arXiv subjects

Spyros Lalis

Publications and source records attributed to Spyros Lalis.

12 recordsLinked to original sources

arXiv2026

Green or Fast? Learning to Balance Cold Starts and Idle Carbon in Serverless Computing

Serverless computing simplifies cloud deployment but introduces new challenges in managing service latency and carbon emissions. Reducing cold-start latency requires retaining warm function instances, while minimizing carbon emissions favors reclaiming idle resources. This balance is further complicated by time-varying grid carbon intensity and varying workload patterns, under which static keep-alive policies are inefficient. We present LACE-RL, a latency-aware and carbon-efficient management framework that formulates serverless pod retention as a sequential decision problem. LACE-RL uses deep reinforcement learning to dynamically tune keep-alive durations, jointly modeling cold-start probability, function-specific latency costs, and real-time carbon intensity. Using the Huawei Public Cloud Trace, we show that LACE-RL reduces cold starts by 51.69% and idle keep-alive carbon emissions by 77.08% compared to Huawei's static policy, while achieving better latency-carbon trade-offs than state-of-the-art heuristic and single-objective baselines, approaching Oracle performance.

cs.DC↗

arXiv2026

DPUConfig: Optimizing ML Inference in FPGAs Using Reinforcement Learning

Heterogeneous embedded systems, with diverse computing elements and accelerators such as FPGAs, offer a promising platform for fast and flexible ML inference, which is crucial for services such as autonomous driving and augmented reality, where delays can be costly. However, efficiently allocating computational resources for deep learning applications in FPGA-based systems is a challenging task. A Deep Learning Processor Unit (DPU) is a parameterizable FPGA-based accelerator module optimized for ML inference. It supports a wide range of ML models and can be instantiated multiple times within a single FPGA to enable concurrent execution. This paper introduces DPUConfig, a novel runtime management framework, based on a custom Reinforcement Learning (RL) agent, that dynamically selects optimal DPU configurations by leveraging real-time telemetry data monitoring, system utilization, power consumption, and application performance to inform its configuration selection decisions. The experimental evaluation demonstrates that the RL agent achieves energy efficiency 95% (on average) of the optimal attainable energy efficiency for several CNN models on the Xilinx Zynq UltraScale+ MPSoC ZCU102.

cs.AR↗

arXiv2025

Using Machine Learning to Take Stay-or-Go Decisions in Data-driven Drone Missions

Drones are becoming indispensable in many application domains. In data-driven missions, besides sensing, the drone must process the collected data at runtime to decide whether additional action must be taken on the spot, before moving to the next point of interest. If processing does not reveal an event or situation that requires such an action, the drone has waited in vain instead of moving to the next point. If, however, the drone starts moving to the next point and it turns out that a follow-up action is needed at the previous point, it must spend time to fly-back. To take this decision, we propose different machine-learning methods based on branch prediction and reinforcement learning. We evaluate these methods for a wide range of scenarios where the probability of event occurrence changes with time. Our results show that the proposed methods consistently outperform the regression-based method proposed in the literature and can significantly improve the worst-case mission time by up to 4.1x. Also, the achieved median mission time is very close, merely up to 2.7% higher, to that of a method with perfect knowledge of the current underlying event probability at each point of interest.

cs.RO↗

arXiv2025

Black-box Adversarial Attacks on CNN-based SLAM Algorithms

Continuous advancements in deep learning have led to significant progress in feature detection, resulting in enhanced accuracy in tasks like Simultaneous Localization and Mapping (SLAM). Nevertheless, the vulnerability of deep neural networks to adversarial attacks remains a challenge for their reliable deployment in applications, such as navigation of autonomous agents. Even though CNN-based SLAM algorithms are a growing area of research there is a notable absence of a comprehensive presentation and examination of adversarial attacks targeting CNN-based feature detectors, as part of a SLAM system. Our work introduces black-box adversarial perturbations applied to the RGB images fed into the GCN-SLAM algorithm. Our findings on the TUM dataset [30] reveal that even attacks of moderate scale can lead to tracking failure in as many as 76% of the frames. Moreover, our experiments highlight the catastrophic impact of attacking depth instead of RGB input images on the SLAM system.

cs.RO↗

arXiv2024

Should I Stay or Should I Go: A Learning Approach for Drone-based Sensing Applications

