EW 7最新文献

{"title":"World wide failures","authors":"W. Vogels","doi":"10.1145/504450.504473","DOIUrl":"https://doi.org/10.1145/504450.504473","url":null,"abstract":"The one issue that unites almost all approaches to distributed computing is the need to know whether certain components in the system have failed or are otherwise unavailable. When designing and building systems that need to function at a global scale, failure management needs to be considered a fundamental building block. This paper describes the development of a system-independent failure management service, which allows systems and applications to incorporate accurate detection of failed processes, nodes and networks, without the need for making compromises in their particular design.","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129821996","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":"Using the web to coordinate distributed applications","authors":"P. Ciancarini, R. Tolksdorf","doi":"10.1145/504450.504486","DOIUrl":"https://doi.org/10.1145/504450.504486","url":null,"abstract":"Original Web middleware does not provide support for multiuser applications, such as groupware or workflow, as its basic nature is that of a passive information system. In fact, all activity is tied to server machines able to execute code using the CGI mechanism.\u0000Enhanced web browsers supporting applet scripting languages, such as Java [1], allow multithread activity at the user interface. However, there is no integrated middleware to coordinate activities tied to multiple, distributed clients. Coordination has to be centralized at some server to which all users participating in an application have to connect to. Thereby, the activity located at the browser does not really make the application distributed, as applets at the browser cannot connect to other applets providing services to them directly.\u0000For instance, think about a service to play chess or poker on the WWW against remote players. Java-enabled browsers can easily be used to build user interfaces including an interactive chessboard and other services, like a text-based messaging system. However, to coordinate a game over the Internet the Java programmer has to use sockets to transmit data and synchronizing events. This is a low-level approach subject to a variety of design and implementation errors.\u0000In order to solve such a class of problems, and simplify the design and implementation of distributed applications over the Web, we present here the architecture of the PageSpace, a software platform based on extending Java with coordination technology [3].\u0000The PageSpace introduces a notion of active, coordinable objects - the agents - which are able to both use and provide services from an to other agents, without requiring centralized coordination from servers.\u0000Distributed applications involve heterogeneous machine-, network- and operating-system architectures. Coordinating agents in these environments means to make these heterogeneities transparent to the programmer. Using the PageSpace, transparency of network heterogeneity is achieved by using the Internet middleware underlying the communication platform. Transparency of different hardware architecture and operating systems is provided by Java. PageSpace is built on a set of existing technologies:\u0000Coordination technology provides the conceptual platform for the coordination of activities amongst asynchronously working collaborating agents. Coordination technology has been initiated by the language Linda [3]. Linda is a minimalistic language,namely it includes a small number of constructs which need to be integrated in a conventional language to get a complete language for distributed programming. The basic conception of uncoupled coordination implicit in Linda has been studied in a large variety of research projects, focusing on parallel, distributed, and open distributed systems, on its theoretical foundations by giving semantics to coordination languages, and on a number of implementation oriented research concerning the embedding of ","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129633127","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 trading function in action","authors":"B. Jacob, T. Mudge","doi":"10.1145/504450.504492","DOIUrl":"https://doi.org/10.1145/504450.504492","url":null,"abstract":"This paper describes a commercial software and hardware platform for telecommunications and multimedia processing. The software architecture loosely follows the CORBA and ODP standards of distributed computing and supports a number of application types on different hardware configurations. This paper is the result of lessons learned in the process of designing, building, and modifying an industrial telecommunications platform. In particular, the use of the trading function in the design of the system led to such benefits as support for the dynamic evolution of the system, the ability to dynamically add services and data types to a running system, support for heterogeneous systems, and a simple design performing well enough to handle traffic in excess of 40,000 busy-hour calls.","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116012228","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":"NILE: wide-area computing for high energy physics","authors":"K. Marzullo, Michael Ogg, Aleta Ricciardi, A. Amoroso, F. A. Calkins, Eric Rothfus","doi":"10.1145/504450.504460","DOIUrl":"https://doi.org/10.1145/504450.504460","url":null,"abstract":"The CLEO project [2], centered at Cornell University, is alarge-scale high energy physics project. The goals of the projectarise from an esoteric question---why is there apparently so littleantimatter in the universe?---and the computational problems thatarise in trying to answer this question are quite challenging.