TOLERANCE TO BYZANTINE FAULTS: ADAPTIVE DYNAMIC SERVICES FOR PARTICIONABLE SYSTEMS (Q3149538): Difference between revisions
Jump to navigation
Jump to search
(Created claim: summary (P836): RELIABLE DISTRIBUTED COMPUTATION IS RAPIDLY EVOLVING INTO SCENARIOS WITH CAPABILITIES OF SELF-ORGANISATION AND ADAPTATION. FAULT-TOLERANT DISTRIBUTED SYSTEMS, BASED ON INTERCONNECTED COMPUTER SETS THAT INCLUDE REPLICATED SERVERS, AD-HOC DEVICE COLLECTIONS, WIRELESS SENSOR NETWORKS, PEER-TO-PEER NETWORKS WITH OR WITHOUT INFRASTRUCTURE, OPPORTUNISTIC NETWORKS OR CLOUD COMPUTING SYSTEMS, ALLOW PARTICIPANTS TO ACCESS SERVICES REGARDLESS OF LOCATION,...) |
(Changed label, description and/or aliases in en: translated_label) |
||
label / en | label / en | ||
TOLERANCE TO BYZANTINE FAULTS: ADAPTIVE DYNAMIC SERVICES FOR PARTICIONABLE SYSTEMS |
Revision as of 15:00, 12 October 2021
Project Q3149538 in Spain
Language | Label | Description | Also known as |
---|---|---|---|
English | TOLERANCE TO BYZANTINE FAULTS: ADAPTIVE DYNAMIC SERVICES FOR PARTICIONABLE SYSTEMS |
Project Q3149538 in Spain |
Statements
48,279.0 Euro
0 references
96,558.0 Euro
0 references
50.0 percent
0 references
30 December 2016
0 references
29 December 2020
0 references
UNIVERSIDAD DEL PAIS VASCO/EUSKAL HERRIKO UNIBERTSITATEA
0 references
20069
0 references
LA COMPUTACION DISTRIBUIDA CONFIABLE ESTA EVOLUCIONANDO RAPIDAMENTE HACIA ESCENARIOS CON CAPACIDADES DE AUTO-ORGANIZACION Y ADAPTACION. LOS SISTEMAS DISTRIBUIDOS TOLERANTES A FALLOS, BASADOS EN CONJUNTOS DE ORDENADORES INTERCONECTADOS QUE INCLUYEN SERVIDORES REPLICADOS, COLECCIONES AD-HOC DE DISPOSITIVOS, REDES DE SENSORES INALAMBRICAS, REDES ENTRE PARES CON O SIN INFRAESTRUCTURA, REDES OPORTUNISTAS O SISTEMAS DE COMPUTACION EN LA NUBE, PERMITEN EN TEORIA A SUS PARTICIPANTES ACCEDER A SERVICIOS INDEPENDIENTEMENTE DE SU LOCALIZACION, DE LA TOPOLOGIA O DE LA MOVILIDAD. EN LA PRACTICA, DISEÑAR SERVICIOS CONFIABLES PARA SISTEMAS DISTRIBUIDOS ALTAMENTE DINAMICOS ES UN RETO AUN POR RESOLVER SATISFACTORIAMENTE. LOS CONTINUOS CAMBIOS EN LA COMPOSICION DEL SISTEMA Y EN LA TOPOLOGIA DE LA RED DEBIDOS A ATAQUES, PERDIDAS DE MENSAJES, CONEXIONES Y DESCONEXIONES, FALLOS Y MOVILIDAD DE LOS NODOS, OBLIGAN A APLICAR ENFOQUES NOVEDOSOS PARA EL SOPORTE DE APLICACIONES, EN PARTICULAR PARA LA RESOLUCION DEL PARADIGMA DEL CONSENSO O ACUERDO DISTRIBUIDO TOLERANTE A FALLOS._x000D_ RECIENTEMENTE NUESTRO GRUPO DE INVESTIGACION HA CONTRIBUIDO EN ESTA AREA CON LA PROPUESTA DE UN MODELO FORMAL PARA CATEGORIZAR SISTEMAS DISTRIBUIDOS CONFIABLES Y ADAPTATIVOS, JUNTO CON UN SERVICIO DE ELECCION DE LIDER PARA SISTEMAS EN LOS QUE SE ALTERNAN PERIODOS DE COMPORTAMIENTO ESTABLE CON PERIODOS DE INESTABILIDAD. ESTE ENFOQUE, CONVENIENTEMENTE