MMMartin Molan

Martin Molan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Martin Molan has a total of 28 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    EXASAGE: The first data center operational data analysis assistant

    by Junaid Ahmed Khan, Martin Molan, Andrea Bartolini


  • Open Access

    ANZIL: Attention-based network for zero-risk inspection of LiDAR point cloud in self-driving cars

    by Grafika Jati, Martin Molan, Francesco Barchi, Andrea Bartolini, Giuseppe Mercurio, Andrea Acquaviva


  • Open Access

    LIDAROC: Realistic LiDAR Cover Contamination Dataset for Enhancing Autonomous Vehicle Perception Reliability

    by Grafika Jati, Martin Molan, Francesco Barchi, Andrea Bartolini, Giuseppe Mercurio, Andrea Acquaviva

    LiDAR is the foundation of many autonomous vehicle perception systems, so it is essential to study and ensure the integrity and robustness of the data collected by LiDAR. To facilitate future research into robust and resilient LiDAR processing, we present a dataset containing a collection of uncontaminated and realistically contaminated LiDAR samples. We have also studied the effect of contaminants on the object detection task. The state-of-the-art object detection algorithms produce catastrophic errors in detection, such as failure to identify objects, detection of ghost objects, and wrong detection with high confidence. Based on the number of such catastrophic errors, we introduce a novel measure for the LiDAR data's contamination level. The results of the empirical evaluation of the effect of the contaminants on object detection motivate the necessity of further research into contaminant detection and contaminant-resilient data processing, which are all enabled by the dataset collected by this work.


  • GRAAFE: GRaph Anomaly Anticipation Framework for Exascale HPC systems

    by Martin Molan, Mohsen Seyedkazemi Ardebili, Junaid Ahmed Khan, Francesco Beneventi, Daniele Cesarini, Andrea Borghesi, Andrea Bartolini


  • Open Access

    TinyLid: a RISC-V accelerated Neural Network For LiDAR Contaminant Classification in Autonomous Vehicle

    by Grafika Jati, Martin Molan, Francesco Barchi, Andrea Bartolini, Andrea Acquaviva

    LiDAR plays a critical role in autonomous car perception. Hence, the robustness of LiDAR data is imperative. However, malfunctions resulting from sensor cover contaminants are unavoidable and can lead to erroneous data that slowly degrade performance. As such, detecting contamination in LiDAR is essential but remains an open challenge due to varying contaminant types, changing properties over time, and deployment aspects. Automatic classification of the contaminants would enable the automated response (like cleaning the sensor) to ensure the integrity of the data collected by the LiDAR sensor. To minimize the effect on the whole vehicle perception system, the contamination classification has to be performed near the sensor and in a computationally efficient way. To address these challenges, we have conducted a feasibility study of developing an efficient near-sensor machine learning-powered contaminant classification running on the RISC-V architecture. This paper proposes a lightweight 2D CNN network, TinyLid, trained to classify contaminants based on the most comprehensive LiDAR contaminant dataset. The results presented in this paper show that the proposed solution can achieve high classification performance while being computationally efficient and running on hardware with negligible power consumption compared to the LiDAR sensor itself. Specifically, implementing a proposed ML model on a reference RISC-V architecture GAP8 achieves the inference time of 2.575 milliseconds, 6.138 operations/cycle, and uses only 6.8% of 512 KiB L2 memory. The results presented in this work showcase the possibility of increasing the reliability and integrity of the LiDAR-collected sensor data without significant computational or energy consumption impact on the broader system.


  • Open Access

    AutoGrAN: Autonomous Vehicle LiDAR Contaminant Detection using Graph Attention Networks

    by Grafika Jati, Martin Molan, Junaid Ahmed Khan, Francesco Barchi, Andrea Bartolini, Giuseppe Mercurio, Andrea Acquaviva

