- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- IoT and Edge/Fog Computing
- Parallel Computing and Optimization Techniques
- Advanced Data Storage Technologies
- Scientific Computing and Data Management
- Experimental Learning in Engineering
- Cloud Data Security Solutions
- Service-Oriented Architecture and Web Services
- Interconnection Networks and Systems
- Distributed systems and fault tolerance
- Caching and Content Delivery
- Blockchain Technology Applications and Security
- Software System Performance and Reliability
- Teaching and Learning Programming
- Online Learning and Analytics
- Educational Technology and Assessment
- Intelligent Tutoring Systems and Adaptive Learning
- Advanced Database Systems and Queries
- Embedded Systems Design Techniques
- Security and Verification in Computing
- Peer-to-Peer Network Technologies
- Cloud Computing and Remote Desktop Technologies
- Mobile Agent-Based Network Management
- Big Data and Business Intelligence
Universität Innsbruck
2016-2024
Saints Cyril and Methodius University of Skopje
2011-2020
Software (Spain)
2020
Pillar (Ukraine)
2020
i2CAT
2016
University of Ss. Cyril and Methodius in Trnava
2013-2015
Institute of Informatics of the Slovak Academy of Sciences
2011
Contemporary solutions for cloud-supported, edge-data analytics mostly apply techniques in a rigid bottom-up approach, regardless of the data's origin. Typically, data are generated at edge infrastructure and transmitted to cloud, where traditional applied. Currently, developers forced resort ad hoc specifically tailored available (for example, devices) when designing, developing, operating applications. Here, novel approach implements real-time edge-computing The authors introduce their...
The ongoing transition towards a networked society requires reliable and secure network infrastructure services. As networks evolve from simple point to systems complex, software-defined, ultra-high capacity reach, distributed cloud environments, new security challenges emerge. EU-funded RECODIS project aims at coordinating fostering research collaboration in Europe on disaster resiliency communication networks. One of the types, considered by Working Group (WG) 4, are deliberate human-made...
Serverless workflow applications or function choreographies (FCs), which connect serverless functions by data- and control-flow, have gained considerable momentum recently to create more sophisticated as part of Function-as-a-Service (FaaS) platforms. Initial experimental analysis the current support for FCs uncovered important weaknesses, including provider lock-in, limited data-flow control-flow constructs. To overcome some these we introduce Abstract Function Choreography Language (AFCL)...
Accurate load forecasting is crucial for predictive control in many energy domain applications, with significant economic and ecological implications. To address these implications, this study provides an extensive benchmark of state-of-the-art deep learning models short-term communities. Namely, LSTM, xLSTM, Transformers are compared benchmarks such as KNNs, synthetic models, persistence models. This comparison considers different scales aggregation (e.g., number household loads) varying...
Infrastructure-as-Code (IaC) frameworks empower developers to swiftly define and provision their infrastructure with a single click. However, the domain-specific languages (DSLs) utilized for coding often lean towards provider specificity rather than being application-centric. This results in increased developer effort, as they are compelled duplicate data when deploying serverless functions across diverse regions providers within federated FaaS environments. To mitigate this challenge, we...
All e-Assessment systems have several challenges, such as accurate evaluation, security and data privacy, performance, scalability etc. In this paper we focus on performance. We propose a SOA architecture of cloud hosted system which uses elasticity in order to achieve sustainable Our solution consists three subsystems, the first for management, second reports, third on-demand activities during assessments. It reduces overall costs since it minimum resources utilized only e-Assessment....
The speedup is usually limited by two main laws in high-performance computing, that is, the Amdahl's and Gustafson's laws.However, sometimes can reach far beyond linear speedup, known as superlinear which means greater than number of processors are used.Although not a new concept many authors have already reported its existence, most them it side effect, without explaining why how happening.In this paper, we analyze several different types define taxonomy for them.Additionally, present...