Multicopter drones are becoming a key platform in several application domains, enabling precise on-the-spot sensing and/or actuation. We focus on the case where the drone must process the sensor data in order to decide, depending on the outcome, whether it needs to perform some additional action, e.g., more accurate sensing or some form of actuation. On the one hand, waiting for the computation to complete may waste time, if it turns out that no further action is needed. On the other hand, if the drone starts moving toward the next point of interest before the computation ends, it may need to return back to the previous point, if some action needs to be taken. In this paper, we propose a learning approach that enables the drone to take informed decisions about whether to wait for the result of the computation (or not), based on past experience gathered from previous missions. Through an extensive evaluation, we show that the proposed approach, when properly configured, outperforms several static policies, up to 25.8%, over a wide variety of different scenarios where the probability of some action being required at a given point of interest remains stable as well as for scenarios where this probability varies in time.

cs.RO↗

arXiv2024

MP-SL: Multihop Parallel Split Learning

Federated Learning (FL) stands out as a widely adopted protocol facilitating the training of Machine Learning (ML) models while maintaining decentralized data. However, challenges arise when dealing with a heterogeneous set of participating devices, causing delays in the training process, particularly among devices with limited resources. Moreover, the task of training ML models with a vast number of parameters demands computing and memory resources beyond the capabilities of small devices, such as mobile and Internet of Things (IoT) devices. To address these issues, techniques like Parallel Split Learning (SL) have been introduced, allowing multiple resource-constrained devices to actively participate in collaborative training processes with assistance from resourceful compute nodes. Nonetheless, a drawback of Parallel SL is the substantial memory allocation required at the compute nodes, for instance training VGG-19 with 100 participants needs 80 GB. In this paper, we introduce Multihop Parallel SL (MP-SL), a modular and extensible ML as a Service (MLaaS) framework designed to facilitate the involvement of resource-constrained devices in collaborative and distributed ML model training. Notably, to alleviate memory demands per compute node, MP-SL supports multihop Parallel SL-based training. This involves splitting the model into multiple parts and utilizing multiple compute nodes in a pipelined manner. Extensive experimentation validates MP-SL's capability to handle system heterogeneity, demonstrating that the multihop configuration proves more efficient than horizontally scaled one-hop Parallel SL setups, especially in scenarios involving more cost-effective compute nodes.

cs.LG↗

arXiv2023

Flexible Computation Offloading at the Edge for Autonomous Drones with Uncertain Flight Times

An ever increasing number of applications can employ aerial unmanned vehicles, or so-called drones, to perform different sensing and possibly also actuation tasks from the air. In some cases, the data that is captured at a given point has to be processed before moving to the next one. Drones can exploit nearby edge servers to offload the computation instead of performing it locally. However, doing this in a naive way can be suboptimal if servers have limited computing resources and drones have limited energy resources. In this paper, we propose a protocol and resource reservation scheme for each drone and edge server to decide, in a dynamic and fully decentralized way, whether to offload the computation and respectively whether to accept such an offloading requests, with the objective to evenly reduce the drones' mission times. We evaluate our approach through extensive simulation experiments, showing that it can significantly reduce the mission times compared to a no-offloading scenario by up to 26.2%, while outperforming an offloading schedule that has been computed offline by up to 7.4% as well as a purely opportunistic approach by up to 23.9%.

cs.RO↗

arXiv2021

Artificial neural networks for online error detection

Hardware reliability is adversely affected by the downscaling of semiconductor devices and the scale-out of systems necessitated by modern applications. Apart from crashes, this unreliability often manifests as silent data corruptions (SDCs), affecting application output. Therefore, we need low-cost and low-human-effort solutions to reduce the incidence rate and the effects of SDCs on the quality of application outputs. We propose Artificial Neural Networks (ANNs) as an effective mechanism for online error detection. We train ANNs using software fault injection. We find that the average overhead of our approach, followed by a costly error correction by re-execution, is 6.45% in terms of CPU cycles. We also report that ANNs discover 94.85% of faults thereby resulting in minimal output quality degradation. To validate our approach we overclock ARM Cortex A53 CPUs, execute benchmarks on them and record the program outputs. ANNs prove to be an efficient error detection mechanism, better than a state of the art approximate error detection mechanism (Topaz), both in terms of performance (12.81% CPU overhead) and quality of application output (94.11% detection coverage).

