arXiv subjects

Ivana Podnar Žarko

Publications and source records attributed to Ivana Podnar Žarko.

11 recordsLinked to original sources

arXiv2026

A Kubernetes-Native Request Router for Quality-Aware Inference Serving in the Computing Continuum

We introduce Adaptive Score-based Routing Balancer (ASRB), a dynamic, score-based request routing mechanism for Kubernetes-based service deployments over the computing continuum. ASRB jointly considers infrastructure-level information, response time measurements, and application-level quality indicators, with a particular focus on serving Machine Learning (ML) workloads. For these workloads, ASRB balances requests over service instances deployed in the continuum, following service provider-defined policies encoded as weighted combinations of QoS criteria to flexibly address latency-accuracy trade-offs. To drive routing decisions and swiftly adapt to changes in the operating environment, ASRB monitors a range of runtime metrics across multiple system layers. To deal with the associated monitoring overhead, particularly important for large-scale deployments, it selectively and adaptively controls monitoring intensity without sacrificing on routing quality. ASRB is implemented without requiring any modifications to Kubernetes, making it straightforward to deploy and operate in existing cluster environments. Our testbed experiments demonstrate the versatility of ASRB: When tuned for latency reduction, it achieves at least 10 ms lower mean response time compared with latency-oriented state-of-the-art routing mechanisms, while it achieves higher accuracy when this is prioritized through specific configurations, thus enabling flexible and operator-controllable trade-offs. At the same time, it attains reduced failure rates, higher responsiveness to changes in the operating environment, and up to ~70% less monitoring cost than relevant state-of-the-art solutions, at the potential expense of only a modest latency penalty in some configurations.

cs.DC↗

arXiv2026

QoS-Aware Load Balancing in the Computing Continuum via Multi-Player Bandits

As computation shifts from the cloud to the edge to reduce processing latency and network traffic, the resulting Computing Continuum (CC) creates a dynamic environment where meeting strict Quality of Service (QoS) requirements and avoiding service instance overload becomes challenging. Existing methods often prioritize global metrics and overlook per-client QoS, which is crucial for latency-sensitive and reliability-critical applications. We propose QEdgeProxy, a decentralized QoS-aware load balancer that acts as a proxy between IoT devices and service instances in the CC. We formulate the load balancing problem as a Multi-Player Multi-Armed Bandit (MP-MAB) with heterogeneous rewards: Each load balancer autonomously selects service instances to maximize the probability of meeting its clients' QoS requirements by using Kernel Density Estimation (KDE) to estimate QoS success probabilities. Our load-balancing algorithm also incorporates an adaptive exploration mechanism to recover rapidly from performance shifts and non-stationary conditions. We present a Kubernetes-native QEdgeProxy implementation and evaluate it on an emulated CC testbed deployed on a K3s cluster with realistic network conditions and a latency-sensitive edge-AI workload. Results show that QEdgeProxy significantly outperforms proximity-based and reinforcement-learning baselines in per-client QoS satisfaction, while adapting effectively to load surges and changes in instance availability.

cs.NI↗

arXiv2025

Adaptive Graph Pruning with Sudden-Events Evaluation for Traffic Prediction using Online Semi-Decentralized ST-GNNs

Spatio-Temporal Graph Neural Networks (ST-GNNs) are well-suited for processing high-frequency data streams from geographically distributed sensors in smart mobility systems. However, their deployment at the edge across distributed compute nodes (cloudlets) createssubstantial communication overhead due to repeated transmission of overlapping node features between neighbouring cloudlets. To address this, we propose an adaptive pruning algorithm that dynamically filters redundant neighbour features while preserving the most informative spatial context for prediction. The algorithm adjusts pruning rates based on recent model performance, allowing each cloudlet to focus on regions experiencing traffic changes without compromising accuracy. Additionally, we introduce the Sudden Event Prediction Accuracy (SEPA), a novel event-focused metric designed to measure responsiveness to traffic slowdowns and recoveries, which are often missed by standard error metrics. We evaluate our approach in an online semi-decentralized setting with traditional FL, server-free FL, and Gossip Learning on two large-scale traffic datasets, PeMS-BAY and PeMSD7-M, across short-, mid-, and long-term prediction horizons. Experiments show that, in contrast to standard metrics, SEPA exposes the true value of spatial connectivity in predicting dynamic and irregular traffic. Our adaptive pruning algorithm maintains prediction accuracy while significantly lowering communication cost in all online semi-decentralized settings, demonstrating that communication can be reduced without compromising responsiveness to critical traffic events.

cs.LG↗

arXiv2025

Semi-decentralized Training of Spatio-Temporal Graph Neural Networks for Traffic Prediction