\u0000To answer this question, the CESR storage ring at Cornell isused to generate a beam of electrons directed at an equally strongbeam of positrons. These two beams meet inside a detector that isembedded in a magnetic field and is equipped with sensors. Thecollisions of electrons and positrons generate several secondarysubatomic particles. Each collision is called an event andis sensed by detecting charged particles (via the ionization theyproduce in a drift chamber) and neutral particles (in the case ofphotons, via their deposition of energy in a crystal calorimeter),as well as by other specialized detector elements. Most events areignored, but some are recorded in what is called raw data(typically 8Kbytes per event). Offline, a second program calledpass2 computes, for each event, the physical properties ofthe particles, such as their momenta, masses, and charges. Thiscompute-bound program produces a new set of records describing theevents (now typically 20Kbytes per event). Finally, a third programreads these events, and produces a lossily-compressed version ofonly certain frequently-accessed fields, written in what is calledroar format (typically 2Kbytes per event).\u0000The physicists analyze this data with programs that are, for themost part, embarrassingly parallel and I/O limited. Such programstypically compute a result based on a projection of a selection ofa large number of events, where the result is insensitive to theorder in which the events are processed. For example, a program mayconstruct histograms, or compute statistics, or cull the rawdata for physical inspection. The projection is either the completepass2 record or (much more often) the smaller roarrecord, and the selection is done in an ad-hoc manner by theprogram itself.\u0000Other programs are run as well. For example, a Monte Carlosimulation of the experiment is also run (called montecarlo) in order to correct the data for detector acceptance andinefficiencies, as well as testing aspects of the model used tointerpret the data. This program is compute bound. Anotherimportant example is called recompress. Roughly every twoyears, improvements in detector calibration and reconstructionalgorithms make it worthwhile to recompute more accuratepass2 data (and hence, more accurate roar data) fromall of the raw data. This program is compute-bound (itcurrently requires 24 200-MIP workstations running flat out forthree months) and so must be carefully worked into the schedule sothat it does not seriously impact the ongoing operations.\u0000Making this more concrete, the current experiment generatesapproximately 1 terabyte of event data a year. Only recentroar data can be kept on disk; all oth","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116240678","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":"Information servers: a scaleable communication paradigm for wide area networks and the information superhighway","authors":"M. Wood, Bradford B. Glade","doi":"10.1145/504450.504501","DOIUrl":"https://doi.org/10.1145/504450.504501","url":null,"abstract":"The predominant use of point-to-point communication protocols on the global Internet, combined with the communication patterns imposed by the World Wide Web, make inefficient use of the Internet's increasingly scarce network bandwidth. While more network fiber helps to belay the impact of these inefficiencies, ultimately more intelligent coordination of data distribution and the use of more efficient distribution protocols will be necessary to support the user population of the Information Superhighway. To this end, we propose the notion of Information Servers. Using the publish-subscribe paradigm, Information Servers distribute information in such way to minimize network traffic.","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128852309","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 architecture for a wide area distributed system","authors":"P. Homburg, M. Steen, A. Tanenbaum","doi":"10.1145/504450.504465","DOIUrl":"https://doi.org/10.1145/504450.504465","url":null,"abstract":"Distributed systems provide sharing of resources and information over a computer network. A key design issue that makes these systems attractive is that all aspects related to distribution are transparent to users. Unfortunately, general-purpose wide area distributed systems that allow users to share and manage arbitrary resources in a transparent way hardly exist. In particular, they generally do not take into account the most important properties that characterize wide area systems: 1) A very large number of users and resources, 2) an inherent latency problem caused by the distance between nodes, 3) heterogeneity due to a variety of underlying operating systems and networks, and 4) involvement of multiple administrative organizations.The research described in this paper is part of the Globe Project (Globe stands for GLobal Object Based Environment). The goal of this project is the design and implementation of a wide area distributed system that provides a convenient programming abstraction and full transparency. The main contribution of this paper is the description of a new system for distributed shared objects. In contrast to other systems, the implementation of distribution, consistency, and replication of state is completely encapsulated in a distributed shared object. This allows for object-specific solutions, and provides the right mechanism for building efficient and truly scalable systems.","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117137774","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":"Server operating systems","authors":"M. F. Kaashoek, D. Engler, G. Ganger, D. Wallach","doi":"10.1145/504450.504478","DOIUrl":"https://doi.org/10.1145/504450.504478","url":null,"abstract":"We introduce server operating systems, which are sets of abstractions and runtime support for specialized, high-performance