Modeling Digital Twins of Kubernetes-Based Applications

Borsatti, Davide; Cerroni, Walter; Foschini, Luca; Grabarnik, Genady Ya.; Poltronieri, Filippo; Scotece, Domenico; Shwartz, Larisa; Stefanelli, Cesare; Tortonesi, Mauro; Zaccarini, Mattia

doi:10.1109/iscc58397.2023.10217853

Cited by 7 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CI solutions essentially build the knowledge of the target function by collecting the outcomes of their sampling in a memory pool that evolves over time to explore the mostpromising parts of the search space, usually through algorithms that leverage metaphors from the evolutionary and biological world. This scheme has proven very effective in the optimization of static systems, but requires some additional attention in systems with strong dynamical components, as is often the case in the Cloud Continuum [36,41]. In fact, applying CI solutions to solve dynamic and expensive optimization problems opens the problem of which parts (if any) of the memory need to be invalidated and discarded.…”

Section: Cloud Continuummentioning

confidence: 99%

Reinforcement Learning vs. Computational Intelligence: Comparing Service Management Approaches for the Cloud Continuum

Poltronieri,

Stefanelli,

Tortonesi

et al. 2023

Future Internet

Self Cite

View full text Add to dashboard Cite

Modern computing environments, thanks to the advent of enabling technologies such as Multi-access Edge Computing (MEC), effectively represent a Cloud Continuum, a capillary network of computing resources that extend from the Edge of the network to the Cloud, which enables a dynamic and adaptive service fabric. Efficiently coordinating resource allocation, exploitation, and management in the Cloud Continuum represents quite a challenge, which has stimulated researchers to investigate innovative solutions based on smart techniques such as Reinforcement Learning and Computational Intelligence. In this paper, we make a comparison of different optimization algorithms and a first investigation of how they can perform in this kind of scenario. Specifically, this comparison included the Deep Q-Network, Proximal Policy Optimization, Genetic Algorithms, Particle Swarm Optimization, Quantum-inspired Particle Swarm Optimization, Multi-Swarm Particle Optimization, and the Grey-Wolf Optimizer. We demonstrate how all approaches can solve the service management problem with similar performance—with a different sample efficiency—if a high number of samples can be evaluated for training and optimization. Finally, we show that, if the scenario conditions change, Deep-Reinforcement-Learning-based approaches can exploit the experience built during training to adapt service allocation according to the modified conditions.

show abstract

Section: Cloud Continuummentioning

confidence: 99%

Reinforcement Learning vs. Computational Intelligence: Comparing Service Management Approaches for the Cloud Continuum

Poltronieri,

Stefanelli,

Tortonesi

et al. 2023

Future Internet

Self Cite

View full text Add to dashboard Cite

show abstract

Graph Attention Networks and Deep Q-Learning for Service Mesh Optimization: A Digital Twinning Approach

Khan,

Taheri,

Kassler

et al. 2024

ICC 2024 - IEEE International Conference on Communications

View full text Add to dashboard Cite

VOICE: Value-of-Information for Compute Continuum Ecosystems

Zaccarini,

Cantelli,

Fazio

et al. 2024

2024 27th Conference on Innovation in Clouds, Internet and Networks (ICIN)

Self Cite

View full text Add to dashboard Cite

The "Compute Continuum" (CC), which encompasses and harmonizes the Edge, Fog, and Cloud computing paradigms, emerges as an opportunity for an efficient deployment and adoption of innovative services ranging from eHealth to virtual reality. However, the effective management of services all along the CC presents the challenge of service deployment in relatively resource scarce or poorly connected locations. We argue that CC environments call for novel resource management strategies with a holistic perspective of the whole computing ecosystem, that also consider the QoS reconfiguration and tuning of services in the resource reallocation process. Tackling these challenges, we introduce VOICE, a cutting-edge platform designed for service management in CC ecosystems. Thanks to a Value of Information (VoI)-based technique that emphasizes the significance of data filtering, VOICE supports dynamic service components allocation throughout the CC. To evaluate the choice of VoI as enablingtechnology for VOICE, we leverage computational intelligence (CI) approaches and compare it with more traditional strategies (e.g. maximization of satisfied requests or latency minimization). Final results show how VoI can be a valuable proposal to optimize the usage of resource scarcity that characterizes CC environments.

show abstract

Modeling Digital Twins of Kubernetes-Based Applications

Cited by 7 publications

References 14 publications

Reinforcement Learning vs. Computational Intelligence: Comparing Service Management Approaches for the Cloud Continuum

Reinforcement Learning vs. Computational Intelligence: Comparing Service Management Approaches for the Cloud Continuum

Graph Attention Networks and Deep Q-Learning for Service Mesh Optimization: A Digital Twinning Approach

VOICE: Value-of-Information for Compute Continuum Ecosystems

Contact Info

Product

Resources

About