{"title":"Genetic allgorithm for dynamic task scheduling","authors":"M. D. Kidwell, D. Cook","doi":"10.1109/PCCC.1994.504094","DOIUrl":"https://doi.org/10.1109/PCCC.1994.504094","url":null,"abstract":"A genetic dynamic scheduling algorithm (GDSA) is applied to the problem of dynamically scheduling multiple tasks. These independent, non- identical tasks must be distributed in a multiprocessor, shared memory system. This problem is known to be NP-complete. The use of a genetic algorithm for dynamic task scheduling offers several advantages to alternative algorithms. A genetic algorithm fids a solution immediately, at initialization. All processing time is devoted to improvement of the solution. This makes them well suited for dynamic applications, where the amount of processing time is likely to vary. Genetic algorithms tend to make the greatest improvements during initial generations. They move quickly to a good solution, although later improvements will become increasingly less fresuent and less dramatic. Schedule makespans using the GDSA are compared to makespans when a first-come, fxst-serve dynamic scheduling approach is used. This opportunistic scheduling algorithm places arriving tasks in a FIFO queue, and pmessors remove tasks from the queue as they become available. Tests were run to schedule 30 and 200 tasks. When 30 tasks were scheduled, the GDSA caused small, but consistent schedule improvements. The GDSA was most effective if the execution times of the tasks were uniformly distributed. However, schedule improvements were also noted with normally and exponentially distributed task execution times. When 200 tasks were scheduled, the GDSA was most effective if the execution times of the tasks were exponentially distributed.","PeriodicalId":203232,"journal":{"name":"Proceeding of 13th IEEE Annual International Phoenix Conference on Computers and Communications","volume":"775 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126631897","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":"Fast Bandwidth Access for Multimedia Traffic Classes Under QOS Constraints","authors":"S. Chatterjee, M. Bassiouni","doi":"10.1109/PCCC.1994.504100","DOIUrl":"https://doi.org/10.1109/PCCC.1994.504100","url":null,"abstract":"Some of the applications for future high-speed networks Several types of connections with widely different bandwidth and delay requirements and a variety of traffic types including voice, video and multimedia data is expected to co-exist in ATM based Broadband-ISDN networks. One viable approach has been to classify traffic types with similar QOS and try to meet their performance requirements. Can the introduction of traffic classes affect the per connection guarantees and their performance? We investigate this issue by providing a framework for efficiently integrating a number of traffic classes while devising schemes to meet the QOS requirements of admitted traffic. We introduce the concept of potential function of a class which is a function that reflects its bandwidth requirements, utilization level changing with time of day and a parameter CY (which might determine the priority of the class) based on which bandwidth is dynamically allocated. We also discuss the operation of a multifunctional scheduler that services each class with a discrete quanta of service determined by its potential function and tries to meet specific class QOS requirements. The scheme is an efficient way of combining constant bit-rate (CBR) services with variable bit-rate (VBR) statistically multiplexed services. The feasibility of this novel bandwidth allocation and scheduling scheme is validated by extensive simulation","PeriodicalId":203232,"journal":{"name":"Proceeding of 13th IEEE Annual International Phoenix Conference on Computers and Communications","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126190403","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":"Adaptive breadth-first search protocols","authors":"Gurdip Singh","doi":"10.1109/PCCC.1994.504124","DOIUrl":"https://doi.org/10.1109/PCCC.1994.504124","url":null,"abstract":"We present adaptive protocols for the problem of breadth-first search in a distributed system. An adaptive protocol can change its behavior in response to changes in the network environment, which might be useful to satisfy certain performance requirements and to enhance the life cycle of a software system. We obtain an adaptive protocol by combining two existing breadth-first search protocols, A and B. A performs better than B if network topology is sparse or if the actual message delays on different links during the execution of the protocol are the same. 0therwise, it is more costly to operate than B. Protocol Comp(A, B), which we obtain by composing A and B, performs well under a variety of network conditions. It behaves like A whenever conditions are favorable and like B otherwise. Its message complexity on any topology is the minimum of the message complexities of A and B on that topology. Using the same technique, we obtain another adaptive protocol by combining A with another protocol C. The performance characteristics of this protocol are similar to those of Comp(d, B). Protocols obtained using our technique exhibit an additional property of local adaptivity i.e., the protocols adapt themselves locally to changes in the network.","PeriodicalId":203232,"journal":{"name":"Proceeding of 13th IEEE Annual International Phoenix Conference on Computers and Communications","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115152319","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 Guarantees on ATM Networks","authors":"Chengyao Tang, S. Murthy, D. Long","doi":"10.1109/PCCC.1994.504101","DOIUrl":"https://doi.org/10.1109/PCCC.1994.504101","url":null,"abstract":"Recent developments in ATM technology has made multiplexing of a wide range of traffic with diverse performance requirements important. The statistical multiplexing of several traffic types such as voice, video and data can lead to network congestion thus violating the quality of service (QOS) guarantees. In order to meet the performance requirements of the applications an efficient transport protocol and congestion control mechanism becomes necessary. We propose a transport mechanism for guaranteeing application required QOS requirements over an ATM network. Leaky-bucket congestion control scheme is used as the traffic policing mechanism. Source characterization of voice and video are modeled as Markov Modulated Poisson Processes (MMPP) and the proposed transport mechanism is evaluated by simulation on a single and multihop network with multimedia traffic. The performance of the scheme is examined as a function of source characteristics and the effect of statistical multiplexing is also investigated. The results indicate that the leaky-bucket scheme and statistical multiplexing are more effective for bursty traffic. This calls for a dynamic control mechanism to achieve optimal system utilization with guaranteed QOS.","PeriodicalId":203232,"journal":{"name":"Proceeding of 13th IEEE Annual International Phoenix Conference on Computers and Communications","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132502972","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}