Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 11

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  rozproszony system komputerowy
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
EN
There is suggested a method for the trust level establishment to the nodes of distributed computer systems (DCS) taken into account the dynamics of the information value changing and with the in-time records of the security incidents from the nodes. The proposed method of the DCS nodes rating establishment allows adaptively and during the DCS functioning to determine a safety configuration of resources (nodes) for the information processing in the DCS. Also, there is described the specifics of the implementation of the mechanisms for the nodes trust level establishment.
PL
W artykule przedstawiono metodę określenia poziomu zaufania w przypadku węzłów w rozproszonych systemach komputerowych. Bierze się w niej pod uwagę dynamikę informacji, w celu zapewnienia bezpieczeństwa węzłów. Proponowana metoda oceny węzłów pozwala określić konfigurację zasobów bezpieczeństwa dla przetważania informacji w rozproszonych systemach komputerowych. Ponadto, opisano specyfikę wdrażania mechanizmów ustanowienia węzłów dla poziomu zaufania.
EN
Introspection is the prerequisite of autonomic behavior, the first step towards performance improvement and resource usage optimization for large-scale distributed systems. In grid environments, the task of observing the application behavior is assigned to monitoring systems. However, most of them are designed to provide general resource information and do not consider specific information for higher-level services. More precisely, in the context of data-intensive applications, a specific introspection layer is required to collect data about the usage of storage resources, data access patterns, etc. This paper discusses the requirements for an introspection layer in a data management system for large-scale distributed infrastructures. We focus on the case of BlobSeer, a large-scale distributed system for storing massive data. The paper explains why and how to enhance BlobSeer with introspective capabilities and proposes a three-layered architecture relying on the MonALISA monitoring framework. We illustrate the autonomic behavior of BlobSeer with a self-configuration component aiming to provide storage elasticity by dynamically scaling the number of data providers. Then we propose a preliminary approach for enabling self-protection for the BlobSeer system, through a malicious client detection component. The introspective architecture has been evaluated on the Grid'5000 testbed, with experiments that prove the feasibility of generating relevant information related to the state and behavior of the system.
EN
A linear quadratic optimal control problem for a class of discrete distributed systems is analyzed. To solve this problem, we introduce an adequate topology and establish that optimal control can be determined though an inversion of the appropriate isomorphism. An example and a numerical approach are given.
PL
W artykule przedstawiono nową metodę szeregowania wiadomości. W zaproponowanej metodzie (Dynamic Priority Group Scheduling) szeregowanie wiadomości odbywa się w oparciu o dynamicznie przydzielany priorytet dla grup wiadomości. Opracowany został także wariant omawianej metody umożliwiający szeregowanie wiadomości z mieszanym przydziałem priorytetu (Mixed Priority Group Scheduling). Przedstawiono sposób sprawdzania dotrzymywania warunków RT dla metod DPGS i MPGS. Dla trzech wybranych magistral miejscowych (PROFIBUS FMS, CAN, InterBus-S) przeprowadzono badania, mające na celu porównanie wiadomości opracowanych metod z metodami GRMS i EDF.
EN
In this paper new methods of messages scheduling are presented. The first method, Dynamic Priority Group Scheduling method is based on assigning dynamic priorities not to individual message, but to group of messages. The second method, Mixed Priority Group Scheduling method enable to assign to group of messages static priorities as well. A procedure of checking RT-constraint fulfilment of messages in distributed real time system based on fieldbus for DPGS and MPGS methods are presented. For three fieldbuses PROFIBUS FMS, CAN and InterBus tests were performed in order to compare properties of created methods to GRMS and EDF methods. New methods take into consideration specificity of distributed real time system based on fieldbus and adapt message deadlines to time properties of distributed system. It is very importent that new methods have less calculating time than GRMS and EDF methods. Tests which were performed give results as follow: * Performances of new methods depend on message model parameters, but obtained resuts have never been worst than results for GRMS method, * In the case, when results are almost the same as the results for EDF method, and it takes into consideration the smallest calculating time of new methods, it can say that new methods are most effectively.
PL
W artykule przedstawiono właściwości i prametry czasowe trzech wybranych magistral miejscowych: PROFIBUS FMS, CAN i InterBus-S. Prezentowane wyniki uzyskano na podstawie badań, polegających na sprawdzeniu dotrzymania ograniczeń czasowych dla wiadomości w rozproszonym systemie czasu rzeczywistego z magistralą miejscową przy założeniu, że wiadomości szeregowane są przy użyciu metod szeregowania : FIFO, GRMS, EDF. Badania przeprowadzone były dla najgorszego przypadku, tzn. zostało założone, że w chwili t = 0 zostały aktywowane wszystkie zadania w systemie. Uwzględniono także czasy blokowania wiadomości. Wyniki badań zaprezentowano w postaci graficznej i tabelarycznej.
EN
In this paper properties and time parameters of three fieldbuses PROFIBUS FMS, CAN, InterBus are presented as the result of research based on checking of RT - constraint fulfilment of messages in distributed real time system based on fieldbus. Research was done for FIFO method and real time methods scheduling GRMS and EDF. Tests were performed for the worst case - it was assumed that all messages in system were activated in instant zero. The message blocking is also considered. Results of testing were obtain by using computer algorithms created by author. On the basis of performing tests we can draw conclusions as follow: * FIFO method is very simple and requires small calculating time, bul its performance are the least from among priorities method, * Dynamic methods have better performance than static priority for each fieldbus (except for CAN fieldbus, becouse there is impossible to apply dynamic method to this fieldbus), * InterBus-S fieldbus has the best effects for transmitting 1 or 2 bytes data length, CAN fieldbus has the best effects for transmitting 3 to 8 bytes data length and PROFIBUS FMS fieldbus has the best effects for transmitting massages with data length above 8 bytes.
