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

{"title":"A Delay-Minimizing Service Policy With Guaranteed Bandwidth For Real-Time Traffic","authors":"P. Pancha, M. Karol","doi":"10.1109/HPCS.1995.662005","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662005","url":null,"abstract":"Future high-spccd networks that will support intcgrated video. voice and data applications are expected to utilize the concepts of ATM fast packct switching (bascd on fixcd-size packets or cells) for transpon of services. The success of these integrated packet-switched networks will dcpcnd to a largc degree on their ability to support. real-time multimedia applications. The operation of current packet-switchcd networks (for example, the Internet) are primarily based on the principles of \"best effort\" transport. Thcse principlcs. which have thcir background in the computer networking origins of these networks, do not provide concrete guarantces for the quality of service (QoS) that is delivered to an application and are therefore inadequate for realtime applications. As a result. one of the main challenges in designing high-speed packet-switched networks is to devise transport tcchniqucs that will guarantce and deliver the required QoS for traffic flows. In thc design of rcal-timc transpon mcchanisms it is necessary to detemine queueing and scheduling algorithms that support specification and control of QoS paramcms for real-time applications. In this context, for real-time applications, the two QoS parameters that must be spccificd (and controlled to some degree) are the maximum delay (where delay is measured from the time a cell arrives lo a qucuc to whcn it begins service) and the delay jitter. It is well known that the round robin (RR) scheduling schemc and its variants for cxamplc. hierarchical round robin (HRR) [ I ] . fair queueing (FQ) [2][3]. packet-by-packet generalized proccssor sharing (PGPS) 141. and self-clocked fair queueing (SCFQ) [ 5 ] can all guarantee a minimum bit rate for an application undcr any traffic condition. It can also be shown that this minimum bit rate guarantee is equivalent to a guaranteed maximum dclay whcn an application is \"well behaved (e.g., there is a bound on the number of cells arrivals in any time period). For cxamplc. in 14) an cxprcssion for the bound on maximum delay of the PGPS scheme is derived for a traffic pattern that conforms to thc paranctcrs of a lcaky bucket filter. Similar bounds for the maximum delay are possible with many of the other fair qucucing schcmc if thc flow is the output of a leaky bucket filter. We can also achieve a maximum delay bound with the service policy introduccd in this papcr. While schemes that guarantcc maximum delay bounds are relatively well known, the control of delay jitter in networks is less straightforward. In this papcr. wc use two statistical measures for jitter that are similar in spirit to the maximum delay difference. These measures. thc standard dcviation and the 99.9 percentile for cell delay, correspond to measures of the width and the tail of thc dclay dcnsity function. rcspcctively. and are statistical indicators for the magnitude of delay jitter. We use these performance measures as the basis for comparison of delay jitter performance between the diffe","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"29 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":"132349339","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":"B(its) T(o) The U(ser): A Communication Benchmark","authors":"K. Malý, A. Gupta, S. Mynam, S. Khanna","doi":"10.1109/HPCS.1995.662012","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662012","url":null,"abstract":"Our benchmark does not test the maximum performance of a workstation in a homogeneous network under ideal circumstances. Rather, we take a vendor supplied workstation running UNIXfootnote{TM of USL, Inc.} operating system, install our benchmark program and connect the workstation to a blackboxtestbed which emulates a LAN connected to a WAN. The benchmark run will submit the workstation to a carefully designed combination of tests. The result is an indicator of what the user, at the application level, can expect in terms of bits sent to or received from a remote host. Our benchmark, in contrast to maximum performance tests, takes into account concurrent activities, such as CPU and I/O activities, which compete for resources on the test machine and concurrent activities on the network which will interfere with the test machine''s communication. On the one hand we intend these results for the workstation buyer who can use the data together with the configuration specification and the list price to make a well reasoned decision if the workstation meets the needs when compared to the results of other workstations. We also believe these results can provide insights to the architects as to the location and nature of the communication bottlenecks of a particular workstation. We present the results of applying the BTU benchmark to four workstations from different vendors which shows that neither price nor SPECmarks are a good predictor for communication performance as measured by BTUs. We conclude this paper with a discussion of our plans for having the community accept this proposal as a communication benchmark.","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133078539","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":"Dynamic Server Selection In The Internet","authors":"M. Crovella, R. Carter","doi":"10.1109/HPCS.1995.662022","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662022","url":null,"abstract":"Abstract As distributed information services like the World Wide Web become increasingly popular on the Internet, problems of scale are clearly evident. A promising technique that addresses many of these problems is service (or document) replication. However, when a service is replicated, clients then need the additional ability to find a ``good'''' provider of that service. In this paper we report on techniques for finding good service providers without a priori knowledge of server location or network topology. We consider the use of two principal metrics for measuring distance in the Internet: hops, and round-trip latency. We show that these two metrics yield very different results in practice. Surprisingly, we show data indicating that the number of hops between two hosts in the Internet is {em not/} strongly correlated to round-trip latency. Thus, the distance in hops between two hosts is not necessarily a good predictor of the expected latency of a document transfer. Instead of using known or measured distances in hops, we show that the extra cost at runtime incurred by dynamic latency measurement is well justified based on the resulting improved performance. In addition we show that selection based on dynamic latency measurement performs much better in practice that any static selection scheme. Finally, the difference between the distribution of hops and latencies is fundamental enough to suggest differences in algorithms for server replication. We show that conclusions drawn about service replication based on the distribution of hops need to be revised when the distribution of latencies is considered instead.","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132128501","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":"Providing Predictable Performance To A Network-Based Application Over ATM","authors":"S. Dharanikota, K. Maly, B. Kvande, C. M. Overstreet, R. Mukkamala","doi":"10.1109/HPCS.1995.662025","DOIUrl":"https://doi.org/10.1109/HPCS.1995.662025","url":null,"abstract":"Currently, applications which use an underlying network can produce highly varied performance for different end-to-end protocols, and at different CPU, and network load conditions. With the proliferation of distributed and collaborative applications, the predictability in an application performance takes on more importance. In this work we attempt to provide predictable performance to different classes of applications.","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126450594","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 Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems (HPCS'95)","authors":"","doi":"10.1109/HPCS.1995.661999","DOIUrl":"https://doi.org/10.1109/HPCS.1995.661999","url":null,"abstract":"","PeriodicalId":174670,"journal":{"name":"Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems","volume":"29 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126989155","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}