A5003618718- Cloud Computing and Resource Management
- Software System Performance and Reliability
- Anomaly Detection Techniques and Applications
- Artificial Intelligence in Games
- Scientific Computing and Data Management
- Distributed and Parallel Computing Systems
- IoT and Edge/Fog Computing
- Reinforcement Learning in Robotics
- Advanced Data Storage Technologies
- Network Security and Intrusion Detection
- Distributed systems and fault tolerance
- Fault Detection and Control Systems
- Data Quality and Management
- Evolutionary Algorithms and Applications
- Data Stream Mining Techniques
- Time Series Analysis and Forecasting
- Multi-Agent Systems and Negotiation
- Smart Grid Security and Resilience
- Parallel Computing and Optimization Techniques
- Wireless Networks and Protocols
- Automated Road and Building Extraction
- Image and Object Detection Techniques
- Simulation Techniques and Applications
- Intelligent Tutoring Systems and Adaptive Learning
- Cloud Data Security Solutions
West University of Timişoara
2013-2024
Institute e-Austria Timisoara
2015-2023
Machine learning has received increasing attention in computer science recent years and many types of methods have been proposed. In networks, little paid to the use ML for fault detection, main reason being lack datasets. This is motivated by reluctance network operators share data about their infrastructure failures. this paper, we attempt fill gap using anomaly detection techniques discern hardware failure events wireless community networks. For purpose 4 unsupervised machine learning,...
Model-driven engineering (MDE) often features quality assurance (QA) techniques to help developers creating software that meets reliability, efficiency, and safety requirements. In this paper, we consider the question of how quality-aware MDE should support data-intensive systems. This is a difficult challenge, since existing models QA largely ignore properties data such as volumes, velocities, or location. Furthermore, requires ability characterize behavior technologies Hadoop/MapReduce,...
Model-driven engineering (MDE) often features quality assurance (QA) techniques to help developers creating software that meets reliability, efficiency, and safety requirements. In this paper, we consider the question of how quality-aware MDE should support data-intensive systems. This is a difficult challenge, since existing models QA largely ignore properties data such as volumes, velocities, or location. Furthermore, requires ability characterize behavior technologies Hadoop/MapReduce,...
The goal of this paper is to create a hybrid system based on Multi-Agent Architecture that will investigate the evolution some neural network methods along with technical and fundamental analysis stock market indexes how information influences behavior in order improve profitability short or medium time period investment. proposed compares results Standard Feed Forward Neural Network, Elman Jordan Recurrent Networks Network evolved Neuro Evolution Augmenting Topologies (NEAT) which gives...
The main goal of this paper is the design a multi-agent system (MAS) that handles unit micromanagement in real time strategy games and able to adapt/learn during game play. To achieve we adopted rtNEAT approach order obtain customized neural network topologies, thus avoiding generation too complex architectures. Also by defining internal external inputs for each agent managed create independent agents are cooperate form teams their mutual benefit at same eliminate unnecessary communication...
This paper makes a short overview of current state the art monitoring tools for cloud and big data frameworks. In order to effectively create, test deploy new algorithms or frameworks one needs suitable solutions. Hence we aim on creating critical some solutions existing market. Also present relevant metrics used applications, focused mainly deployment scenarios
Deploying and running applications in multi-cloud environments is a challenging task for number of reasons:different configuration parameters are needed differentcloud environments, application's artefacts may vary acrosstechnologies or cloud providers, services provided at IaaSlevel from one provider to another. This paperintroduces run-time platform that enables the deploymentand execution on multi-clouds with guaranteedquality service (QoS), details underlying ofthe unified layer...
Wireless community networks, WCN, have proliferated around the world. Cheap off-the-shelf WiFi devices enabled this new network paradigm where users build their own infrastructure in a do-it-yourself alternative to traditional operators. The fact that are responsible for administration of nodes makes very dynamic. There frequent reboots networking devices, and join leave network. In addition, unplanned deployment it heterogeneous, with both high low capacity links. Therefore, anomaly...
Latest advances in information technology and the widespread growth different areas are producing large amounts of data. Consequently, past decade a number distributed platforms for storing processing datasets have been proposed. Whether development or production, monitoring applications running on these is not an easy task dedicated tools were proposed this scope. In paper we present distributed, scalable, highly available platform able to collect, store, query process data obtained from...
Nowadays, massive amounts of data are acquired, transferred, and analyzed nearly in real-time by utilizing a large number computing storage elements interconnected through high-speed communication networks. However, one issue that still requires research effort is to enable efficient monitoring applications infrastructures such complex systems. In this paper, we introduce an Integer Linear Programming (ILP) model called M3AT for optimized assignment agents aggregators on large-scale We...
We are witnessing a wave of emerging cloud computing technologies and services that empower advanced applications from different vertical sectors, with diverse requirements. These trends give rise to number fundamental challenges relate the application deployment, support heterogeneous infrastructures provided security. In this setting, SERRANO project steps in define an intent-based paradigm operating federated consisting edge, HPC resources, which will be realized through platform....
This paper presents a multi-agent system that handles unit micromanagement using online machine learning in real time strategy games. We used rtNEAT algorithm order to obtain customized neural network topologies, thus avoiding complex architecture. use an ontology based template create suitable input and outputs for agents enabling them cooperate form teams their mutual benefit eliminating communication overhead. The AI was implemented the JADE framework BWAPI handled between our game. have...
Extreme Data is an incarnation of Big concept distinguished by the massive amounts data that must be queried, communicated and analyzed in near real-time using a very large number memory or storage elements exascale computing systems.Immediate examples are scientific produced at rate hundreds gigabits-per-second stored, filtered analyzed, millions images per day parallel, one billion social posts queried on in-memory components database.Traditional disks commercial nowadays cannot handle...
The collection and aggregation of monitoring data from distributed applications are an extremely important topic. scale these applications, such as those designed for Big Data, makes the performance services responsible parsing aggregating logs a key issue. Logstash is well-known open source framework centralizing both structured unstructured data. As with many throttling common issue due to incoming exceeding processing ability. conventional approach improving usually entails increasing...
Exascale systems are a hot topic of research in computer science. These contrast to current Cloud, Big Data and HPC will routinely contain hundreds thousand nodes generating millions events. At this scale hardware fault anomalous behaviour is not only more likely but be expected. In paper we describe the architecture monitoring solution coupled with an event detection component. The latter component extremely important order handle multitude potential We major lacking that needs done, which...
Adaptive Game AI has been one of the key topics being researched in field academic game research. In this paper we present a comparison several domain independent machine learning methods with aid which extract expert knowledge from logs. Each log is represented as feature vector that encodes cardinality and timing for player actions. We compare wide variety classification highlight ones are best deployment an adaptive systems.
A Distributed Application Topology is a valuable commodity built on the strength of long and iterative design process. topology generally refined over time, other topologies can use it as component, community may share it. To reproduce deployment, several properties must be recorded such data origin, processing steps, configuration settings, hardware requirements. This paper presents an extension to TOSCA specification that allows for definition accelerator-aware services span from Cloud...