Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems最新文献

{"title":"Distributed Explicit Rate Allocation","authors":"A. Charny, K. Ramakrishnan","doi":"10.1109/HPCS.1995.662007","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662007","url":null,"abstract":"The problem of congestion niaiiagement has been a particularly active area of research for the last decade. Networks are getting faster. arc more richly interconnected and are growing in scale everyday. The larger bandwidthdelay products for these high speed networks poses a continuing challenge for flow and congestion management algorithms. Traditional work i n congestion control and avoidance algorithms has been to layer them on top of window flow control. which is an end-to-end mechanism to assure no loss in the receiving end-system. When used in the context of the large bandwidth-delay product seen in high speed networks, window based flow control mechanisms result in the use of large windows and the potential for large bursts of packets arriving at intermediate nodes. This requires significant additional buffering that is often considered undesirable. One of the changes for congestion management in high speed networks is the use of rate based congestion/flow control schemes. The demarkation between flow and congestion control also tends to be blurred with the use of rate based schemes. However, using rate based mechanisms results in a fresh set of problems to solve in the congestion control and avojdance domain. Another fundamental change over the same time frame is the development of switched networks, exemplified by ATM. The issue of dealing with congestion in connection oriented networks, while apparently easier, is complicated by the need to address qualities of service. A lot of the recent work in congestion control has been in the context of ATM flow and congestion control. M u c h of the work has been in the context of the ABR (Available Bit Rate) service. These typically arc used for data transfer, and are considered to be lower priority than the other classes of service, CHR and VIjH. Left over bandwidth from the use of a resource (link, switch or end-system) by CBR and VBR is used for AHR service. The timely use of the available resources by ABR without causing congestion (which results either in significantly increased queueing delays or cell loss) is an important goal. Another equally important goal is to ensure that the resources in the network are shared in a ‘fair and equitable’ manner. A definition of ‘fairness‘ is the notion of max-min fairness, which has been adopted in the ATM Forum for congestion management of the ABIi service. A simplistic notion of inax-min fairness viewed by a flow at a local resource is: If my flow is limited at this resource for lack of adequate capacity. then there should be no other flow passing through this resource that will be given a greater share of the resource’s capacity. There are several ways of computing the max-min fair allocation for a flow in a network of multiple resources and multiple flows with different resource demands [l], [ 7 ] , [8], [12],","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116238214","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":"A Compiler-Based Approach To Protocol Optimization","authors":"C. Castelluccia, P. Hoschka","doi":"10.1109/HPCS.1995.662010","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662010","url":null,"abstract":"The Hipparcli project [CasSlb] uses the Esterel language [Dio95] for describing protocol building blocks. These blocks cui be configured to form an application specific protocol implementation. Using application specific knowledge should improve the performance of such ai iniplciiiciitation when compared to the use of a standard protocol. However, the C code generated by tlie currently existing Esterel compiler is not very fast, and moreover has a largc code size lCas94aI. This is a problem the compiler shares with many similar tools that gencratc executable code starting from a protocol defined in a formal description technique (FDT). This paper describes our design for an optimizer that we incorporated into the Esterel compiler. After analyzing tlic code generated by the Esterel compiler, we decided that two optiiiiizatioii techniques nierit further investigation:","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114910116","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":"Event-Signaling Within Higher Performance Network Subsystems","authors":"J. Chung, C. Brendan, S. Traw, J.M. Smith","doi":"10.1109/HPCS.1995.662033","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662033","url":null,"abstract":"The afterburner ATM link Adapter has allowed us to evaluate three event-signaling schemes: polling, traditional interrupts and the clocked interrupts first investigated in our operating system work in AURORA. The schemes are evaluated in the context of a single-copy TCP/IP stack. The experimental results indicate that clocked interrupts can provide throughput comparable with traditional interrupts for dedicated machines (up to over 144 Mbps, the highest TCP/IP/ATM throughput reported), and better performance when the machines are loaded with an artificial workload. Polling, implemented to be used with an unmodified netperf measurement tool, was competitive for small TCP/IP socket buffer sizes (i32KB). We concluded that clocked interrupts may be preferable for applications requiring high throughput on systems with heavy processing workloads, such as servers. Comments University of Pennsylvania Department of Computer and Information Science Technical Report No. MSCIS-96-15. This technical report is available at ScholarlyCommons: http://repository.upenn.edu/cis_reports/188 Event-signaling within Higher Performance Network Subsystems Jeffrey D. Chung C. Brendan S. Traw Jonathan M. Smith University of Pennsylvania School of Engineering and Applied Science Computer and Information Science Department Philadelphia, PA 19104-6389 Event-signaling within Higher Performance Network Subsystems Jeffrey D. Chung, C. Brendan S. Traw and Jonathan M. Smith* University of Pennsylvania {jdchung , traw , jms)@cis. upenn. edu","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132721654","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 Superhighway Enters The Manufacturing World","authors":"J. Guillaud, M. R. Pokam, G. Michel","doi":"10.1109/HPCS.1995.662031","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662031","url":null,"abstract":"","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"190 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121729466","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":"Performance Issues Of A Digital Video-on-Demand Gateway","authors":"C. Bisdikian, F. Schaffa","doi":"10.1109/HPCS.1995.662026","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662026","url":null,"abstract":"","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122327833","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":"A Multimedia Synchronization Protocol For High Speed Networks","authors":"Shi-Kyu Bae, K. Han","doi":"10.1109/HPCS.1995.662017","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662017","url":null,"abstract":"In this paper, we propose an MASP(Multimedia Association Synchronization Rotocol) to