cs.DC↗

arXiv2021

IPLS : A Framework for Decentralized Federated Learning

The proliferation of resourceful mobile devices that store rich, multidimensional and privacy-sensitive user data motivate the design of federated learning (FL), a machine-learning (ML) paradigm that enables mobile devices to produce an ML model without sharing their data. However, the majority of the existing FL frameworks rely on centralized entities. In this work, we introduce IPLS, a fully decentralized federated learning framework that is partially based on the interplanetary file system (IPFS). By using IPLS and connecting into the corresponding private IPFS network, any party can initiate the training process of an ML model or join an ongoing training process that has already been started by another party. IPLS scales with the number of participants, is robust against intermittent connectivity and dynamic participant departures/arrivals, requires minimal resources, and guarantees that the accuracy of the trained model quickly converges to that of a centralized FL framework with an accuracy drop of less than one per thousand.

cs.NI↗

arXiv2018

RF Jamming Classification using Relative Speed Estimation in Vehicular Wireless Networks

Wireless communications are vulnerable against radio frequency (RF) jamming which might be caused either intentionally or unintentionally. A particular subset of wireless networks, vehicular ad-hoc networks (VANET) which incorporate a series of safety-critical applications, may be a potential target of RF jamming with detrimental safety effects. To ensure secure communication and defend it against this type of attacks, an accurate detection scheme must be adopted. In this paper we introduce a detection scheme that is based on supervised learning. The machine-learning algorithms, KNearest Neighbors (KNN) and Random Forests (RF), utilize a series of features among which is the metric of the variations of relative speed (VRS) between the jammer and the receiver that is passively estimated from the combined value of the useful and the jamming signal at the receiver. To the best of our knowledge, this metric has never been utilized before in a machine-learning detection scheme in the literature. Through offline training and the proposed KNN-VRS, RF-VRS classification algorithms, we are able to efficiently detect various cases of Denial of Service Attacks (DoS) jamming attacks, differentiate them from cases of interference as well as foresee a potential danger successfully and act accordingly.

cs.CR↗

arXiv2016

Coordinated Broadcast-based Request-Reply and Group Management for Tightly-Coupled Wireless Systems

As the domain of cyber-physical systems continues to grow, an increasing number of tightly-coupled distributed applications will be implemented on top of wireless networking technologies. Some of these applications, including collaborative robotic teams, work in a coordinated fashion, whereby a distinguished node takes control decisions and sends commands to other nodes, which in turn perform the requested action/operation and send back a reply/acknowledgment. The implementation of such interactions via reliable point-to-point flows may lead to a significant performance degradation due to collisions, especially when the system operates close to the capacity of the communication channel. We propose a coordinated protocol which exploits the broadcast nature of the wireless medium in order to support this application-level interaction with a minimal number of message transmissions and predictable latency. The protocol also comes with group management functionality, allowing new processes to join and existing processes to leave the group in a controlled way. We evaluate a prototype implementation over WiFi, using a simulated setup as well as a physical testbed. Our results show that the proposed protocol can achieve significantly better performance compared to point-to- point approaches, and remains fully predictable and dependable even when operating close to the wireless channel capacity.

cs.DC↗

arXiv2014

A Programming Model and Runtime System for Significance-Aware Energy-Efficient Computing

Reducing energy consumption is one of the key challenges in computing technology. One factor that contributes to high energy consumption is that all parts of the program are considered equally significant for the accuracy of the end-result. However, in many cases, parts of computations can be performed in an approximate way, or even dropped, without affecting the quality of the final output to a significant degree. In this paper, we introduce a task-based programming model and runtime system that exploit this observation to trade off the quality of program outputs for increased energy-efficiency. This is done in a structured and flexible way, allowing for easy exploitation of different execution points in the quality/energy space, without code modifications and without adversely affecting application performance. The programmer specifies the significance of tasks, and optionally provides approximations for them. Moreover, she provides hints to the runtime on the percentage of tasks that should be executed accurately in order to reach the target quality of results. The runtime system can apply a number of different policies to decide whether it will execute each individual less-significant task in its accurate form, or in its approximate version. Policies differ in terms of their runtime overhead but also the degree to which they manage to execute tasks according to the programmer's specification. The results from experiments performed on top of an Intel-based multicore/multiprocessor platform show that, depending on the runtime policy used, our system can achieve an energy reduction of up to 83% compared with a fully accurate execution and up to 35% compared with an approximate version employing loop perforation. At the same time, our approach always results in graceful quality degradation.

cs.PL↗