“…With Instant-Grid it is possible to try out grid-and cluster technologies [29], and develop and test own applications in a distributed research infrastructure. Its original environment [28] was based on the Globus Toolkit 4 (GT4) grid middleware, but as a part of this thesis extensions were made to enable the fully dynamic deployment of the Portable Batch System (PBS) and the UNiform Interface to COmputing REsources v6 (UNICORE6) grid middleware [135]. The extensions provide maximum compatibility to "real", production-level distributed research infrastructures, since the same technology and software is used.…”
Section: Instant-gridmentioning
confidence: 99%
“…This chapter is partly adapted from our previous works on the topic of automated middleware deployment [28,29,30], as well as on the topic of self-configured demonstration and testing environments [135,188].…”
Section: A Framework For the Simulation Of Heterogeneous Information ...mentioning
confidence: 99%
“…In this work we call this version the Instant-Grid Globus Toolkit 4 edition. We published our results in [28,29,30].…”
Section: A Framework For the Simulation Of Heterogeneous Information ...mentioning
confidence: 99%
“…We gave a detailed technical description of our approach to deploy pre-configured middleware systems in a fully automated process in our previous works [28,29,30,135]. Here we just highlight the main concept of the automated configuration at (a) the startup process and (b) during runtime: Automated Configuration at Startup Technically, services to turn a computer lab into a grid system are based on the well established PXE mechanism, which is also used by several cloud environments as well as in Live-CD projects aiming at cluster setups.…”
Section: Automated Self-configurationmentioning
confidence: 99%
“…Neither Internet connection nor global grid connectivity is required. The ready-to-use applications were presented in our previous works [28,29,30,135,188]. Here we just highlight the most important ones:…”
Section: Ready-to-use Features and Applicationsmentioning
Recently, distributed research infrastructures were built up to feed the emerging resource demand of research institutes and enable researchers from various disciplines to access cutting edge technologies. More recently, the different research disciplines began to approach each other in the mutual goal to answer increasingly complex research problems. It has therefore become increasingly important for research infrastructures to interoperate. Exchanging resource descriptions coming from the information and monitoring systems of different research infrastructures is among the most important steps to reach interoperability.The aim of this thesis is to develop and realise a concept of an interoperable information and monitoring system, which can be used transparently with different existing distributed research infrastructures. This concept also extends the scope of existing information services by including resource properties relating to quality of service and organisational items. We focus on the information and monitoring systems, because of their importance for the other components of distributed research infrastructures. Exchanging resource descriptions, discovering resources, and monitoring the quality and availability of computing and storage resources are essential for executing jobs or transferring data in distributed research infrastructures.Based on a requirement analysis of application scenarios we identify the information needed for exchanging resource descriptions and provide a theoretical model which is capable to include that information. We also present a theoretical approach for a schema mediation process. For modeling resource descriptions in distributed research environments many information schemas and data models exist. In addition to the theoretical analysis, we discuss the evaluation of information schemas being utilized in productional environments. Based on these results, a concept of an automated resource exchange process and a respective generic monitoring architecture supporting it are provided.In addition to our theoretical analysis, concept, and models, a proof of concept for distributed research infrastructures has been developed. We also design and develop a simulation framework for service-oriented, multi-middleware environments that is based on automated system deployment and service provisioning. The evaluation is done by artificial simulation experiments in the simulation framework and in real system scenarios, which demonstrate the applicability of our concept. Instant-Grid, MediGrid, DARIAH and ePIC.Moreover, I want to thank my current and prior colleagues at the GWDG for the nice work atmosphere, as well as my friends and fellow doctoral students for their moral support, which made my day-to-day work balanced and varied. I would like to give special thanks to the two proof-readers of this work.Last, but not least, I want to thank my family for their patience and endless support.
“…With Instant-Grid it is possible to try out grid-and cluster technologies [29], and develop and test own applications in a distributed research infrastructure. Its original environment [28] was based on the Globus Toolkit 4 (GT4) grid middleware, but as a part of this thesis extensions were made to enable the fully dynamic deployment of the Portable Batch System (PBS) and the UNiform Interface to COmputing REsources v6 (UNICORE6) grid middleware [135]. The extensions provide maximum compatibility to "real", production-level distributed research infrastructures, since the same technology and software is used.…”
Section: Instant-gridmentioning
confidence: 99%
“…This chapter is partly adapted from our previous works on the topic of automated middleware deployment [28,29,30], as well as on the topic of self-configured demonstration and testing environments [135,188].…”
Section: A Framework For the Simulation Of Heterogeneous Information ...mentioning
confidence: 99%
“…In this work we call this version the Instant-Grid Globus Toolkit 4 edition. We published our results in [28,29,30].…”
Section: A Framework For the Simulation Of Heterogeneous Information ...mentioning
confidence: 99%
“…We gave a detailed technical description of our approach to deploy pre-configured middleware systems in a fully automated process in our previous works [28,29,30,135]. Here we just highlight the main concept of the automated configuration at (a) the startup process and (b) during runtime: Automated Configuration at Startup Technically, services to turn a computer lab into a grid system are based on the well established PXE mechanism, which is also used by several cloud environments as well as in Live-CD projects aiming at cluster setups.…”
Section: Automated Self-configurationmentioning
confidence: 99%
“…Neither Internet connection nor global grid connectivity is required. The ready-to-use applications were presented in our previous works [28,29,30,135,188]. Here we just highlight the most important ones:…”
Section: Ready-to-use Features and Applicationsmentioning
Recently, distributed research infrastructures were built up to feed the emerging resource demand of research institutes and enable researchers from various disciplines to access cutting edge technologies. More recently, the different research disciplines began to approach each other in the mutual goal to answer increasingly complex research problems. It has therefore become increasingly important for research infrastructures to interoperate. Exchanging resource descriptions coming from the information and monitoring systems of different research infrastructures is among the most important steps to reach interoperability.The aim of this thesis is to develop and realise a concept of an interoperable information and monitoring system, which can be used transparently with different existing distributed research infrastructures. This concept also extends the scope of existing information services by including resource properties relating to quality of service and organisational items. We focus on the information and monitoring systems, because of their importance for the other components of distributed research infrastructures. Exchanging resource descriptions, discovering resources, and monitoring the quality and availability of computing and storage resources are essential for executing jobs or transferring data in distributed research infrastructures.Based on a requirement analysis of application scenarios we identify the information needed for exchanging resource descriptions and provide a theoretical model which is capable to include that information. We also present a theoretical approach for a schema mediation process. For modeling resource descriptions in distributed research environments many information schemas and data models exist. In addition to the theoretical analysis, we discuss the evaluation of information schemas being utilized in productional environments. Based on these results, a concept of an automated resource exchange process and a respective generic monitoring architecture supporting it are provided.In addition to our theoretical analysis, concept, and models, a proof of concept for distributed research infrastructures has been developed. We also design and develop a simulation framework for service-oriented, multi-middleware environments that is based on automated system deployment and service provisioning. The evaluation is done by artificial simulation experiments in the simulation framework and in real system scenarios, which demonstrate the applicability of our concept. Instant-Grid, MediGrid, DARIAH and ePIC.Moreover, I want to thank my current and prior colleagues at the GWDG for the nice work atmosphere, as well as my friends and fellow doctoral students for their moral support, which made my day-to-day work balanced and varied. I would like to give special thanks to the two proof-readers of this work.Last, but not least, I want to thank my family for their patience and endless support.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.