server applications. We have designed and are implementing a prototype server OS with support for aggressive specialization, direct device-to-device access, an event-driven organization, and dynamic compiler-assisted ILP. Using this server OS, we have constructed an HTTP server that outperforms servers running on a conventional OS by more than an order of magnitude and that can safely timeshare the hardware platform with other applications.","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122071395","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":"DSM-CC for world wide applications","authors":"Liam M. Casey","doi":"10.1145/504450.504485","DOIUrl":"https://doi.org/10.1145/504450.504485","url":null,"abstract":"DSM-CC (Digital Storage Media - Command and Control) is a very recent ISO/IEC standard developed for the delivery of interactive multimedia broadband services. This paper develops two themes of DSM-CC and explores issues related to its applicability to World Wide use. The first theme is that of Sessions and Resource control: tracking and controlling all resource usage goes against the prevailing approach of the Internet but will, we advocate, be required if quality of service is to be guaranteed effectively. The second theme is effect on distributed computing models of downloadable Information Appliances and network resource control. In this model the acquisition of an object reference also becomes the acquisition of the network resources needed to use the object's operations. Finally we suggest areas of further work.","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"143 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114665838","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":"Hinted caching in the web","authors":"J. Mogul","doi":"10.1145/504450.504470","DOIUrl":"https://doi.org/10.1145/504450.504470","url":null,"abstract":"The World Wide Web, like any practical distributed system, benefits greatly from caching. Many studies have shown relatively poor cache performance, because existing caches depend on temporal locality, and reference patterns in the Web do not have particularly high temporal locality. We can improve Web caching by exploiting spatial locality. This requires protocol changes to allow servers to provide hints to caches, both to support prefetching and to improve allocation and replacement policies.","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"168 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133176829","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":"Legion: flexible support for wide-area computing","authors":"A. Grimshaw, W. Wulf","doi":"10.1145/504450.504487","DOIUrl":"https://doi.org/10.1145/504450.504487","url":null,"abstract":"1 The Oppor tun i t y Recent technological changes have made the realization of nationwide, and indeed worldwide, virtual computers possible. A nationwide virtual computer would consist of workstations, personal computers, graphics rendering engines, supercomputers, and possibly non-traditional computing and rendering devices such as TV's, all connected via the Nil and an integrated software infrastructure. Such a system, if constructed, may well revolutionize how we work, how we play, and in general, how we interact with one another. The staggering success of the World Wide Web only hints at the potential of such a system. The technological changes that enable the realization of a worldwide virtual computer are (1) the introduction of gigabit/second fiber-optic backbones, which effectively \"shrink\" the distance between geographically separate computers, (2) the maturation of parallel compilers that support parallel execution on distributed memory machines, (3) advances in distributed systems software over twenty years which have increased our understanding of how to manage complex distributed environments, (4) the widespread acceptance in industry of the object-oriented paradigm, and (5) advances in cryptography and authentication protocols. These changes challenge the computer science community to provide a solid integrated software foundation on which to build applications, and with which to unleash the potential of so many diverse, powerful resources. The foundation must at least hide the underlying physical infrastructure from users and from the vast majority of programmers, support access, location, and fault transparency, enable inter-operability of components, support construction of larger integrated components using existing components, and provide a secure environment for both resource owners and users. Further, the services must be provided in the context of tens of thousands to millions of autonomous hosts; thus, the software infrastructure must be scalable or it will fail to meet users' performance expectations. It is this software foundation that will make the collection of hosts and fiber connections a virtual computer. We call such software metasystems software. Whether or not metasystems software is explicitly designed, the nation (and the world) will eventually end up with a metasystem. The reason is simple--both individual and organizational users will be required to deal with increasingly obvious shortcomings of the nationwide computer system. The World Wide Web and gopher show that tools will be developed in an ad hoc fashion to paper over the gaps between local systems. The issue is not whether metasystems will be developed; clearly they will. It is, rather, whether they will come about by design in a coherent, seamless system, or painfully and in an ad hoc manner by patching together independently developed systems, each with different objectives, design philosophies, and computation models.","PeriodicalId":137590,"journal":{"name":"EW 7","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129588599","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}