In this paper, a deviceless edge computing solution is analysed in contrast to the "traditional" server solution. We compare centralized with distributed approaches for horizontal offloading of data and computations, analyze requirements protocols realize such solutions. The proposed solutions are more energy-efficient (IoT devices may work longer period without recharging), provide scalable elastic environment extended fault tolerance.
Hardware-based courses in computer science studies require much effort from both students and teachers. The most important part of students' learning is attending person actively working on laboratory exercises hardware equipment. This paper deals with a specific group students, those who are marginalized by not being able to regularly attend lessons perform the exercises. "busy students" includes older travel, employed or parents. In particular it describes how such achieved impressive...
Cloud applications often benefit from deployment on serverless technology Function-as-a-Service (FaaS), which may instantly spawn numerous functions and charges users for the period when are running. Maximum is achieved orchestrated in a workflow or function choreographies (FCs). However, many provider limitations specific FaaS, such as maximum concurrency duration increase failure rate, can severely hamper execution of entire FCs. Current support resilience limited to retries try-catch,...
Amazon EC2's spot instances (SIs) represent a competitive Cloud resource in terms of price compared to reliable and fixed options. The drawback, however, is that SIs may not always be available they can revoked at any given time. In this paper, we describe comprehensive experimental evaluation for EC2 characterize their performance behavior three different regions each which continent. We the life cycle with most important phases an SI, introduce relevant events prevent user from obtaining...
Resilience against disaster scenarios is essential to network operators, not only because of the potential economic impact a but also communication networks form basis crisis management. COST RECODIS aims at studying measures, rules, techniques and prediction mechanisms for different scenarios. This paper gives an overview solutions in context technology-related disasters. After general overview, focuses on resilient Software Defined Networks.
Many researchers reported considerable delay of up to a few seconds when invoking serverless functions for the first time. This phenomenon, which is known as cold start, affects even more users are running multiple orchestrated in workflow. However, many cases need instantly spawn numerous functions, usually part parallel loops. In this paper, we introduce start and analyze behavior three Function-as-a-Service (FaaS) providers AWS Lambda, Google Cloud Functions, IBM Functions loops, both...
Serverless is a popular paradigm for expressing compute-intensive applications as serverless workflows. In practice, significant portion of the computing typically offloaded to various Backend-as-a-Service (BaaS) cloud services. There cent rise federated and Sky offers cost performance advantages these BaaS-enabled However, due vendor lock-in lack service interoperability, many challenges remain that impact development, deployment, scheduling workflows in infrastructures. This paper...
SUMMARY The realization of modern processors is based on a multicore architecture with increasing number cores per processor. Multicore are often designed such that some level the cache hierarchy shared among cores. Usually, last several or all (e.g., L3 cache) and each core possesses private low caches L1 L2 caches). Superlinear speedup possible for matrix multiplication algorithm executed in memory multiprocessor due to existence superlinear region. It region where requirements storage...
Cloud, fog and dew computing concepts offer elastic resources that can serve scalable services. These be scaled horizontally or vertically. The former is more powerful, which increases the number of same machines (scaled out) to retain performance service. However, this scaling tightly connected with existence a balancer in front will balance load among end points. In paper, we present successful implementation low-level balancer, implemented on network layer. scalability tested by series...
<p style="margin: 0px;">The computer architecture and organization course is essential in all science engineering programs, the most selected liked elective for related disciplines. However, attractiveness brings a new challenge, it requires lot of effort by instructor, to explain rather complicated concepts beginners or those who study The usage visual simulators can improve both teaching learning processes. overall goal twofold: 1)~to enable environment basic 2)~to increase student's...
Migrating the virtual machines from on-premise to cloud raises new security challenges for a company. A potential threat tenants are not only Internet hackers, but also service provider and other co-tenants, due multi-tenancy feature. The provider's is challenged by tenants, as well. Deploying open source additional since intruders have access code can assess its vulnerabilities. In this paper, we thoroughly vulnerabilities of OpenStack framework, one most used frameworks today. We server...