In smart mobility, large networks of geographically distributed sensors produce vast amounts of high-frequency spatio-temporal data that must be processed in real time to avoid major disruptions. Traditional centralized approaches are increasingly unsuitable to this task, as they struggle to scale with expanding sensor networks, and reliability issues in central components can easily affect the whole deployment. To address these challenges, we explore and adapt semi-decentralized training techniques for Spatio-Temporal Graph Neural Networks (ST-GNNs) in smart mobility domain. We implement a simulation framework where sensors are grouped by proximity into multiple cloudlets, each handling a subgraph of the traffic graph, fetching node features from other cloudlets to train its own local ST-GNN model, and exchanging model updates with other cloudlets to ensure consistency, enhancing scalability and removing reliance on a centralized aggregator. We perform extensive comparative evaluation of four different ST-GNN training setups -- centralized, traditional FL, server-free FL, and Gossip Learning -- on large-scale traffic datasets, the METR-LA and PeMS-BAY datasets, for short-, mid-, and long-term vehicle speed predictions. Experimental results show that semi-decentralized setups are comparable to centralized approaches in performance metrics, while offering advantages in terms of scalability and fault tolerance. In addition, we highlight often overlooked issues in existing literature for distributed ST-GNNs, such as the variation in model performance across different geographical areas due to region-specific traffic patterns, and the significant communication overhead and computational costs that arise from the large receptive field of GNNs, leading to substantial data transfers and increased computation of partial embeddings.

cs.LG↗

arXiv2025

Reactive Orchestration for Hierarchical Federated Learning Under a Communication Cost Budget

Deploying a Hierarchical Federated Learning (HFL) pipeline across the computing continuum (CC) requires careful organization of participants into a hierarchical structure with intermediate aggregation nodes between FL clients and the global FL server. This is challenging to achieve due to (i) cost constraints, (ii) varying data distributions, and (iii) the volatile operating environment of the CC. In response to these challenges, we present a framework for the adaptive orchestration of HFL pipelines, designed to be reactive to client churn and infrastructure-level events, while balancing communication cost and ML model accuracy. Our mechanisms identify and react to events that cause HFL reconfiguration actions at runtime, building on multi-level monitoring information (model accuracy, resource availability, resource cost). Moreover, our framework introduces a generic methodology for estimating reconfiguration costs to continuously re-evaluate the quality of adaptation actions, while being extensible to optimize for various HFL performance criteria. By extending the Kubernetes ecosystem, our framework demonstrates the ability to react promptly and effectively to changes in the operating environment, making the best of the available communication cost budget and effectively balancing costs and ML performance at runtime.

cs.DC↗

arXiv2024

Reducing Communication Overhead in the IoT-Edge-Cloud Continuum: A Survey on Protocols and Data Reduction Strategies

The adoption of the Internet of Things (IoT) deployments has led to a sharp increase in network traffic as a vast number of IoT devices communicate with each other and IoT services through the IoT-edge-cloud continuum. This network traffic increase poses a major challenge to the global communications infrastructure since it hinders communication performance and also puts significant strain on the energy consumption of IoT devices. To address these issues, efficient and collaborative IoT solutions which enable information exchange while reducing the transmitted data and associated network traffic are crucial. This survey provides a comprehensive overview of the communication technologies and protocols as well as data reduction strategies that contribute to this goal. First, we present a comparative analysis of prevalent communication technologies in the IoT domain, highlighting their unique characteristics and exploring the potential for protocol composition and joint usage to enhance overall communication efficiency within the IoT-edge-cloud continuum. Next, we investigate various data traffic reduction techniques tailored to the IoT-edge-cloud context and evaluate their applicability and effectiveness on resource-constrained and devices. Finally, we investigate the emerging concepts that have the potential to further reduce the communication overhead in the IoT-edge-cloud continuum, including cross-layer optimization strategies and Edge AI techniques for IoT data reduction. The paper offers a comprehensive roadmap for developing efficient and scalable solutions across the layers of the IoT-edge-cloud continuum that are beneficial for real-time processing to alleviate network congestion in complex IoT environments.

cs.NI↗

arXiv2024

Inference Load-Aware Orchestration for Hierarchical Federated Learning

Hierarchical federated learning (HFL) designs introduce intermediate aggregator nodes between clients and the global federated learning server in order to reduce communication costs and distribute server load. One side effect is that machine learning model replication at scale comes "for free" as part of the HFL process: model replicas are hosted at the client end, intermediate nodes, and the global server level and are readily available for serving inference requests. This creates opportunities for efficient model serving but simultaneously couples the training and serving processes and calls for their joint orchestration. This is particularly important for continual learning, where serving a model while (re)training it periodically, upon specific triggers, or continuously, takes place over shared infrastructure spanning the computing continuum. Consequently, training and inference workloads can interfere with detrimental effects on performance. To address this issue, we propose an inference load-aware HFL orchestration scheme, which makes informed decisions on HFL configuration, considering knowledge about inference workloads and the respective processing capacity. Applying our scheme to a continual learning use case in the transportation domain, we demonstrate that by optimizing aggregator node placement and device-aggregator association, significant inference latency savings can be achieved while communication costs are drastically reduced compared to flat centralized federated learning.