EXTENDIDO A TODOS LOS ASPECTOS DEL SISTEMA DISTRIBUIDO, PERMITE REFLEJAR DE MANERA ADECUADA LA REALIDAD DE MUCHOS SISTEMAS ACTUALES. ASI, EL ACUERDO DISTRIBUIDO NECESARIO PARA PROPORCIONAR SERVICIOS CONFIABLES ESTA SOPORTADO POR UN PROTOCOLO DE ELECCION DE LIDER QUE PROGRESA EN PERIODOS DE ESTABILIDAD Y PRESENTA RESILIENCIA EN LOS PERIODOS PROVOCADOS POR CUALQUIER FORMA DE INESTABILIDAD._x000D_ EL OBJETIVO GENERAL DEL PROYECTO BFT-DYNASTIE ES EXTENDER EL MODELO BASADO EN LA ALTERNANCIA DE PERIODOS DE COMPORTAMIENTO ESTABLE E INESTABLE A TODOS LOS ASPECTOS DE LOS SISTEMAS DISTRIBUIDOS TOLERANTES A FALLOS, INCLUYENDO LOS MODELOS DE SINCRONIA, DE FALLOS DE PROCESOS Y CANALES DE COMUNICACION, DE COMPOSICION DEL SISTEMA, DE LA MOVILIDAD DE LOS NODOS, Y DEL PARTICIONADO DE LA RED. LOS DOS RETOS PRINCIPALES Y NOVEDOSOS DE ESTE PROYECTO SON: LA CONSIDERACION DEL MODELO DE FALLO MAS GENERAL Y COMPLEJO DE ABORDAR, CONOCIDO COMO BIZANTINO, ARBITRARIO O MALICIOSO, QUE REQUIERE MAYORIAS CUALIFICADAS Y LA UTILIZACION DE TECNICAS DEL AREA DE LA SEGURIDAD; Y EL FUNCIONAMIENTO DEL SISTEMA EN MODO PARTICIONADO._x000D_ COMO RESULTADO DEL PROYECTO SE DESARROLLARA UN CONJUNTO DE SERVICIOS DE ACUERDO DISTRIBUIDO TOLERANTES A FALLOS BIZANTINOS SIGUIENDO UN ENFOQUE MODULAR. ASI, SE IDENTIFICAN TRES SERVICIOS BASICOS (DETECCION DE FALLOS, ELECCION DE LIDER Y RADIADO FIABLE) Y OTROS TANTOS SERVICIOS AVANZADOS BASADOS EN LOS ANTERIORES (CONSENSO, PUBLISH/SUBSCRIBE Y RADIADO CON ORDEN TOTAL), QUE A SU VEZ DARAN SOPORTE AL DESARROLLO DE APLICACIONES TOLERANTES A FALLOS._x000D_ CON EL OBJETIVO DE ESTABLECER LA VALIDEZ DEL ENFOQUE PROPUESTO Y VERIFICAR SU APLICABILIDAD PRACTICA, SE EVALUARAN LOS SERVICIOS DESARROLLADOS EN DOS AMBITOS DE APLICACION: LA GENERACION DISTRIBUIDA DE CRIPTOMONEDAS TIPO BITCOIN Y EL CONTROL DE TRANSACCIONES UTILIZANDO LOS SERVICIOS DE CONSENSO Y DE RADIADO CON ORDEN TOTAL, Y LA UTILIZACION DEL SERVICIO PUBLISH/SUBSCRIBE EN LA DENOMINADA INTERNET DE LOS OBJETOS, MAS CONCRETAMENTE EN EL AMBITO DE LOS SISTEMAS CIBERFISICOS BASADOS EN REDES DE SENSORES INALAMBRICAS. (Spanish)
0 references
RELIABLE DISTRIBUTED COMPUTATION IS RAPIDLY EVOLVING INTO SCENARIOS WITH CAPABILITIES OF SELF-ORGANISATION AND ADAPTATION. FAULT-TOLERANT DISTRIBUTED SYSTEMS, BASED ON INTERCONNECTED COMPUTER SETS THAT INCLUDE REPLICATED SERVERS, AD-HOC DEVICE COLLECTIONS, WIRELESS SENSOR NETWORKS, PEER-TO-PEER NETWORKS WITH OR WITHOUT INFRASTRUCTURE, OPPORTUNISTIC NETWORKS OR CLOUD COMPUTING SYSTEMS, ALLOW PARTICIPANTS TO ACCESS SERVICES REGARDLESS OF LOCATION, TOPOLOGY OR MOBILITY. IN PRACTICE, DESIGNING RELIABLE SERVICES FOR HIGHLY DYNAMIC DISTRIBUTED SYSTEMS IS A CHALLENGE YET TO BE SOLVED SATISFACTORILY. Continuums in the