B. Clifford Neuman. The Virtual System Model: A Scalable Approach to Orga- nizing Large Systems. Technical Report TR-90-05-01, University of Washington, May 1990.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....area networks to world wide internetworks the logistics of discovering resources of interest have become correspondingly more complex. The growing complexity of finding information information in one such large network, the Internet, has motivated a wide variety of new tools for naming support [2, 7, 9, 4]. For example, the Archie system [1] provides an on line index of the files that are available at over 200 public FTP sites. The Alex system [3] provides a uniform naming system along with caching services that span all of the hosts in the Internet, and interoperates with an underlying file system ....

B. Clifford Neuman. The Virtual System Model: A Scalable Approach to Orga- nizing Large Systems. Technical Report TR-90-05-01, University of Washington, May 1990.

....for global middleware. Also note that DNS is used to name 56 Theta 10 6 hosts in July 1999. 1 which is significantly less than the 10 12 objects we intend to support. We doubt that DNS can scale to very large numbers of hosts. The Prospero file system supports a naming model similar to ours [9, 8]. However, the focus of the Prospero system is completely different. The goal of Prospero is to examine how scale affects users, specifically how scale affects the usability of a large system. To enhance usability, Prospero allows users to build their own personal virtual system by customizing ....

B. C. Neuman. The Virtual System Model: A Scalable Approach to Organizing Large Systems. PhD thesis, University of Washington, June 1992.

....the system must also be highly transparent as well as highly scalable. The Prospero file system is a distributed system designed for local organization of information that is distributed across an Internet and for easy access to information at remote sites organized similarly (using Prospero) [Neu92]. It has been successfully used to organize information from Internet archive sites, including software releases, archives of Internet mailing lists, copies of technical reports, and conference papers. It has also been used as the preferred method of remote access to archie , a database that ....

B. C. Neuman. The virtual system model: A scalable approach to organizing large systems. PhD Thesis 92-06-04, University of Washington, Seattle, Washington, June 1992.

....Petri net for modeling TCP behavior. the HTTP protocol, the use of persistent connections for several requests (P HTTP) is possible and recommended, thus minimizing the number of necessary slow start phases and connection setups. Other approaches like T TCP [16] and UDP based mechanisms like ARDP [17] have been addressed in [3] B. Model Development The model presented in this section has been developed having three main aims in mind. First, the model should account for the window flow control mechanism and capture the influence of the slow start mechanism on transmission performance. ....

B. C. Neuman, The virtual system model: A scalable approach to organizing large systems, Ph.D. thesis, University of Washington, Seattle, 1992, available at ftp://prospero.isi.edu/pub/papers/prospero/prosperoneuman -thesis.ps.Z.

....users often like to customize their namespaces, while applications expect to access resources through names of their choice. The need for custom namespaces has been acknowledged in two recent systems: Plan 9 [Pike, 93] a successor of UNIX which provides per process namespaces, and Prospero [Neuman, 92] a naming system which allows users to customize namespaces. The key consequence of the need for separate and possibly customizable namespaces is that treating naming and data access together, as in the UNIX file system, is no longer a good proposition. 81 Like naming, data access should also ....

B. Neuman, The Virtual System Model: Scalable Approach to Organizing Large Systems, Ph.D. Dissertation, Dept. of Computer Science and Eng., University of Washington, WA.

....converts the raw site listings into a format that can be added to the filenames database. The user access component allows users on the Internet to access and query the archie servers. Currently, there are three possible access methods: the telnet interface, the e mail interface, and the Prospero [18, 17, 16, 19] interface. The telnet interface to archie consists of connecting to an archie server through the telnet command, and then submitting queries. Users can then submit queries to either the whatis and filenames databases. The telnet interface has proven to be very resource consuming, since each ....

....little more than a year in service, with over 1,000 registered FTP archive sites, making available 2,100,000 files and 3,500 different software packages, archie is accessed from 47 different countries, reporting 3,500 search queries per day at the Canada server. Prospero The Prospero File System [18, 17, 16, 19] is a tool for organizing distributed information. Prospero lets users build customized views, or virtual systems, of directories distributed throughout the Internet. The Prospero name space forms a generalized directed graph, where intermediate nodes are directories, leaves are files, and edges ....

B. Clifford Neuman. The virtual system model: A scalable approach to organizing large systems. Technical Report 90-05-01, Department of Computer Science and Engineering, University of Washington, Seattle, Washington, May 1990.