6
Content available remote Cost-efficient synthesis of multiprocessor heterogeneous systems
EN
In this paper an algorithm for co-synthesis of distributed embedded systems is presented. The algorithm is based on iterative improvement heuristics, taking into consideration sophisticated modifications and possibilities of further improvements. Starting from the solution with the highest performance, architecture of the system is modified until it achieves the lowest cost. It has been observed that the algorithm presented has the capacity of getting out of the local minima. Experimental results showed high efficiency of the algorithm. Almost all results obtained with the help of the algorithm were significantly better than the results obtained with the help of Yen-Wolf algorithm presented in the literature.
EN
The problem of locating pointwise sensor measurements so as to optimally estimate unknown parameters in a class of distributed systems is studied. Based on a scalar measure of performance defined on the corresponding Fisher information matrix, two approaches are developed for this problem: introduction of continuous designs, which allows for adaptation of well-known sequential algorithms of classical optimum experimental design, and application of standard non-linear programming techniques. In each case, particular algorithms are delineated and analysis of the appropriate sensor placements is made. The relative advantages and shortcomings of both the approaches are discussed and demonstrated by applying them to a two-dimensional diffusion process.
PL
Jednym z popularnych rozwiązań stosowanych w obszarze systemów sterujących jest użycie rozproszonego systemu komputerowego. Poszczególne węzły takiego systemu są często łączone poprzez magistralę miejscową (fieldbus). Istotnym zadaniem staje się wówczas sprawdzenie w takim systemie dochowania ograniczeń czasu rzeczywistego, w szczegolności dla przesyłanych wiadomości poprzez sieć. W ostatnim okresie nastąpił rozwój analitycznych metod szeregowania i sprawdzania spełnienia warunków czasu rzeczywistego dla środowiska scentralizowanego systemów operacyjnych czasu rzeczywistego, a jednocześnie podejmowane są działania mające na celu modyfikację tych metod, aby zastosować je do badania systemów rozproszonych czasu rzeczywistego, a w szczególności do szeregowania wiadomosci przesyłanych poprzez sieci. W pierwszej części artykułu dokonano porównania pomiędzy szeregowaniem zadań a szeregowaniem wiadomości. Następnie przedyskutowano prosty model systemu rozproszonego, na podstawie którego został przyjęty model strumienia wiadomości. Jako metodę szeregowania i sprawdzenia spełnienia warunkow czasu rzeczywistego wybrano metodę Generalised Rate Monotonic Scheduling (GRMS). Przedstawiono zastosowanie tej metody dla badania spełnienia warunków czasu rzeczywistego wiadomości przesyłanych poprzez magistrale miejscowe stosujące metodę odpytań i metodę przekazywania żetonu w warstwie dostępu do medium fizycznego. Powyższe rozważania zobrazowane zostały przykładem obliczeniowym dla magistrali PROFIBUS DP.
EN
The popular solution for control systems is to use distributed computer system. The computers in such a system are often connected by fieldbus network. Investigation of real time behaviour of distributed computer control systems based on fieldbus is an important task. Recently, analytic methods for test compliance with real time requirements have been developed for centralised real time operating systems. At the same time attempts to modify these methods have been undertaken, so as to use them to examine distributed real time systems, especially for scheduling the message transmitted through the network. In the first part of the paper a comparison is made between task and message scheduling. The model of distributed computer system for control is discussed. The message scheduling basing on GRMS method is discussed for two types of network protocols used in fieldbuses : polling and token passing. The message blocking is also considered in the model. An example illustrates method for checking of RT - constraint fulfilment for PROFIBUS DP fieldbus.
9
Content available remote Spreadability and spray actuators
EN
Spreadability of distributed systems has been recently introduced (El Jai and Kassara, 1994; El Jai et al., 1995). The aim of this paper is to propose a new approach to spreadability based on considering actuators as generators of spreadability. This approach is related to ecological and environmental problems, like vegetation dynamics, where the actuators are assumed to be governed by a transport equation while the system is described by a diffusion model.
10
Content available remote Source identification in distributed parameter systems
EN
In this paper, we solve the problem of the pointwise source identification of the convection-diffusion transport processes. This is done by converting the identification problem into an optimization problem of finding a spatial location and the capacity of a point source which results in the best match of model-predicted measurements to actual observed measurements.
11
Content available remote Observer design for distributed-parameter dissipative bilinear systems
EN
This paper deals with the problem of observer synthesis for a class of infinitedimensional dissipative bilinear systems working for a class of inputs. First, a simple observer (the estimation error converges strongly asymptotically to zero) for strongly persistent input signals is presented. Next, sufficient conditions are given which guarantee the existence of an exponential observer for skew-adjoint bilinear systems based on the time-varying differential Riccati (or Lyapunov) equations working for strongly regularly persistent inputs. These results are illustrated by means of partial-differential systems.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.