We describe a software mechanism that simplifies the implementation of real time distributed object oriented applications. The distributed observable/observer mechanism extends the Observable/Observer concept of Java to a distributed environment encapsulating the communication protocol used by networking. The mechanism supports the distributed model/view controller (M/VC) paradigm, allowing view controllers to be attached to remote models by distributed structural object composition (DSOC). DSOC simplifies the development of distributed real time applications significantly since DSOC objects need only be interconnected to interact properly with each other. We illustrate the distributed observable/observer mechanism with DSOC by examples.

article Free Access Share on Structural active object systems for simulation Authors: Toshimi Minoura View Profile , Shirish S. Pargaonkar View Profile , Kurt Rehfuss View Profile Authors Info & Claims ACM SIGPLAN NoticesVolume 28Issue 10Oct. 1, 1993 pp 338–355https://doi.org/10.1145/167962.165921Published:01 October 1993Publication History 12citation301DownloadsMetricsTotal Citations12Total Downloads301Last 12 Months12Last 6 weeks3 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 SiteeReaderPDF

It is argued that the design and analysis of a concurrent system can be made simpler and more intuitive if execution times of abstract operations are arbitrarily but systematically defined. This technique (time abstraction) is complementary to data abstraction and is more effective when used in combination with data abstraction. As examples, a bounced-buffer monitor and a multilevel concurrency scheme for a database system are analyzed by using data and time abstraction.>

This paper deals with the serializability theory for single-version and multiversion database systems. We first introduce the concept of disjoint-interval topological sort ( DITS , for short) of an arc-labeled directed acyclic graph. It is shown that a history is serializable if and only if its transaction IO graph has a DITS. We then define several subclasses of serializable histories, based on the constraints imposed by write-write, write-read, read-write, or read-read conflicts, and investigate inclusion relationships among them. In terms of DITS, we give a sufficient condition for a class of serializable histories to be polynomially recognizable, which is then used to show that a new class of histories, named WRW, can be recognized in polynomial time. We also present NP-completeness results for the problem of testing membership in some other classes. In the second half of this paper, we extend these results to multiversion database systems. The inclusion relationships among multiversion classes defined by constraints, such as write-write and write-read, are investigated. One such class coincides with class DMVSR, introduced by Papadimitriou and Kanellakis, and gives a simple characterization of this class. It is shown that for most constraints, multiversion classes properly contain the corresponding single-version classes. Complexity results for the membership testing are also discussed.

Performance of a reliable storage subsystem for a centralized database system was studied by simulation. The reliable storage subsystem studied consists of three redundant disk units that are updated one at a time from a consistent database state to another consistent database state. Thus, even if a central processor and/or one disk unit fail simultaneously, at least one disk unit will contain a consistent database state. Redundant multiple disk units allow simultaneous processing of multiple read operations. Our simulation result shows that when all transactions are read-only transactions, the throughput of a three-unit system is 2.5 times higher than that of a single-unit system. On the other hand, redundant disk units slows down write operation because each disk unit must be updated for each virtual page updated. The major result in this paper shows that this adverse effect can be mitigated if updates of multiple transactions are merged and if an efficient disk scheduling policy (e.g., CSCAN) is employed. When the ratio of write operations to read operations is between 10 to 30%, the throughput of the three-unit system is still 89 to 144% higher than that of the single-unit system, which is not reliable. This result is significant when we consider that the performance of a reliable storage subsystem based on logging cannot exceed that of a single-unit storage subsystem.

Version-based access capabilities (V-capabilities) that can be used for concurrency control of a database system are introduced. A V-capability is an access capability for a version of a data item. If transactions are allowed to access only those versions whose V-capabilities are possessed by them, consistent operation of a database system can be realized by properly controlling the issuance of V-capabilities. Since the issue times of V-capabilities can be different from their use times, V-capabilities add another dimension to the concurrency control of a database system.

Several new algorithms that utilize wmt-for relations mduced by processes are developed for deadlock avoidance in operating systems.One of the algorithms allows more concurrency than Havender's scheme and Habermann's algorithm; nonetheless, it can still be computed in polynomial time.Further, the deadlock,prediction problem for a set of processes using locks is reduced to a new NP-complete problem, and the deadlock-avoidanceproblemfor processes with branches is shown to be PSPACE-complete, Finally, the relative power of various algorithms is compared.

A new resiliency scheme for a distributed database system with replicated data is presented. One salient feature of the scheme is that it does not employ a log subsystem; hence, it can be used for a highly reliable system that must tolerate a total crash of a site. In addition, the scheme supports system partitioning without any consistency problem. The new scheme is based on a precise treatment of logical data; in particular, a precise definition of resilient system operation is given in terms of logical data.