Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 8

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
EN
The paper presents an original method of dynamic classication of objects from a new domain which lacks an expert knowledge. The method relies on analysis of attributes of objects being classied and their general quality Q, which is a combination of particular object's attributes. The method uses a test of normality as a basis for computing the reliability factor of the classication (rfc), which indicates whether the classication and the model of quality Q are reliable. There is no need to collect data about all objects before the classication starts and possibly the best objects ale selected dynamically (on-the-y) while data concerning consecutive objects are gathered. The method is implemented as a software tool called Articial Classication Adviser (ACA). Moreover, the paper presents a case study, where the best candidates for reghting mobile robot operators are selected.
EN
The problem of effective storing and processing of data is a very important aspect of every IoT application. To fulfill these requirements, NoSQL scalable datastores are frequently used. Scalable Distributed 2-layer Data Structures (SD2DS) are examples of such systems. SD2DS is a general purpose structure that can be used as a distributed datastore, easy to adapt to many different needs. In the paper, an enhanced architecture of IoT systems (denoted as IoT*) supplemented with SD2DS is proposed and evaluated. The combination brings many advantages, such as the possibility to create one unified structure consisting of many servers accessible by many clients, usage of different media for storage (such as RAM, hard disk, databases and cloud) and a single access method for both data and data sources.
3
Content available remote Efficient data management on a multicomputer
EN
High performance, fault tolerance and scalability are usual requirements for an application running on a multicomputer. The paper presents different variants of centralized SDDS LH* architecture in the light of all the requirements. Hence, the paper briefly summarizes already published features of SDDS that concern data scalability and fault tolerance, and then introduces a new option for SDDS called throughput scalability that can balance workload of nodes of a multicomputer. Finally, having met all the requirements for efficient management of data on a multicomputer the SDDS schemes are estimated as for the time and memory overhead.
PL
Wysoka wydajność, odporność na błędy i skalowalność to typowe wymagania aplikacji dla multikomputerów. W artykule zaprezentowano różne odmiany struktur SDDS LH* o architekturze scentralizowanej w świetle wszystkich tych wymagań. Podsumowano znane już możliwości struktur SDDS dotyczące skalowalności danych i odporności na błędy oraz przedstawiono nową funkcjonalność SDDS nazwaną skalowalnością przepustowości, pozwalającą na zrównoważenie obciążenia węzłów multikomputera. Ostatecznie, po spełnieniu wszystkich wymagań w kwestii efektywnego zarządzania danymi w obrębie multikomputera, struktury SDDS są analizowane pod względem kosztów czasowych i pamięciowych.
4
EN
Efficient data management and distribution in a multicomputer is a subject of much research. Distributed file systems are the most common solution of this problem, however, recent works are focused on more general data distribution protocols. Scalable, Distributed Data Structures (SDDS) are another promising approach to this issue. In this paper we discuss the efficiency of an implementation of SDDS in various applications. The results of experiments are presented.
EN
A description which summarizes entire and usually big set of data is called its model. The problem investigated in the paper consists in verification of models of data coming from a simulation experiment of selecting candidates for operators of mobile robot (more strictly building reliable predictive model of the data). The models are validated using train-and-test method and verified with the help of the EM (expectation-maximization) algorithm which was originally designed for solving clustering problems with missing data. Actually, the selecting is a clustering problem because the candidates are assigned to ‘chosen’, ‘accepted’ or ‘rejected’ subgroups. For such a case the missing data is the category (the subgroup) for which a candidate should be assigned on the basis of his activity measured during the simulation experiment. The paper explains the procedure of model verification. It also shows experimental results and draws conclusions.
6
Content available remote Experimental evaluation of two touring simulators for training operators of mobot
EN
In the paper two versions of a touring simulator are presented and evaluated. Both versions were developed for training mobot operators. Operators improve their skills playing a game based on the simulator. For simulators the problem of the precision of simulation usually appears. The same is here. Therefore, the simulators differ in quality of graphics. In the paper a usefulness of the simulators for training is evaluated. Experiments answered the question how important for efficient training is high quality of graphics generated by the simulator.
PL
W artykule przedstawiono i porównano dwie wersje symulatora marszruty. Obie wersje zostały przygotowane w celu trenowania operatorów mobilnego robota (mobota). Operatorzy poszerzają swoje umiejętności w trakcie gry opartej na symulatorze. W przypadku symulatorów zawsze pojawia się problem precyzji symulacji. Z tego powodu wersje symulatora różnią się precyzją odwzorowania trasy mobota. W artykule porównano przydatność obu wersji do trenowania operatorów. Przeprowadzone eksperymenty pozwoliły odpowiedzieć na pytanie, jak ważna dla wydajnego szkolenia jest precyzja symulatora.
EN
Scalable Distributed Data Structures (SDDS) consists of two components dynamically spread across a multicomputer: data records belonging to a file and a mechanism controlling record placement in file space. Record (data) faults may lead to invalid computations at most, while faults concerning record placement mechanisms may lead whole SDDS file to crash. In this paper, cause-effect analysis of record placement faults concerning SDDS RP* (Range Partitioning) file is given.
PL
Skalowalne Rozproszone Struktury Danych (SDDS) składają się z dwóch komponentów rozproszonych dynamicznie w obrębie multikomputera: danych należących do pliku oraz mechanizmu kontroli położenia rekordów w pliku. Błędy rekordów (danych) mogą co najwyżej doprowadzić do błędnych obliczeń, podczas gdy błędy związane z mechanizmem rozmieszczania rekordów mogą doprowadzić cały plik SDDS do zniszczenia. W niniejszym artykule dokonano analizy przyczynowo-skutkowej błędów rozmieszczania rekordów dotyczących struktur SDDS typu RP* (z podziałem na zakresy).
PL
Skalowane, rozproszone struktury danych (SDDS - Scalable Distributed Data Structures) mają zastosowanie w multikomputerach. Na SDDS składają się dwa składniki rozpraszane dynamicznie w obrębie multikomputera: rekordy należące do pliku oraz sterowanie umieszczaniem rekordów w pliku. W artykule przedstawiono metodę uodporniania SDDS na błędy sterowania umieszczaniem rekordów w pliku, oraz sposób weryfikacji zaproponowanej metody za pomocą symulatora SDDS z programowym wstrzykiwaniem błędów.
EN
Scalable Distributed Data Structures (SDDS) consists of two components dynamically spread across a multicomputer: records belonging to a file and a mechanism controlling record placement in file space. Faults in mechanism controlling record placement in the file may lead an application to crash, while record data faults may cause invalid computations at most. In the paper fault-tolerant architecture for mechanism controlling record placement in SDDS file is presented and evaluated with the help of SDDS oriented software fault injector. The architecture uses Job Comparison Technique along with TMR. Moreover, 'backwarding' to correct every client's addressing fault is applied. Time overhead due to redundancy introduced is estimated.
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.