A new look at an old protocolSyverson, Paul F.
doi: 10.1145/230908.230911pmid: N/A
Analyses of the Needham-Schroeder protocol using the logics SVO and BAN are discussed. It is shown that the protocol may fail to meet goals derived in the BAN89 analysis of it. The features of SVO analysis revealing the limitations on appropriateness of these goals are indicated.
VP: a new operation for semaphoresTai, K. C.; Carver, Richard H.
doi: 10.1145/230908.250717pmid: N/A
A semaphore is a non-negative integer variable on which only operations P and V are allowed. The semaphore construct (with various extensions) is supported by many operating systems and is often used to implement other synchronization constructs. Since the semaphore construct is low-level, a semaphore-based program may contain synchronization errors that are very difficult to detect. In this paper, we propose a new operation VP (s1,s2), where s1 and s2 are distinct or identical semaphores. An execution of VP(s1,s2) by a process T is equivalent to that of "V (s1); P(s2)" except that when T starts the V (s1) operation, T is guaranteed to be the next process to access s2 (i.e., T will execute the P(s2) operation before another process executes a P(s2) or V(s2) operation.) Section 2 shows several examples to illustrate the advantages of the VP operation. Section 3 concludes this paper. In this paper, we assume that the queue for a semaphore is first-in-first-out and that the order of processes blocked in a semaphore queue is the same order in which they executed P operations on the semaphore. We also assume that the scheduling of processes in a concurrent program is fair, i.e., each process ready for execution will eventually be selected as the running process. Programs in this paper use Ada-like syntax.
Scheduling PVM tasksJu, Jiubin; Wang, Yong
doi: 10.1145/230908.230914pmid: N/A
This paper describes a PVM task scheduler designed and implemented by the authors. The scheduler supports selecting idle workstations, scheduling pool tasks and dynamically produced subtasks. It can improve resource utilization, reduce job response time and simplize programming.
Joint actions based authorization schemesVaradharajan, Vijay; Allen, Phillip
doi: 10.1145/230908.230915pmid: N/A
Authorization policy requirements in commercial applications are often richer compared to military applications in terms of the types of privileges required, and more complex in terms of both the nature and degree of interactions between participating objects. Delegation and joint action mechanisms allow a more flexible and dynamic form of access control, thereby enabling the representation of sophisticated authorization policies. This paper explores some issues that need to be addressed when designing such joint actions based authorization policies. We describe some approaches to supporting joint actions based authorization policies, and their ramifications for trust of various components of the implementation. We consider an example from the medical field, and define attributes relevant to the design of joint action schemes and present three schemes for supporting joint action based authorization policies.
Distributed mutual exclusion on hypercubesNaimi, Mohamed
doi: 10.1145/230908.230917pmid: N/A
Hypercube have been commercially available in the past few years to their high degree of connectivity, symmetry, and low degree of diameter. In this paper, we analyse the performance in number of messages on d-dimensional hypercube, for two groups of distributed algorithms for mutual exclusion, a permission-based mutual exclusion group, and a token-based mutual exclusion group.In the permission-based mutual exclusion algorithm, a node enters in the critical section only after receiving permission from all others nodes, this algorithm requires d2 d messages.In the token-based mutual exclusion algorithm, a node is allowed to access its critical section if and only if it holds the token. In this algorithm, there is a node, called root, which knows the last node to get the token among the current requesting nodes. When a node wants to enter critical section, it sends request message to the root, which in turn informs the last node that new node will get the token next, and updates its last node. As result, the requesting nodes form a distributed queue, each of which records only the element next to it, this algorithm requires 2d messages in the worst case.
Decidability of a class of dual communicating finite state machinesMountassir, H.
doi: 10.1145/230908.230920pmid: N/A
A number of techniques have been proposed to validate systems of communicating machines which exchange exclusively messages through FIFO channels. The reachability analysis is a well-known and most popular technique implemented in several tools of validation. Two majors questions are discussed : the finiteness of the reachability graph and the state explosion problem. In this paper we clarify a restricted class of dual machines and its applicability of real protocols. Our investigation at the first time concern sufficient conditions to ensure the finiteness of the reachability graph. In the second part of the paper we confirm the results using the fair reachability analysis in which global states are generated by forcing the two machines in equal speeds.
Real-time scheduling of multimedia data retrieval to minimize buffer requirementTsai, Wen-Jiin; Lee, Suh-Yin
doi: 10.1145/230908.230921pmid: N/A
Continuous display is an important issue in the domain of multimedia applications. Especially, to ensure this continuity in the presence of multiusers, a feasible scheduling algorithm is prerequisite for real time data retrieval from the I/O subsystem. I/O scheduling techniques can be classified into two types: meta-I/O scheduling which arranges the sequence of data retrieval before issuing physical I/O requests, and disk scheduling which determines the order of processing I/O requests that have been issued. In disk scheduling, there are several elegant algorithms that had been discussed such as Scan, C-Scan, shortest seek time first and Scan-EDF. All of them focused on improving I/O throughput by serving requests closer to disk head first 78. We focus this paper, however, on solving the real time meta-I/O scheduling.For real-time scheduling, several algorithms had been addressed such as earliest-deadline-first (EDF) 4, least-laxity-first (LLF) 9, earliest-ready-time first (LRF) 6, and so on, which had shown to be elegant for task scheduling to promote system throughput. When applying to meta-I/O scheduling, however, these algorithms would result in large amount of buffer requirement for accommodating the retrieved data. In this paper, we proposed two real-time algorithms and a technique, called object migration, to minimize buffer requirement for meta-I/O scheduling. A buffer measurement approach was also proposed in this paper to estimate the performance of a real-time scheduling algorithm, which is based upon the well-known graph coloring technique. Simulation experiments were conducted to analyze the performance of algorithms. The results indicate that our approaches perform much better than existing real-time algorithms in terms of reducing buffer requirement.