....assumption is not valid for the new query mechanism, then a new index utility would have to be created. Given this relative inflexibility for defining new access engines, WAIS scores a on this scenario. WAIS also scores a on the augmented version of this scenario. 15 7. 3 Prospero Prospero [6] is a system that enables viewing files across the Internet according to a locally defined view. Its primary focus is in the organization of information rather than the searching or retrieving of information [7] It is intended to be used in conjunction with other types of Internet systems such as ....

B.C.Neuman, "The Virtual System Model: A Scalable Approach to Organizing Large Systems", Ph. D. Thesis, University of Washington, 1992.

....of the stored object. A view of the repository desired by a user may be different from the organization of the storage. Grouping of logically related but separately stored objects is supported to a very limited extent by symbolic links. The Semantic File System [Gifford, 91] and Prospero [Neuman, 92] provide a much better, customizable organization, albeit under the file system interface. Information Access in Mobile Computing Environments 4 The standard repository is limited because it is attached to a machine or a set of machines. A user can normally only customize a part of the ....

....is limited because it is attached to a machine or a set of machines. A user can normally only customize a part of the hierarchy and a few other aspects like the contexts used for name resolution. The per process file hierarchy in Plan 9 [Pike, 93] and the virtual system model used in Prospero [Neuman, 92] aim at remedying this problem. A logical view may involve multiple servers, possibly speaking different protocols. A keyword based search may utilize utilities like find and grep on a UNIX file system and the WAIS interface for other logically related objects available from a different server. ....

Neuman B., The Virtual System Model: A Scalable Approach to Organizing Large Systems, Ph.D. Dissertation, University of Washington.

....upon and or support. The defining criteria for membership here is the meta nature of the information. The actual artifacts do not reside within these systems, but rather only information about how the artifacts are referenced and accessed via the underlying file space. 5. 1 Prospero Prospero [30, 31] is a virtual file system, that is, a file system whose organization is independent of the physical file system(s) upon which it resides. It was designed with the intent of supporting the organization of information in an adaptable, collaborative manner, with many individuals and sites ....

....of supporting the organization of information in an adaptable, collaborative manner, with many individuals and sites contributing to a shared pool of artifacts, while still supporting an individual s ability to create a personal perspective on that information organization. The original prototype [30] accomplished this through the definition of a set of virtual file commands (vls, vcd, etc. that corresponded with the equivalent set of UNIX commands (ls, cd, etc. These surrogate commands, rather than interrogating the local file system, instead interrogated virtual file servers on ....

Neuman, B. C., The Virtual System Model: A Scalable Approach to Organizing Large Systems, PhD dissertation, The University of Washington, 1992. Also available as technical report UW-CSE-92-06-04.

....from those used by the resource owners. Thus, the persona includes a resource view; a complete, well ordered name mapping to the real files, devices, resources and so forth. It is used both by the user and, perhaps more importantly, by his or her applications. This technique is used by Prospero [2] to manage access to data files in the Internet; in a related way, Plan 9 [3] does name space mapping to deal with files, devices and services. However, since the persona is mobile and may be resource constrained, its remapping is more complex than either of these systems. 4. Computing Persona ....

....be available due to storage limitations. Thus, the namespace has to zoom onto selected parts of the ocean of resources that are both interesting to the user and accessible within the dynamic resource limitations. The basic idea of customized namespaces has been investigated in several projects [2], 3] Cache Management Efficient realization of such a namespace has to rely on innovative caching. Caching algorithms designed for static, multi user LAN environments won t fit mobile computers as the user s view of resources is decoupled from the storage view. What a file server sees as a ....

B. Neuman, The Virtual System Model: A Scalable Approach to Organizing Large Systems, Ph.D. Dissertation, University of Washington, Seattle, 1992.

....that allow the user to personalize their access to the Internet WWW. With these tools, users can catalog and organize sites according to their preferences. Furthermore, when it seems to be useful, they can make those organizations available to others. Such tools include the Prospero file system [Neuman 1992a, Neuman 1992b] and the Alex file system [Cate 1992] 1.1.4 Which to Choose All three of the above approaches address important problems associated with better, easier to use, net access. However, in this paper, we address the third kind of approach, that of personalizing net access. ....

....which would provide personal modification to sites based on use as well as user preference. 2 Models for Personalized Access 2. 1 An Existing Approach, Prospero and the Virtual System Model As mentioned above, one of the systems addressing the personalized access approach to the net is Prospero [Neuman 1992a, Neuman 1992b] Neuman s thesis stated that existing approaches, based on a centralized server, were inappropriate for organizing access to an intrinsically distributed, large scale system. As such he developed the Virtual System Model, an approach that presented four principal features, listed ....