transmit multimedia data with a temporal relationship on general networks not guranteeing a constant delay time. As existing synchronization schemes have been proposed for future network not to be implemented or for specific environments to their applications, they are difficult to be implemented in general network such as packet networks or Internet which is composed of heterogeneous networks. Our protocol adopts a model which has group connections called associations between two hosts. The transmitter part schedules transmission times considering the delay time of each transmission path for packet to arrive on time. The receiver part playouts the packet with intermedia synchronization function, if necessary. T h e protocol frequently measures delay on the transmission path for exact scheduling, and provides a playout capability in environments where no global clock is maintained.","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126461978","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":"Multi-Subsystem Protocol Architectures: Motivation And Experience With An Adapter-Based Approach","authors":"B. Krupczak, M. Ammar, K. Calvert","doi":"10.1109/HPCS.1995.662030","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662030","url":null,"abstract":"Protocol software can be difficult, cumbersome, and expensive to develop and test. In practice, several techniques are used to reduce this complexity. First. protocol code is subdivided into layers or modules and organized into a graph structure representing how those modules combine to provide communication services. Second, protocols are developed within programming environments or subsystems (such as BSD [4], Streams[l. 81, or thex-Kemel[6,23) which providean overall protocol model and offer organized, consistent access to operating system resources. Third, protocol code is often ported from one subsystem to another rather than developed from scratch.","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122552323","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":"A Hierarchical Multicast Routing Protocol","authors":"I. Akyildiz, W. Yen","doi":"10.1109/HPCS.1995.662018","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662018","url":null,"abstract":"Multimedia applications will become more and more important and popular in the future networks [5]. Many multimedia applications including electronic newspaper and remote education require point to multipoint communication instead of point to point communication. The point to multipoint comniuniration is refered as multicast in this paper. The problem of multicast routing is considered as a graph theoretical optimal delivery tree problem. Several highly complex multicast routing algorithms have been proposed along these lines in recent years. These algorithms require intensive computation and message exchanges; thus they are unsuitable for practical protocol implcmciit ation. Kxisting multicast routing protocols were designed to trade optimality for time efliciency [2. 3. 61. Still, the existing protocols have either scalability or performance problems when applied i n the wide area network environment [l, 41. In this paper, a hierarchical multicast routing ( I I M R ) algorithm is proposed which uses the concepts of clusters and local cores to address tlicscl problems. This algorithm is simple, yet flexible enough to provide various routing modes to satisfy different users' delay and scalability requirements. A set of protocols is designed to implement the algorithm. These protocols (a) are independent of the underlying unicast protocol; ( b ) exhibit low complexity and good scalability; (c) provide the upper bound for the delay performance; a n d ( d ) support dynamic membership change. Through simulations, the performance of the new protocol is compared to that of existing protocols and its advantages are demonstrated.","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131230109","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":"A Reflective Object-Oriented Distributed System For Heterogeneous Multimedia Environments","authors":"Larry T. Chen, Leslee Xu, T. Suda, Tetsuya Yamamoto, K. Obinata","doi":"10.1109/HPCS.1995.662003","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662003","url":null,"abstract":"In this paper, object-oriented techniques and concepts (e.g. object migration, inheritance, delegation) and reflection are applied to the design of a distributed system called the Reflective Object-oriented Distributed System (RODS). RODS hides object distribution and underlying platform heterogeneity to simplify the development of distributed multimedia applications. RODS has two major components: The Flexible Multimedia Object Adaptor (FMOA) and the Reflective Object Request Broker (RORB). The FMOA is an application framework of classes to facilitate the construction of distributed multimedia applications. The RORB is a CORBA Object Request Broker that integrates additional features such as smart proxies, delegation, migration, and sharing. By utilizing a reflective model, RORB allows the application programmer and the system to dynamically modify distributed object management policies.","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130134572","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":"Delivering High-Performance Communication To The Application-Level","authors":"W. Vogels, T. V. Eicken","doi":"10.1109/HPCS.1995.662011","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662011","url":null,"abstract":"A crucial part of the futurc succcss of the Information Superhighway will bc the ability lo deliver the technological capabilities of high-speed networks to the application level. Electronic w orkplaccs w ill be di recti y connected through high-speed local area network to gigabit backbones providing global connectivity. It seems very attractive to build these workplaces out of standard available workstations, which, combined with off-theshelf communication hardware and running a regular operating system. will need to be capable of handling the required high-bandwidth and low-latency communication. Experiences with these standard environments have been very disappointing until now, especially with the focus on the low-latency delivery of small messages that play such an important role in state-of-the-art distributed systcms. This paper prcscnts an analysis of the current problems, describes an architccturc (U-Net) that is able to overcome these problems by implcmcnting the communication system at uscr-level, and reports on the performance of a prototype implemcnlation on an off-the-shelf hardware platform (SparcStations with Fore Systems ATM interfaces) running a standard operating system. The &Net communication architecture1 11 virtualizes the nctwork device so that each process has the illusion of owning the interface to the network. hotection is assured through kernel control of the setup and teardown of connections. The architecture is able to support both. legacy protocols and novel networking abstractions: TCP and UDP a$ wcll as Active Messages are implemented and exhibit pcrformancc that is only limited by the processing capabilities of the network interface","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125936862","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}