    Extreme conditions and the integrity of LiDAR sensors influence AI perception models in autonomous vehicles. Lens contamination caused by external particles can compromise LiDAR object detection performance. Automatic contaminant detection is important to improve reliability of sensor information propagated to the user or to object detection algorithms. However, dynamic conditions such as variations in location, distance, and types of objects around the autonomous vehicle make robust and fast contaminant detection significantly challenging. We propose a method for contaminant detection using voxel-based graph transformation to address the challenge of sparse LiDAR data. This method considers LiDAR points as graph nodes and employs a graph attention layer to enhance the accuracy of contaminant detection. Additionally, we introduce cross-environment training and testing on real-world contaminant LiDAR data to ensure high generalization across different environments. Compared with the current state-of-the-art approaches in contaminant detection, our proposed method significantly improves the performance by as much as 0.1575 in F1-score. Consistently achieving F1 scores of 0.936, 0.902, and 0.920 across various testing scenarios, our method demonstrates robustness and adaptability. Requiring 128 milliseconds on a AMD EPYC 74F3 CPU for the end-to-end process, our method is well-suited for an early warning system, outperforming human reaction times, which require at least 390 milliseconds to detect hazards. This significantly contributes to enhancing safety and reliability in the operations of autonomous vehicles.


  • Open Access

    ExaQuery: Proving Data Structure to Unstructured Telemetry Data in Large-Scale HPC

    by Junaid Ahmed Khan, Martin Molan, Matteo Angelinelli, Andrea Bartolini

    High-performance computing (HPC) is the cornerstone of technological advancements in our digital age, but its management is becoming increasingly challenging, particularly as systems approach exascale. Operational data analytics (ODA) and holistic monitoring frameworks aim to alleviate this burden by collecting live telemetry from HPC systems. ODA frameworks rely on NoSQL databases for scalability, with implicit data structures embedded in metric names, necessitating domain knowledge for navigating telemetry data relations. To address the imperative need for explicit representation of relations in telemetry data, we propose a novel ontology for ODA, which we apply to a real HPC installation. The proposed ontology captures relationships between topological components and links hardware components(compute nodes, rack, systems) with job's execution and allocations collected telemetry. This ontology forms the basis for constructing a knowledge graph, enabling graph queries for ODA. Moreover, we propose a comparative analysis of the complexity (expressed in lines of code) and domain knowledge requirement (qualitatively assessed by informed end-users) of complex query implementation with the proposed method and NoSQL methods commonly employed in today's ODAs. We focused on six queries informed by facility managers' daily operations, aiming to benefit not only facility managers but also system administrators and user support. Our comparative analysis demonstrates that the proposed ontology facilitates the implementation of complex queries with significantly fewer lines of code and domain knowledge required as compared to NoSQL methods.


  • Open Access

    Exploring the Utility of Graph Methods in HPC Thermal Modeling

    by Bruno Guindani, Martin Molan, Andrea Bartolini, Luca Benini

    This work critically examines several approaches to temperature prediction for High-Performance Computing (HPC) systems, focusing on component-level and holistic models. In particular, we use publicly available data from the Tier-0 Marconi100 supercomputer and propose models ranging from a room-level Graph Neural Network (GNN) spatial model to node-level models. Our results highlight the importance of correct graph structures and suggest that while graph-based models can enhance predictions in certain scenarios, node-level models remain optimal when data is abundant. These findings contribute to understanding the effectiveness of different modeling approaches in HPC thermal prediction tasks, enabling proactive management of the modeled system.


  • Open Access

    Model for Quantitative Estimation of Functionality Influence on the Final Value of a Software Product

    by Gregor Molan, Gregor Dolinar, Jovan Bojkovski, Radu Prodan, Andrea Borghesi, Martin Molan

    Purpose: The gap between software development requirements and the available resources of software developers continues to widen. This requires changes in the development and organization of software development. Objectives: Presented is a model introducing a quantitative software development management methodology that estimates the relative importance and risk of functionality retention or abundance, which determines the final value of the software product. Method: The final value of the software product is interpreted as a function of the requirements and functionalities, represented as a computational graph (called a software product graph). The software product graph allows the relative importance of functionalities to be estimated by calculating the corresponding partial derivatives of the value function. The risk of not implementing the functionality is estimated by reducing the final value of a product. Validation: This model has been applied to two EU projects: CareHD and vINCI. In vINCI, the functionalities with the most significant added value to the application were developed based on the implemented model and those that brought the least value were abandoned. Optimization was not implemented in the CareHD project and proceeded as initially designed. Consequently, only 71% of the CareHD’s potential value has been realized. Conclusions: Presented model enables rational management and organization of software product development with real-time quantitative evaluation of functionalities impacts, assessment of the risks of omitting them without a significant impact. A quantitative evaluation of the impacts and risks of retention or abundance is possible based on the proposed algorithm, which is the core of the model. This model is a tool for rational organization and development of software products.