cs.DC↗

arXiv2024

QEdgeProxy: QoS-Aware Load Balancing for IoT Services in the Computing Continuum

While various service orchestration aspects within Computing Continuum (CC) systems have been extensively addressed, including service placement, replication, and scheduling, an open challenge lies in ensuring uninterrupted data delivery from IoT devices to running service instances in this dynamic environment, while adhering to specific Quality of Service (QoS) requirements and balancing the load on service instances. To address this challenge, we introduce QEdgeProxy, an adaptive and QoS-aware load balancing framework specifically designed for routing client requests to appropriate IoT service instances in the CC. QEdgeProxy integrates naturally within Kubernetes, adapts to changes in dynamic environments, and manages to seamlessly deliver data to IoT service instances while consistently meeting QoS requirements and effectively distributing load across them. This is verified by extensive experiments over a realistic K3s cluster with instance failures and network variability, where QEdgeProxy outperforms both Kubernetes built-in mechanisms and a state-of-the-art solution, while introducing minimal computational overhead.

cs.DC↗

arXiv2024

Autonomous Driving with a Deep Dual-Model Solution for Steering and Braking Control

The technology of autonomous driving is currently attracting a great deal of interest in both research and industry. In this paper, we present a deep learning dual-model solution that uses two deep neural networks for combined braking and steering in autonomous vehicles. Steering control is achieved by applying the NVIDIA's PilotNet model to predict the steering wheel angle, while braking control relies on the use of MobileNet SSD. Both models rely on a single front-facing camera for image input. The MobileNet SSD model is suitable for devices with constrained resources, whereas PilotNet struggles to operate efficiently on smaller devices with limited resources. To make it suitable for such devices, we modified the PilotNet model using our own original network design and reduced the number of model parameters and its memory footprint by approximately 60%. The inference latency has also been reduced, making the model more suitable to operate on resource-constrained devices. The modified PilotNet model achieves similar loss and accuracy compared to the original PilotNet model. When evaluated in a simulated environment, both autonomous driving systems, one using the modified PilotNet model and the other using the original PilotNet model for steering, show similar levels of autonomous driving performance.

cs.RO↗

arXiv2021

Aurora: a probabilistic algorithm for distributed ledgers enabling trustless synchronization and transaction inclusion verification

A new node joining a blockchain network first synchronizes with the network to verify ledger state by downloading the entire ledger history. We present Aurora, a probabilistic algorithm that \textit{identifies honest nodes} for transient or persistent communication in the presence of malicious nodes in a blockchain network, or ceases operation if it is unable to do so. The algorithm allows a node joining the network to make an informed decision about its next synchronization step or to verify that a transaction is contained in a valid ledger block without downloading the entire ledger or even the header chain. The algorithm constructs a Directed Acyclic Graph on the network topology to select a subset of nodes including a predefined number of honest nodes with a given probability. It is evaluated on a Bitcoin-like network topology using an open-source blockchain simulator. We investigate algorithm performance and analyze its communication complexity. Our results show that the algorithm facilitates trustless interactions of resource-constrained nodes with a blockchain network containing malicious nodes to enable a leaner initial blockchain download or an efficient and trustless transaction inclusion verification. Moreover, the algorithm can be implemented without any changes to the existing consensus protocol.

cs.CR↗

arXiv2018

Distributed Ledger Technology: Blockchain Compared to Directed Acyclic Graph

Nowadays, blockchain is becoming a synonym for distributed ledger technology. However, blockchain is only one of the specializations in the field and is currently well-covered in existing literature, but mostly from a cryptographic point of view. Besides blockchain technology, a new paradigm is gaining momentum: directed acyclic graphs. The contribution presented in this paper is twofold. Firstly, the paper analyzes distributed ledger technology with an emphasis on the features relevant to distributed systems. Secondly, the paper analyses the usage of directed acyclic graph paradigm in the context of distributed ledgers, and compares it with the blockchain-based solutions. The two paradigms are compared using representative implementations: Bitcoin, Ethereum and Nano. We examine representative solutions in terms of the applied data structures for maintaining the ledger, consensus mechanisms, transaction confirmation confidence, ledger size, and scalability.

cs.DC↗