We propose a novel hybrid planning approach for the automated generation of IT change plans. The algorithm addresses an abstraction mismatch between refinement of tasks and reasoning about the lifecycle and state-constraints of domain objects. To the best of our knowledge, it is the first approach to address this abstraction mismatch for IT Change Management and to be based on Artificial Intelligence planning techniques. This has several advantages over previously existing research including increased readability, expressiveness, and maintainability of the descriptions. We developed the foundations of the approach and successfully validated it by applying it to change request planning for TikiWiki, a Content Management System.
This paper presents a scalable and elastic distributed system for monitoring Cloud infrastructure based on a pure peer-to-peer architecture. Its distributed nature enables to deploy long-living queries across the network to monitor a diverse set of entities and metrics, spanning across all layers of a Cloud stack which can change rapidly. This allows for aggregating low-level metrics from operating systems, to higher level application-specific metrics derived from services, databases, or application log files. The observed metrics and information can be evaluated and used to reliably trigger policies to automate complex management tasks within a Cloud environment. The architecture incorporates a query framework for obtaining high-level information and a policy framework to provide self-management capabilities to monitored Cloud infrastructure. The system has been implemented as a proof of concept. Details and statistical results are provided to validate the scalability of the underlying architecture.
Cloud computing infrastructure services enable the flexible creation of virtual infrastructures on-demand. However, the creation of infrastructures is only a part of the process for provisioning services. Other steps such as installation, deployment, configuration, monitoring and management of software components are needed to fully provide services to end-users in the cloud. This paper describes a peer-to-peer architecture to automatically deploy services on cloud infrastructures. The architecture uses a component repository to manage the deployment of these software components, enabling elasticity by using the underlying cloud infrastructure provider. The life cycle of these components is described in this paper, as well as the language for defining them. We also describe the open-source proof-of-concept implementation. Some technical information about this implementation together with some statistical results are also provided.
FREEBASE contains entities and relation information but is highly incomplete. Relevant information is ubiquitous in web text, but extraction deems challenging. We present JEDI, an automated system to jointly extract typed named entities and FREEBASE relations using dependency pattern from text. An innovative method for constraint solving on entity types of multiple relations is used to disambiguate pattern. The high precision in the evaluation supports our claim that we can detect entities and relations together, alleviating the need to train a custom classifier for an entity type 1 .
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.