Recently published algorithms for matching concurrent sets of events have the problem of unbounded message queue growth if events arrive in an undesirable order. This paper presents some algorithms that mitigate this problem by examining events waiting to be processed and removing those that cannot be part of a concurrent set.

Abstract We recount and examine experiences with Hermes, an experimental language for programming distributed systems. Hermes has several unusual language features, including compile‐time checking of data initialization, representation independent data aggregates, and an integrated process model. To facilitate compile‐time initialization checking, Hermes stores data in tables and does not expose pointers. We study these features in light of the experiences of Hermes users around the world, analyzing the strengths and weaknesses of the language.

Detection of global event predicates is an important issue for distributed systems, particularly for debugging and monitoring of such systems. The paper defines event normal form (ENF) event predicates, and two online algorithms to detect the first occurrence of such predicates. These algorithms differ in the technique used to match concurrent groups of events. The first builds and searches a tree of a possible matches. The second algorithm achieves the same result as the first, but needs only a single node of the tree at any time, resulting in better space and time complexity. We outline correctness proofs, and provide performance measurements from an implementation.>

Article ENF event predicate detection in distributed systems Share on Authors: Hsien-Kuang Chiou Computer Science Dept., Polytechnic University, Brooklyn, NY Computer Science Dept., Polytechnic University, Brooklyn, NYView Profile , Willard Korfhage Computer Science Dept., Polytechnic University, Brooklyn, NY Computer Science Dept., Polytechnic University, Brooklyn, NYView Profile Authors Info & Claims PODC '94: Proceedings of the thirteenth annual ACM symposium on Principles of distributed computingAugust 1994 Pages 91–100https://doi.org/10.1145/197917.197964Online:14 August 1994Publication History 3citation172DownloadsMetricsTotal Citations3Total Downloads172Last 12 Months0Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

It is suggested that if the large numbers of idle computers and workstations in distributed systems could be used then considerable computing power could be harnessed at low cost. Such systems are analyzed using Brownian motion with drift to model the execution of a program distributed over the idle computers in a network of idle and busy processors. The ways in which the use of these transient processors affects a program's execution time is determined. The probability density of a program's finishing time on both single and multiple transient processors is found. These results are explored for qualitative insight. Some approximations for the finishing time probability density are suggested.>

Distributed systems having large numbers of idle computers and workstations are analyzed using a very simple model of a distributed program (a fixed amount of work) to see how the use of transient processors affects the program's service time. The probability density of the length of time it takes to finish a fixed amount of work is determined. An equation is given for the main result for an M-processor network. Simulations confirm that Brownian motion with drift is an accurate model of system performance. With large programs that run for a long time relative to the length of available and nonavailable periods, the central limit-theorem applies, and the Brownian-motion-with-drift model remains good regardless of the distributions of the available and the nonavailable periods. Under these assumptions, the distribution of finishing time is very tight about its mean and well approximated by a normal distribution.>