Isaac Lera

Performance-related ontologies and semantic web applications for on-line performance assessment intelligent systems

Several techniques and applications have been proposed to aid the decision taking process in the system performance domain. Most of these techniques have depicted the performance model of systems through annotations of performance measurements coming from specific software descriptive syntactical languages. However, the semantic representation of performance information provides the possibility of its ulterior machine-processable logical interpretation and therefore the applicability of inference rules about a particular domain. Moreover, ontologies ease the interchange and reuse of knowledge of particular domains, e.g. system performance. In this work, we propose a performance ontology together with the system performance analysis technique as an example of framework building for intelligent applications based on semantic web. The paper also shows the construction of performance rules through OWL to automatically infer new performance constraints and QoS knowledge about the system on execution.

Performance assessment on ambient intelligent applications through ontologies

This paper brings together the performance assessment of ambient intelligence architectures systems with ontology engineering. Thus, firstly appropriate description methods for distributed intelligent applications are summarized. Derived from the system characterization, typical software performance engineering techniques are based on the augmented description of the model regarding performance annotations. However, these annotations are only related with the syntactical view of the software architecture. In the next generation of performance assessment tools for ambient intelligent systems, the description of the system would be capable of reasoning and acquiring knowledge about performance. Having an appropriate architectural description including performance aspects, any possible design options for intelligent distributed applications can be evaluated according to their performance impact. Therefore, we propose the use of an ontology with performance-related information not only to possible evaluate the architecture through the common off-line procedure but also the first step to build a broker that assesses the performance of the system during its execution.

Performance assessment of intelligent distributed systems through software performance ontology engineering (SPOE)

In the computer science community there is a growing interest in the field of Ambient Intelligent Systems. This systems surround their human users with computing and networking technology unobtrusively embedded in their environment. This technology is aimed to provide the users with useful information and to take action to make the environment more convenient for them. As the number of users increases the resources that make Ambient Intelligence possible can be easily saturated making the system unstable and projecting an image of poor QoS to the users. The main goal of this paper is to provide the means for the Ambient Intelligent Systems to monitor themselves and take corrective action automatically if performance starts to drop. Our approach uses a Performance Ontology that structures the knowledge about Software Performance Engineering, and a reasoning engine that acts like an expert system with the Performance Ontology as its foundation. The case study at the end shows the applicability of the developed techniques.

Web operational analysis through performance-related ontologies in OWL for intelligent applications

Web performance engineering techniques are classically based on the augmented description of the model regarding performance annotations. However, these annotations are only related with the syntactical view of the system. The next generation of performance assessment tools for intelligent systems would be capable of acquiring knowledge and even reasoning about performance as the systems will work. The use of ontologies with performance-related information may be used to build on-line performance brokers that assess the performance of the system during its execution. We present the web operational analysis as an example of its future utilization by a framework for building intelligent applications based on semantic web.

OWL-M Extension for Semantic Representations of Ontology Alignments

Ontology Matching (OM) consists of set of processes that decide and establish if ontology elements share lexical or semantic relationships. For a correct alignment, when we establish semantic relationships, it is necessary to find out the meaning of the elements that depends on the context. In this paper, we propose an extension of OWL called OWL-M to improve the representation of multiple types of relationships among ontology elements taking account the context. In function of each OWL-M alignment, a similarity degree is computed. Moreover, to assist in decision making when human participation is required, we define a set of annotations to describe the logical decisions used by OWL-M processes.

Genetic Algorithm for Multi-Objective Optimization of Container Allocation in Cloud Architecture

The use of containers in cloud architectures has become widespread, owing to advantages such as limited overheads, easier and faster deployment, and higher portability. Moreover, they present a suitable architectural solution for the deployment of applications created using a microservice development pattern. Despite the large number of solutions and implementations, there remain open issues that have not been completely addressed in container automation and management. Container resource allocation influences system performance and resource consumption, and so it is a key factor for cloud providers. We propose a genetic algorithm approach, using the Non-dominated Sorting Genetic Algorithm-II (NSGA-II), to optimize container allocation and elasticity management, motivated by the good results obtained with this algorithm in other resource management optimization problems in cloud architectures. Our optimization algorithm enhances system provisioning, system performance, system failure, and network overhead. A model for cloud clusters, containers, microservices, and four optimization objectives is presented. Experimental results demonstrate that our approach is a suitable solution for addressing the problem of container allocation and elasticity, and it obtains better objective values than the container management policies implemented in Kubernetes.

Performance-related Ontologies for Ubiquitous Intelligence based on Semantic Web Applications

Several techniques and applications have been proposed to aid the decision taking process in the system performance domain. Most of these techniques have depicted the performance model of systems through annotations of performance measurements coming from specific software descriptive syntactical languages. However, the semantic representation of performance information provides the possibility of its ulterior machine-processable logical interpretation and therefore the applicability of inference rules about a particular domain. In this work, we propose a performance ontology together with the system performance analysis technique as an example of framework building for intelligent applications based on semantic Web

Migration-Aware Genetic Optimization for MapReduce Scheduling and Replica Placement in Hadoop

This work addresses the optimization of file locality, file availability, and replica migration cost in a Hadoop architecture. Our optimization algorithm is based on the Non-dominated Sorting Genetic Algorithm-II and it simultaneously determines file block placement, with a variable replication factor, and MapReduce job scheduling. Our proposal has been tested with experiments that considered three data center sizes (8, 16 and 32 nodes) with the same workload and number of files (150 files and 3519 file blocks). In general terms, the use of a placement policy with a variable replica factor obtains higher improvements for our three optimization objectives. On the contrary, the use of a job scheduling policy only improves these objectives when it is used along a variable replication factor. The results have also shown that the migration cost is a suitable optimization objective as significant improvements up to 34% have been observed between the experiments.

Performance improvement of web caching in Web 2.0 via knowledge discovery

Web 2.0 systems are more unpredictable and customizable than traditional web applications. This causes that performance techniques, such as web caching, limit their improvements. Our study was based on the hypotheses that the use of web caching in Web 2.0 applications, particularly in content aggregation systems, can be improved by adapting the content fragment designs. We proposed to base this adaptation on the analysis of the characterization parameters of the content elements and on the creation of a classification algorithm. This algorithm was deployed with decision trees, created in an off-line knowledge discovery process. We also defined a framework to create and adapt fragments of the web documents to reduce the user-perceived latency in web caches. The experiment results showed that our solution had a remarkable reduction in the user-perceived latency even losses in the cache hit ratios and in the overhead generated on the system, in comparison with other web cache schemes.

Semantic layered architecture to integrate FR/NFR in software performance engineering

The integration process using UML Profile for Schedulability, Performance, and Time Specification (SPT) or MARTE specification presents some deficiencies due to the inability to formalize the representation criteria and to relate information in a transparent way between the outputs of performance models and software designs. The drawbacks of FR and NFR model integration appears to be crucial for the real application of several SPE techniques, since the performance modelling and the software modelling are not semantically related because there are not explicit and formal semantic interpretation of the concepts. To avoid this, we propose a semantic architecture based on well-known Web Semantic Layered Architecture