The realization of end-to-end quality of service (QoS) guarantees in emerging network-based applications requires mechanisms that support first dynamic discovery and then advance or immediate reservation of resources that will often be heterogeneous in type and implementation and independently controlled and administered.We propose the Globus Architecture for Reservation and Allocation (GARA) to address these four issues.GARA treats both reservations and computational elements such as processes, network flows, and memory blocks as first class entities, allowing them to be created, monitored, and managed independently and uniformly.It simplifies management of heterogeneous resource types by defining uniform mechanisms for computers, networks, disk, memory, and other resources. Layering on these standard mechanisms, GARA enables the construction of application-level co-reservation and coallocation libraries that applications can use to dynamically assemble collections of resources, guided by both application QoS requirements and the local administration policy of individual resources.We describe a prototype GARA implementation that supports three different resource types— parallel computers, individual CPUs under control of the Dynamic Soft Real-Time scheduler, and Integrated Services networks—and provide performance results that quantify the costs of our techniques.

Although the typical process-layout manufacturing environment is susceptible to deadlocks, the problem of deadlock resolution in this context has only lately been undertaken by the scientific community. Previous studies have found that deadlock avoidance methodologies seem to be the most appropriate for this particular context. Unfortunately, in the general case, these methods suffer from high computational complexity which results in heuristic solutions and/or reduced performance. Taking the position that any solution to the problem should be scalable and provably correct, this paper proposes an analytical framework for designing deadlock avoidance policies for a subclass of Resource Allocation Systems (RAS). Specifically, this subclass is characterized by the fact that jobs in the system are defined by deterministic job-step sequences with every step in the sequence requiring a single unit of the system resources. Job-step models are appropriate for the study of the deadlo...

Modern production facilities (i.e., flexible manufacturing systems) exhibit a high degree of resource sharing, a situation in which deadlocks (circular waits) can arise. Using digraph theoretic concepts we derive necessary and sufficient conditions for a deadlock occurrence and rigorously characterize highly undesirable situations (second level deadlocks), which inevitably evolve to circular waits in the next future. We assume that the system dynamics is described by a discrete event dynamical model, whose state provides the information on the current interactions job-resources. This theoretic material allows us to introduce some control laws (named restriction policies) which use the state knowledge to avoid deadlocks by inhibiting or by enabling some transitions. The restriction policies involve small on-line computation costs, so they are suitable for realtime implementation. For a meaningful class of systems one of these policies is the least restrictive deadlock-free policy one ca...

Parallel and distributed programming languages often include explicit synchronization primitives, such as rendezvous and semaphores. Such programs are subject to synchronization anomalies; the program behaves incorrectly because it has a faulty synchronization structure. A deadlock is an anomaly in which some subset of the active tasks of the program mutually wait on each other to advance; thus, the program cannot complete execution. In static anomaly detection, the source code of a program is automatically analyzed to determine if the program can ever exhibit a specific anomaly. Static anomaly detection has the unique advantage that it can certify programs to be free of the tested anomaly; dynamic testing cannot generally do this. Though exact static detection of deadlocks is NP-hard [Tay83a], many researchers have tried to detect deadlock by ...

Deadlock is an increasingly pressing concern as the multicore revolution forces parallel programming upon the average programmer. Existing approaches to deadlock impose onerous burdens on developers, entail high runtime performance overheads, or offer no help for unmodified legacy code. Gadara automates dynamic deadlock avoidance for conventional multithreaded programs. It employs whole-program static analysis to model programs, and Discrete Control Theory to synthesize lightweight, decentralized, highly concurrent logic that controls them at runtime. Gadara is safe, and can be applied to legacy code with modest programmer effort. Gadara is efficient because it performs expensive deadlock-avoidance computations offline rather than online. We have implemented Gadara for C/Pthreads programs. In benchmark tests, Gadara successfully avoids injected deadlock faults, imposes negligible to modest performance overheads (at most 18%), and outperforms a software transactional memory system. Tests on a real application show that Gadara identifies and avoids both previously known and unknown deadlocks while adding performance overheads ranging from negligible to 10%. 1

Abstract not found

Introspection or the ability to observe one's own behavior is one of the most powerful capabilities of human intelligence; it is the basis for understanding and improvement of one's behavior and of human progress. Similarly, introspective computer systems, introduced in this thesis, examine, reason about, and change their own behavior in powerful new ways. Because the complexity of computers is rapidly increasing, yet is restricted by limited human resources, the most attractive quality of introspective computers is their ability to manage this growing complexity themselves. Self-managing computer systems would greatly expand the rational power and complexity of computer systems that can be successfully built. The main difficulty in constructing introspective computer systems is enabling the system to obtain a description of its complete behavior in a dynamic and unobtrusive way. This thesis proposes the partition of the system into two threads of control. The first thread performs the...

The common features of third generation operating systems are surveyed from a general view, with emphasis on the common abstractions that constitute at least the basis for a "theory " of operating systems. Properties of specific systems are not discussed except where examples are useful. The technical aspects of issues and concepts are stressed, the nontechnical aspects mentioned only briefly. A perfunctory knowledge of third generation systems is presumed. Key words and phrases: multiprogramming systems, operating systems, supervisory systems, time-sharing systems, programming, storage allocation, memory allocation, processes, concurrency, parallelism, resource allocation, protection CR categories: 1.3, 4.0, 4.30, 6.20 It has been the custom to divide the era of electronic computing into "generations" whose approximate dates are:

Abstract. We study the problem of thread allocation in asynchronous distributed real-time and embedded systems. Each distributed node handles a limited set of resources, in particular a limited thread pool. Different methods can be invoked concurrently in each node, either by external agents or as a remote call during the execution of a method. In this paper we study thread allocation under a WaitOnConnection strategy, in which each nested upcall made while a thread is waiting must be made in a different thread. We study protocols that control the allocation of threads to guarantee the absence of deadlocks. First, we introduce a computational model in which we formally describe the different protocols and their desired properties. Then, we study two scenarios: a single agent performing sequential calls, and multiple agents with unrestricted concurrency. For each scenario we present (1) algorithms to compute the minimum amount of resources to avoid deadlocks, and (2) run-time protocols that control the allocation of these resources. 1

Currently, conflict-free routing in AGV systems is established by means of one of the following three approaches: (i) the problem elimination through the adoption of a segmented path flow or tandem queue configuration, (ii) the identification of imminent collisions through forward sensing and their aversion through vehicle backtracking and/or rerouting, or (iii) the imposition of zone control and extensive route pre-planning, typically based on deterministic timing of the vehicle traveling and docking stages. Among these three approaches, the segmented path flow-based approach presents the highest robustness to the system stochasticities/randomness, but at the cost of restricted vehicle routings and the need for complicated handling operations. This paper proposes an alternative conflict resolution strategy that will ensure robust AGV conflict resolution, while maintaining the operational flexibility provided by free vehicle travel on arbitrarily structured guidepath networks. Specific...