Multi-cloud computing has been proposed as a way to reduce vendor dependence, comply with location regulations, and optimize reliability, performance and costs. Meanwhile, microservice architectures are becoming increasingly popular in cloud computing as they promote decomposing applications into small services that can be independently deployed and scaled, thus optimizing resources usage. However, setting up a multi-cloud environment to deploy a microservices-based application is still a very complex and time consuming task. Each microservice may require different functionality (e.g. software platforms, databases, monitoring and scalability tools) and have different location and redundancy requirements. Selection of cloud providers should take into account the individual requirements of each service, as well as the global requirements of reliability and scalability. Moreover, cloud providers can be very heterogeneous and offer disparate functionality, thus hindering comparison. In this paper we propose an automated approach for the selection and configuration of cloud providers for multi-cloud microservices-based applications. Our approach uses a domain specific language to describe the application's multi-cloud requirements and we provide a systematic method for obtaining proper configurations that comply with the application's requirements and the cloud providers' constraints. Index Terms-multi-cloud; microservices; cloud management; variability management; software product lines Cloud Provider Ontology Cloud Provider Mappings Cloud Provider Variability Model (Feature Model) Cloud Experts Application Ontology Developer Multi-Cloud Application Requirements Global Cloud Ontology Multi-Cloud Configuration IT operations
Due to the number of cloud providers, as well as the extensive collection of services, cloud computing provides very flexible environments, where resources and services can be provisioned and released on demand. However, reconfiguration and adaptation mechanisms in cloud environments are very heterogeneous and often exhibit complex constraints. For example, when reconfiguring a cloud system, a set of available services may be dependent on previous choices, or there may be alternative ways of adapting the system, with different impacts on performance, costs or reconfiguration time. Cloud computing systems exhibit high levels of variability, making dynamic software product lines (DSPLs) a promising approach for managing them. However, in DSPL approaches, verification is often limited to verifying conformance to a variability model, but this is insufficient to verify complex reconfiguration constraints that exist in cloud computing systems. In this paper, we propose the use of temporal constraints and reconfiguration operations to model a DSPL's reconfiguration lifecycle. We demonstrate how these concepts can be used to model the variability of cloud systems, and we use our approach to identify reconfigurations that meet given criteria.
Feature modeling is widely used to capture and manage commonalities and variabilities in software product lines. Cardinality-based feature models are used when variability applies not only to the selection or exclusion of features but also to the number of times a feature can be included in a product. Feature cardinalities are usually considered to apply in either a local or global scope. However, we have identified that these interpretations are insufficient to capture the variability of cloud environments. In this paper, we redefine cardinality-based feature models to allow multiple relative cardinalities between features and we discuss the effects of relative cardinalities on feature modeling semantics, consistency and cross-tree constraints. To evaluate our approach we conducted an analysis of relative cardinalities in four cloud computing providers. In addition, we developed tools for reasoning on feature models with relative cardinalities and performed experiments to verify the performance and scalability of the approach. The results from our study indicate that extending feature models with relative cardinalities is feasible and improves variability modeling, particularly in the case of cloud environments.
The drastic decrease in mobile SMS costs turned phone users more prone to spam messages, usually with unwanted marketing or questionable content. As such, researchers have proposed different methods for detecting SMS spam messages. This paper presents a technique for embedding SMS messages into vector spaces that is suitable for spam detection. The proposed approach relies on mining patterns that are relevant for distinguishing spam from legitimate messages. A subset of those patterns is used to construct a function that maps text messages into a multidimensional vector space. The extracted patterns are represented as skip-grams of token attributes, where a skip-gram can be seen as a generalization of the n-gram model that allows a distance greater than one between matched tokens in the text. We evaluate the proposed approach using the generated vectors for spam classification on the UCI Spam Collection dataset. The experiments showed that our method combined with shallow networks reached accuracy that is competitive with state-of-the-art approaches.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.