B.C. Neuman, "The Virtual System Model: A Scalable Approach to Organizing Large Systems", Ph.D. Dissertation, University of Washington, 1992.

....an environment, most Internet information is only minimally organized. The challenge for the designers of information systems is to help the user find the information that is of interest. Many of the issues here are similar to those that arise in naming research [Bowman, Peterson Yeatts 1990, Neuman 1992a, Schwartz 1987, Sollins 1985] ################## 1 FTP is the Internet standard File Transfer Protocol. Anonymous FTP is a convention for allowing Internet users to transfer files to and from machines on which they do not have accounts, for example to support distribution of free software and ....

B. C. Neuman. The Virtual System Model: A Scalable Approach to Organizing Large Systems. Tech. Rep. 9206 -04, Dept. Comput. Sci. and Engr., Univ. Washington, Seattle, WA, June 1992. Ph.D. Diss.

....provides. 1 Introduction Roughly speaking, resource discovery tools either organize or search information distributed across many repositories. Tools like the Internet gopher [1] the WorldTo appear in Computing Systems: Volume 5, Number 4, 1992. ALso USC TR 92 519) Wide Web [3] and Prospero [11] organize information into a distributed hypertext. With these tools, people create links between relevant information which may reside on different servers. Tools like archie [7] Nomenclator [14] and netfind [15] build indices from information scavenged from various repositories and sources. ....

B. Clifford Neuman. The virtual system model: A scalable approach to organizing large systems: A thesis proposal. Technical Report TR-90-05-01, University of Washington, Seattle, May 1990.

....to give an index broker a copy of the object, nor is it prevented from doing so. Users of the index broker may have to contact the primary databases to retrieve the object itself. Previous work on yellow page services, known as attribute based naming, has been restricted to network name servers [17, 4, 19, 20]. Our architecture extends attribute based naming to bibliographic databases. The system s final component, the replicated Topic Broker, describes every site and index broker [12] Users employ an instance of the Topic Broker to identify brokers relevant to their queries. Creators of brokers use ....

B. Clifford Neuman. The virtual system model: A scalable approach to organizing large systems: A thesis proposal. Technical Report TR-90-05-01, University of Washington, Seattle, May 1990.

....is not valid for the new query mechanism, then a new index utility would have to be created. Given this relative inflexibility for defining new access engines, WAIS scores a on this scenario. WAIS also scores a on the augmented version of this scenario. 15 7. 3 Prospero Prospero [6] is a system that enables viewing files across the Internet according to a locally defined view. Its primary focus is on the organization of information rather than the searching or retrieving of information [7] It is intended to be used in conjunction with other types of Internet systems such as ....

B.C.Neuman, "The Virtual System Model: A Scalable Approach to Organizing Large Systems", Ph. D. Thesis, University of Washington, 1992.

....(the UUnet archives at uunet.uu.net take up roughly 1GB of disk storage) Recently however, a number of methods for resource discovery have been developed and are available on the Internet ( ftp list file compiled by John Granose odin pilot.njin.net, Archie at archie.cs. mcgill.ca and Prospero [NEU]) It is desirable that the mid level networks be able to provide up to date pointers to the distribution hosts for available public software archives. Coordinating the distribution of a static list is difficult (though not impossible) and the use of automated resource discovery mechanisms such as ....

B.C. Neuman "The Virtual System Model: A Scalable Approach to Organizing Large Systems", Department of Computer Science, University of Washington, FR-35, Seattle, WA, May 1990.

....must makeavailable directory information from many sources including existing indexing schemes 1 and directory information speci#ed by users. It should be possible for directory information from di#erent sources to be combined in useful ways. The Virtual System Model The Virtual System Model #Neuman, 1992# provides a framework for organizing large systems within which users construct their own #virtual systems by selecting objects and services that are available over the network; users then treat the selected resources as a single system, ignoring those resources that were not selected. The ....

Neuman, B. Cli#ord. #1992#. The Virtual System Model: A scalable approach to organizing large systems. Doctoral dissertation #in preparation#. Department of Computer Science and Engineering, UniversityofWashington.

....distributed cyclic garbage. Ladin and Liskov, 1992] uses a logically centralized service which kept all information about inter systems refer1 ences, and [Jull and Jul, 1992] uses a global comprehensive tracing algorithm. Both of them do not scale well because they require global synchronization. [Neuman, 1992] uses a local lower bound instead of a global lower bound, that is, the algorithm sacrifices safety for performance. Lang et al. 1992] introduced tracing within a group instead of global systems to avoid global synchronization. However, it cannot collect distributed cyclic garbage completely. ....