composition of the system and in the topology of the network due to attacks, lost from messages, connections and disconnections, fades and mobility of the nodes, obliging to apply new approaches to the support of applications, in preparation for the resolution of the consensus paradigm or agreement disTRIBUTED TO FALLOS._x000D_ RECIENTLY OUR RESEARCH GROUP HAS IN THIS AREA WITH THE PROPOSAL OF A FORMAL MODEL FOR CATEGORISING DISTRIBUTED AND ADAPTATITED SYSTEMS, together with a leadership service for systems in which periods of stable transport are alternated with periods of unrest. THIS APPROACH, WHICH IS CONVENIENTLY EXTENDED TO ALL ASPECTS OF THE DISTRIBUTED SYSTEM, MAKES IT POSSIBLE TO ADEQUATELY REFLECT THE REALITY OF MANY EXISTING SYSTEMS. Thus, the NECESARY DISTRIBUTED AGREEMENT to provide reliable services is supported by a protocollo de elECTION de l'leader that will provide in periods of stability and subject to resilience in the periods provided by any unrest FORM._x000D_ the GENERAL OBJECT OF THE PROJECT BFT-Dynastie is to extend the model based on the alternance of stably and unsustainable periods of transport to all the aspectos of the distributed systems to falls, including synchronous models, of process and communication channels, of system composition, of the mobility of the nodes, and of the partitioning of the network. THE TWO MAIN AND NOVEL CHALLENGES OF THIS PROJECT ARE: CONSIDERATION OF THE MOST GENERAL AND COMPLEX FAILURE MODEL TO ADDRESS, KNOWN AS BYZANTINE, ARBITRARY OR MALICIOUS, WHICH REQUIRES QUALIFIED MAJORITY AND THE USE OF TECHNIQUES FROM THE SECURITY AREA; And the SYSTEM’S FUNCTIONING IN PARTICIONED MODE._x000D_ AS THE PROJECT RESULTS A CONJUNT OF SERVICES will be developed in accordance with distribiated TOLERANTES TOLERANTES TOLERANTED BIZANTIN FALLS Following a modular focus. Thus, three BASIC SERVICES (DETECTION OF FALLS, ELECTION OF LIDER AND FIABLE Radiate) and other AVAANZED SERVICES BASED IN ANTERIORS (CONSENSE, PUBLISH/Subscribing and Radiated with TOTAL ORDEN), that your time will be supported by the development of applications to fallos._x000D_ WITH THE OBJECTIVE OF STAFFING THE VALIDECE OF THE FOLLOWING PROPOSAL AND VERIFICATE YOUR PRACTICAL APPLICABILITY, the services developed in two areas of application will be assessed: THE DISTRIBUTED GENERATION OF BITCOIN-TYPE CRYPTOCURRENCIES AND THE CONTROL OF TRANSACTIONS USING CONSENSUS AND RADIATING SERVICES WITH TOTAL ORDER, AND THE USE OF THE PUBLISH/SUBSCRIBE SERVICE IN THE SO-CALLED INTERNET OF OBJECTS, MORE SPECIFICALLY IN THE SCOPE OF CIBERFISIC SYSTEMS BASED ON WIRELESS SENSOR NETWORKS. (English)
12 October 2021
0 references
Donostia/San Sebastián
0 references
Identifiers
TIN2016-79897-P
0 references