TY - JOUR AU - Al-Haboobi, Ali AU - Kecskeméti, Gábor TI - Structure-Aware Scheduling Algorithm for Deadline-Constrained Scientific Workflows in the Cloud JF - INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS J2 - IJACSA VL - 15 PY - 2024 IS - 2 SP - 792 EP - 802 PG - 11 SN - 2158-107X DO - 10.14569/IJACSA.2024.0150280 UR - https://m2.mtmt.hu/api/publication/34695021 ID - 34695021 AB - Cloud computing provides pay-per-use IT services through the Internet. Although cloud computing resources can help scientific workflow applications, several algorithms face the problem of meeting the user’s deadline while minimising the cost of workflow execution. In the cloud, selecting the appropriate type and the exact number of VMs is a major challenge for scheduling algorithms, as tasks in workflow applications are distributed very differently. Depending on workflow requirements, algorithms need to decide when to provision or de-provision VMs. Therefore, this paper presents an algorithm for effectively selecting and allocating resources. Based on the workflow structure, it decides the type and number of VMs to use and when to lease and release them. For some structures, our proposed algorithm uses the initial rented VMs to schedule all tasks of the same workflow to minimise data transfer costs. We evaluate the performance of our algorithm by simulating it with synthetic workflows derived from real scientific workflows with different structures. Our algorithm is compared with Dyna and CGA approaches in terms of meeting deadlines and execution costs. The experimental results show that the proposed algorithm met all the deadline factors of each workflow, while the CGA and Dyna algorithms met 25% and 50%, respectively, of all the deadline factors of all workflows. The results also show that the proposed algorithm provides more cost-efficient schedules than CGA and Dyna. LA - English DB - MTMT ER - TY - JOUR AU - Gavua, Ebenezer Komla AU - Kecskeméti, Gábor TI - ASM-based Formal Model for Analysing Cloud Auto-Scaling Mechanisms JF - INFORMATICA-JOURNAL OF COMPUTING AND INFORMATICS J2 - INFORMATICA (LJUBLJANA) VL - 47 PY - 2023 IS - 6 SP - 75 EP - 96 PG - 22 SN - 0350-5596 DO - 10.31449/inf.v47i6.4622 UR - https://m2.mtmt.hu/api/publication/34025369 ID - 34025369 LA - English DB - MTMT ER - TY - JOUR AU - Sallo, Dilshad Hassan AU - Kecskeméti, Gábor TI - Towards a DISSECT-CF extension for simulating function-as-a-service JF - INTERNATIONAL JOURNAL OF PARALLEL EMERGENT AND DISTRIBUTED SYSTEMS J2 - INT J PARALLEL EMERGENT DISTRIB SYST VL - 38 PY - 2023 IS - 4 SP - 288 EP - 300 PG - 13 SN - 1744-5760 DO - 10.1080/17445760.2023.2220141 UR - https://m2.mtmt.hu/api/publication/33975156 ID - 33975156 LA - English DB - MTMT ER - TY - JOUR AU - MOHAMMED ALI, SHALLAW AU - Kecskeméti, Gábor TI - SeQual: an unsupervised feature selection method for cloud workload traces JF - JOURNAL OF SUPERCOMPUTING J2 - J SUPERCOMPUT VL - 79 PY - 2023 SP - 15079 EP - 15097 PG - 19 SN - 0920-8542 DO - 10.1007/s11227-023-05163-w UR - https://m2.mtmt.hu/api/publication/33809072 ID - 33809072 AB - One challenge of studying cloud workload traces is the lack of available users’ identities. Therefore, clustering methods were used to address this challenge through extracting these identities from workload traces. For better extraction, it is beneficial to select attributes (columns in the traces) for clustering by using feature selection methods. However, the use of general selection methods requires details that are not available for workload traces (e.g. predefined number of clusters). Therefore, in this paper, we present an unsupervised feature selection method for cloud workload traces to identify good candidate attributes for clustering. This method uses Silhouette coefficients to rank attributes that are best for users’ extraction through clustering. The performance of our SeQual method is evaluated in comparison with commonly used (supervised and unsupervised) feature selection methods with the help of clustering quality metrics (i.e. adjusted rand index, entropy and precision). The results show that the SeQual method can compete with the supervised methods and perform better than unsupervised ones, with an average accuracy between 90% and 99%. LA - English DB - MTMT ER - TY - JOUR AU - Sallo, Dilshad Hassan AU - Kecskeméti, Gábor TI - Enriching computing simulators by generating realistic serverless traces JF - JOURNAL OF CLOUD COMPUTING-ADVANCES SYSTEMS AND APPLICATIONS J2 - J CLOUD COMPUT-ADV S VL - 12 PY - 2023 IS - 1 PG - 13 SN - 2192-113X DO - 10.1186/s13677-023-00397-8 UR - https://m2.mtmt.hu/api/publication/33697049 ID - 33697049 LA - English DB - MTMT ER - TY - JOUR AU - Gavua, Ebenezer Komla AU - Kecskeméti, Gábor TI - Improving MapReduce Speculative Executions with Global Snapshots_JOURNAL PAPER JF - INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS J2 - IJACSA VL - 14 PY - 2023 IS - 1 SP - 12 EP - 22 PG - 11 SN - 2158-107X DO - 10.14569/IJACSA.2023.0140102 UR - https://m2.mtmt.hu/api/publication/33615905 ID - 33615905 LA - English DB - MTMT ER - TY - JOUR AU - Márkus, András AU - Al-Haboobi, Ali AU - Kecskeméti, Gábor AU - Kertész, Attila TI - Simulating IoT Workflows in DISSECT-CF-Fog JF - SENSORS J2 - SENSORS-BASEL