....of calculating and maintaining backward references is too heavy. For the solution, we designed and implemented a practical and efficient distributed garbage collection mechanism. We have implemented a garbage collection mechanism which uses timeouts, last referenceable timestamp propagation [Neuman, 1992] and backward inquiry. The primary method used for local and remote garbage collection is timeouts, which is similar to leases in Java RMI [RMI, 1997] and pinging in DCOM [Chappell, 1996] Each object has a Time To Live (TTL) and an expiration, and each link maintains an expiration time of its ....

[Article contains additional citation context not shown here]

Barry Clifford Neuman, The virtual system model: A scalable approach to organizing large systems, Ph.D. dissertation, University of Washington, 1992.

....single organization that controls the system makes organization of these resources difficult. These characteristics can be addressed in part through support for customization. Users should be able to choose the resources and objects of interest, and treat the selected resources as a single system [5]. The directory service will play a critical role tying together the components of future systems. The requirements for such a directory service are greatly affected by the scale of the system, and the mobility of its users. One of the biggest problems to be addressed is support for transient ....

B. Clifford Neuman. The Virtual System Model: A Scalable Approach to Organizing Large Systems. PhD thesis, University of Washington, June 1992. Department of Computer Science and Engineering Technical Report 92-06-04.

....must make available directory information from many sources including existing indexing schemes 1 and directory information specified by users. It should be possible for directory information from different sources to be combined in useful ways. The Virtual System Model The Virtual System Model (Neuman, 1992) provides a framework for organizing large systems within which users construct their own virtual systems by selecting objects and services that are available over the network; users then treat the selected resources as a single system, ignoring those resources that were not selected. The ....

Neuman, B. Clifford. (1992). The Virtual System Model: A scalable approach to organizing large systems. Doctoral dissertation (in preparation). Department of Computer Science and Engineering, University of Washington.

....are not persistent. Instead, they are constructed by a process (often using a configuration file) and they only live as long as the processes that use them. The problems just described are addressed by Prospero and the Virtual System Model. 3 The Virtual System Model The Virtual System Model [12] provides a framework for organizing large systems within which users construct their own virtual systems by selecting objects and services that are available over the network; users then treat the selected resources as a single system, ignoring those resources that were not selected. By ....

B. Clifford Neuman. The Virtual System Model: A Scalable Approach to Organizing Large Systems. PhD thesis, University of Washington, 1992. Department of Computer Science and Engineering (in preparation).

....hidden from view. The organization of such systems should be based on the conceptual relationship between resources and the mapping to physical locations should be hidden from the user. These concepts form the basis of the Virtual System Model, a new model for organizing large distributed systems [12]. Appears in Concurrency: Practice and Experience, Vol6(4) 6 4200 June 1994. To apply the concepts of the Virtual System Model to the allocation of resources in large systems, we have chosen to divide the functions of resource management across three types of managers: the system manager, the ....

B. Clifford Neuman. The Virtual System Model: A Scalable Approach to Organizing Large Systems. PhD thesis, University of Washington, June 1992. Available as Department of Computer Science and Engineering Technical Report 92-06-04.

....File System User s Manual Version Alpha.4.4 B. Clifford Neuman Department of Computer Science and Engineering University of Washington, FR 35 Seattle, Washington 98195 bcn cs.washington.edu Draft of June 8, 1991 1 Introduction The Prospero file system is based on the Virtual System Model[1, 2, 3, 4]. It differs from traditional distributed file systems in several ways. In traditional file systems, the mapping of names to files is the same for all users. Prospero supports user centered naming: users construct their own view of the files that are accessible. The mapping from names to files is ....

B. Clifford Neuman. The Virtual System Model: A scalable approach to organizing large systems (a thesis proposal). Technical Report 90-05-01, Department of Computer Science, University of Washington, May 1990.

No context found.

B. C. Neuman, The virtual system model: A scalable approach to organizing large systems, Ph.D. thesis, University of Washington, Seattle, 1992, available at ftp://prospero.isi.edu/pub/papers/prospero/prosperoneuman -thesis.ps.Z.

No context found.

B. Neuman, "The Virtual System Model: A Scalable Approach to Organizing Large Systems", PhD Thesis, University of Washington, 1992.

First 50 documents

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