP, Mover Subcommittee.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131971793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
E. May, D. Lifka, E. Lusk, L. Price, C. Day, S. Loken, J. MacFarlane, A. Baden, R. Grossman, X. Qin, L. Cormell, A. Gauthier, Phil Leibold, J. Marstaller, U. Nixdorf, B. Scipioni
{"title":"Requirements for a system to analyze high energy physics events using database computing","authors":"E. May, D. Lifka, E. Lusk, L. Price, C. Day, S. Loken, J. MacFarlane, A. Baden, R. Grossman, X. Qin, L. Cormell, A. Gauthier, Phil Leibold, J. Marstaller, U. Nixdorf, B. Scipioni","doi":"10.1109/MASS.1993.289780","DOIUrl":"https://doi.org/10.1109/MASS.1993.289780","url":null,"abstract":"The authors describe the requirements for the design and prototyping of an object-oriented database designed to analyze data in high energy physics (HEP). The goal is to satisfy the data processing and analysis needs of a generic HEP experiment to be proposed for the Superconducting SuperCollider (SSC). The experiment requires the collection and analysis of between 100 million and one billion sets of vectors (events), each approximately one megabyte in length. The authors outline how this analysis would proceed using an object-oriented database which supports the basic data types used in HEP.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131182467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Security in the IEEE Mass Storage System Reference Model","authors":"A. Hanushevsky","doi":"10.1109/MASS.1993.289775","DOIUrl":"https://doi.org/10.1109/MASS.1993.289775","url":null,"abstract":"The author discusses the origins of the distribution topology of the IEEE Mass Storage System Reference Model, how it affects security, and the methods that can be used to provide secure resource access. He also shows how certain security considerations emphasize the importance of particular Model-defined services and impose architectural constraints on those services. It is noted that the model's security problems stem from the client-centric distribution of various Model functions. These problems are heightened in Version 5 of the Model, in which any Model component is allowed to be distributed in a client-centric manner. While a distributed security scheme can provide an appropriate security environment, the relationships between Model components must be fully defined before any particular scheme can be recommended. Furthermore, it is critical that the correct authorization agents be identified with security, usability, and performance in mind. It is clear that, in certain cases, the Name Server becomes an important, if not focal, authorization agent. The challenge is to define a security protocol appropriate to the Model, as opposed to modifying the Model to fit current security schemes.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132574148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An introduction to the Mass Storage System Reference Model, version 5","authors":"R. Coyne, H. Hulen","doi":"10.1109/MASS.1993.289778","DOIUrl":"https://doi.org/10.1109/MASS.1993.289778","url":null,"abstract":"Brief histories of the IEEE Mass Storage System Reference Model and the IEEE Storage System Standards Working Group are presented. The Reference Model is explained in terms of layers of abstraction, including storage-container layers, mapping-service layers, and storage-service layers.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124328267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Toward a standard IEEE Mover","authors":"J. Merrill","doi":"10.1109/MASS.1993.289776","DOIUrl":"https://doi.org/10.1109/MASS.1993.289776","url":null,"abstract":"It is noted that, since the early work on the IEEE Mass Storage System Reference Model (MSSRM) in the fall of 1983, mover functionality was recognized as an important part of a data storage system. After giving a brief overview of the Version 4 Mover, the changes and enhancements worked out by the Mover Subcommittee in the process of upgrading the Mover to Version 5, along with the rationale for these changes, are described. In Version 5 of the MSSRM, the Mover has changed from being almost totally dependent on an intimate connection to either the Storage Server or the Bitfile Client to being an independent, self-contained entity. The new Mover also has become somewhat of a guardian of the physical devices in a storage system and the bitfiles stored there. By adding security features and client authentication, the Mover now has the ability to protect data and allow various types of restricted access to devices. The author also discusses some of the issues that still need to be worked out by the IEEE Storage System Standards Working Group in integrating the Mover into the overall model and working toward a standard.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114600796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jonathan S. Goldick, Kathy Benninger, Woody Brown, C. Kirby, Christopher Maher, D. S. Nydick, Bill Zumach
{"title":"An AFS-based supercomputing environment","authors":"Jonathan S. Goldick, Kathy Benninger, Woody Brown, C. Kirby, Christopher Maher, D. S. Nydick, Bill Zumach","doi":"10.1109/MASS.1993.289769","DOIUrl":"https://doi.org/10.1109/MASS.1993.289769","url":null,"abstract":"The Pittsburgh Supercomputing Center (PSC) has been using the Andrew File System (AFS) to integrate its supercomputing environment. Work has been completed to make AFS the PSC's primary, permanent file-storage system. The PSC is using many of the components of this system in production. The PSC has extended AFS to meet its mass storage needs. These extensions include adding archival services to the AFS server and porting the AFS client to the Unicos operating system. The authors discuss the advantages of this system as well as how it performs in a supercomputing environment.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129305231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Network Storage Service usage characteristics","authors":"R. A. Haynes","doi":"10.1109/MASS.1993.289753","DOIUrl":"https://doi.org/10.1109/MASS.1993.289753","url":null,"abstract":"A study of the usage of the UNIX-based Network Storage Service (NSS) at Sandia National Laboratories is described. This NSS usage characterization study was conducted to determine how the NSS is being used and to identify opportunities to improve its usage. The determination of the NSS usage characteristics was made possible by the collection of File Transport Protocol (FTP) and Network File System (NFS) activity data. The average daily usage for the period used in the study indicated that approximately 5 gigabytes are transferred daily with FTP and 0.5 gigabytes are transferred daily with NFS. Since the sample period used in the study is typically a slow time, this daily rate is expected to increase in the coming months. It is shown that the NSS is being used primarily as an archive, with a 4 to 1 ratio of bytes stored to bytes retrieved.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133644148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The storage server as virtual volume manager","authors":"A. Buck, R. Coyne","doi":"10.1109/MASS.1993.289774","DOIUrl":"https://doi.org/10.1109/MASS.1993.289774","url":null,"abstract":"The storage server function has undergone considerable change since Version 4 of the Mass Storage System Reference Model. The authors survey the major changes in the storage server, and motivate a number of storage server concepts and functions through a survey of commercial storage volume management products. A classification of physical storage devices by their major attributes is presented as a physical volume model. The virtual volume model is presented as a recursive composition of physical extents of physical volumes. Finally, the authors propose a restructuring of the service primitives between model layers to simplify interfaces and support arbitrary device-to-device transfers without intermediate copies.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114999201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The IEEE Storage System Standards Working Group Physical Volume Repository","authors":"J. D. Wishner","doi":"10.1109/MASS.1993.289773","DOIUrl":"https://doi.org/10.1109/MASS.1993.289773","url":null,"abstract":"The author describes the current status of the Physical Volume Repository (PVR) development and some of the issues that the PVR subcommittee is studying. The PVR, a module of the IEEE Mass Storage System Reference Model, is responsible for managing removable media volumes stored at a single location. Its architecture is that of a server which provides either manual or automated removable media services to clients using a client-server model. The primary client of the PVR is a Physical Volume Library, but the model does not impose restrictions on the number or nature of clients of the PVR. Several abstractions are used in the model to encapsulate the manufacturer-specific access to physical devices that handle media, so that clients are presented with a device-independent, logical interface.<<ETX>>","PeriodicalId":225568,"journal":{"name":"[1993] Proceedings Twelfth IEEE Symposium on Mass Storage systems","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126732297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}