VL - 23 PY - 2023 IS - 3 PG - 16 SN - 1424-8220 DO - 10.3390/s23031294 UR - https://m2.mtmt.hu/api/publication/33585957 ID - 33585957 N1 - Department of Software Engineering, University of Szeged, Szeged, 6720, Hungary Institute of Information Technology, University of Miskolc, Miskolc, 3515, Hungary Faculty of Computer Science and Mathematics, University of Kufa, Najaf, 54001, Iraq AB - The modelling of IoT applications utilising the resources of cloud and fog computing is not straightforward because they have to support various trigger-based events that make human life easier. The sequence of tasks, such as performing a service call, receiving a data packet in the form of a message sent by an IoT device, and managing actuators or executing a computational task on a virtual machine, are often associated with and composed of IoT workflows. The development and deployment of such IoT workflows and their management systems in real life, including communication and network operations, can be complicated due to high operation costs and access limitations. Therefore, simulation solutions are often applied for such purposes. In this paper, we introduce a novel simulator extension of the DISSECT-CF-Fog simulator that leverages the workflow scheduling and its execution capabilities to model real-life IoT use cases. We also show that state-of-the-art simulators typically omit the IoT factor in the case of the scientific workflow evaluation. Therefore, we present a scalability study focusing on scientific workflows and on the interoperability of scientific and IoT workflows in DISSECT-CF-Fog. LA - English DB - MTMT ER - TY - JOUR AU - Al-Haboobi, Ali AU - Kecskeméti, Gábor TI - Developing a Workflow Management System Simulation for Capturing Internal IaaS Behavioural Knowledge JF - JOURNAL OF GRID COMPUTING J2 - J GRID COMPUT VL - 21 PY - 2023 IS - 1 PG - 26 SN - 1570-7873 DO - 10.1007/s10723-022-09638-7 UR - https://m2.mtmt.hu/api/publication/33395482 ID - 33395482 AB - Scientific workflows are becoming increasingly important for complex scientific applications. Conducting real experiments for large-scale workflows is challenging because they are very expensive and time consuming. A simulation is an alternative approach to a real experiment that can help evaluating the performance of workflow management systems (WMS) and optimise workflow management techniques. Although there are several workflow simulators available today, they are often user-oriented and treat the cloud as a black box. Unfortunately, this behaviour prevents the evaluation of the infrastructure level impact of the various decisions made by the WMSs. To address these issues, we have developed a WMS simulator (called DISSECT-CF-WMS) on DISSECT-CF that exposes the internal details of cloud infrastructures. DISSECT-CF-WMS enables better energy awareness by allowing the study of schedulers for physical machines. It also enables dynamic provisioning to meet the resource needs of the workflow application while considering the provisioning delay of a VM in the cloud. We evaluated our simulation extension by running several workflow applications on a given infrastructure. The experimental results show that we can investigate different schedulers for physical machines on different numbers of virtual machines to reduce energy consumption. The experiments also show that DISSECT-CF-WMS is up to 295× faster than WorkflowSim and still provides equivalent results. The experimental results of auto-scaling show that it can optimise makespan, energy consumption and VM utilisation in contrast to static VM provisioning. LA - English DB - MTMT ER - TY - GEN AU - MOHAMMED ALI, SHALLAW AU - Kecskeméti, Gábor TI - The analysis of cloud workload traces for users' behaviour extraction PY - 2022 UR - https://m2.mtmt.hu/api/publication/34720450 ID - 34720450 LA - English DB - MTMT ER - TY - JOUR AU - Rjeib, Hasanein AU - Kecskeméti, Gábor TI - An investigation on implementing a scenario on different cloud simulators JF - MULTIDISZCIPLINÁRIS TUDOMÁNYOK: A MISKOLCI EGYETEM KÖZLEMÉNYE J2 - MULTIDISZCIPLINÁRIS TUDOMÁNYOK VL - 12 PY - 2022 IS - 3 SP - 256 EP - 263 PG - 8 SN - 2062-9737 DO - 10.35925/j.multi.2022.3.23 UR - https://m2.mtmt.hu/api/publication/33270854 ID - 33270854 AB - Recently, many algorithms have been introduced to provide solutions for Virtual Machine Consolidation in Cloud Data Center. Such Algorithms are usually implemented using many Cloud Simulators, which may have different representation and behavior of cloud infrastructure entities like Data Centers, Physical Machines(PMs), and Virtual Machines(VMs). In this paper, we investigated the impact of using a simulator on the implementation of Cloud infrastructure that is necessary for the Consolidation process, we chose Cloudsim and DISSECTCF simulators in particular since they both support Cloud infrastructure entities. Our aim is to find common entities for cloud simulators that are necessary to build the same infrastructure with particular interest on monitoring the CPU utilization and energy consumption. We report our experience with the implementation on the two Simulators, in addition to the limitation and differences we found during the reproducing process. LA - English DB - MTMT ER -