The automotive industry faces increasing pressure to accelerate vehicle design and production while adhering to stringent industry regulations. This process relies heavily on resource-intensive Computer-Aided Engineering (CAE) software, such as LS-Dyna, Ansys, and Abaqus, which require substantial computational power and costly, limited-availability software licenses. Priced based on the number of CPU cores utilized, these licenses can become a critical bottleneck, causing job delays, increased costs, and missed deadlines. To address limited computational resources, hybrid cloud infrastructures—integrating public cloud resources with on-premise High-Performance Computing (HPC) clusters—have emerged as a promising solution for scalable CAE job execution. However, while hybrid environments alleviate hardware constraints, they do not resolve software license limitations, which remain a significant hurdle to efficient job scheduling. Effectively leveraging hybrid infrastructures thus requires intelligent scheduling strategies that account for both compute and license constraints—an aspect largely overlooked in existing literature. In this paper, we propose License-Aware Deadline Miss Minimization Backfilling (LADMM_backfill), a novel scheduling algorithm designed to minimize deadline violations for CAE jobs under both license and compute resource constraints across hybrid cloud infrastructures. Our approach extends the traditional EASY backfilling algorithm by integrating real-time license availability into scheduling decisions and introducing dynamic heuristics to prioritize jobs for backfilling. This ensures efficient license utilization while maintaining strict compliance with licensing limits, preventing underutilization and overcommitment. Experimental evaluations show that LADMM_backfill improves license utilization by up to 94

Iterative optimization workflows whose resource demands evolve across iterations, spanning domains such as Computer-Aided Engineering (CAE) and Bayesian Optimization increasingly run on hybrid infrastructures that combine on-premise HPC clusters with public cloud resources. These workflows follow a propose-evaluate-adapt loop with iterationvarying parallelism. But today's batch schedulers and workflow engines treat them as static jobs: resource allocations are fixed at submission time, and hybrid tiers are managed independently. This paper presents ElastiFlow, a runtime framework that implements bidirectional resource negotiation between workflow engines and the scheduler, in response to how workflows grow and shrink, enabling resources to be shifted within the selected tier at iteration boundaries, tracking how the degree of parallelism increases or decreases across iterations. To enable this, ElastiFlow introduces a three-stage allocation methodology that (i) computes a feasible set of resource shapes under per-workflow deadline and budget constraints and a snapshot of tier-aware system state, (ii) selects allocations within a single tier, using an empirical runtime and cost model for heterogeneous nodes and (iii) atomically commits allocation changes while releasing unused resources. We lift classical FCFS and EDF into ELASTIC-FCFS and ELASTIC-EDF, preserving their queueing semantics while allowing runtime resizing driven by negotiation. We evaluate ElastiFlow on seismic optimization workflows, representative of iterationvarying campaigns on a realistic hybrid HPC-cloud testbed. Compared to other baselines, ElastiFlow reduces deadline misses by up to 66.5%, improves cost-performance ratio by up to 84%, lowers average queue time by 39.6%, and simultaneously reduces total cost, demonstrating that negotiation-driven elasticity is an effective abstraction for hybrid resource management.

In the fast-developing fields of smart manufacturing, achieving efficiency is of utmost importance. Cloud Manufacturing (CMfg), in particular, has been widely discussed among various researchers in the recent past. However, it presents the challenge of optimizing cost-effective job assignments across numerous manufacturing providers. To this end, this paper proposes a prediction-enabled scheduling framework (PE-SF) that combines Particle Swarm Optimization-based meta-heuristic algorithm (PSO) and Random Forest (RF) prediction algorithm, represented as PSO-RF, to identify a better manufacturing schedule minimizing the makespan, material cost, and delivery distance of the entire manufacturing workflow. The framework incorporates a cloud-based workflow approach to collaboratively handle the manufacturing requirements of clients across distributed geographical locations. Additionally, the article investigated the performance of PSO-RF with other traditional algorithms such as Genetic Algorithm (GA) and First-In First-Out (FIFO). The evaluation results of scheduling various workload sizes across ten distributed manufacturing units indicate that the proposed PSO-RF algorithm demonstrated a significant efficiency improvement of over 13.3

IoT devices with sensors and actuators are frequently deployed in environments without access to the power grid. These devices are battery powered and might make use of energy harvesting if battery lifetime is too limited. This article focuses on automatically adapting the duty cycle frequency to the predicted available solar energy so that a continuous operation of IoT applications is guaranteed. The implementation is based on a low-cost solar control board that is integrated with the Serverless IoT Framework (SIF), which provides an event-based programming paradigm for microcontroller-based IoT devices. The paper presents a case study where the IoT device sleep time is pro-actively adapted to a predicted sequence of cloudy days to guarantee continuous operation.

Serverless Computing (FaaS) has become a popular paradigm for deep learning inference due to the ease of deployment and pay-per-use benefits. However, current serverless inference platforms encounter the coarse-grained and static GPU resource allocation problems during scaling, which leads to high costs and Service Level Objective (SLO) violations in fluctuating workloads. Meanwhile, current platforms only support horizontal scaling for GPU inferences, thus the cold start problem further exacerbates the problems. In this paper, we propose HAS-GPU, an efficient Hybrid Auto-scaling Serverless architecture with fine-grained GPU allocation for deep learning inferences. HAS-GPU proposes an agile scheduler capable of allocating GPU Streaming Multiprocessor (SM) partitions and time quotas with arbitrary granularity and enables significant vertical quota scalability at runtime. To resolve performance uncertainty introduced by massive fine-grained resource configuration spaces, we propose the Resource-aware Performance Predictor (RaPP). Furthermore, we present an adaptive hybrid auto-scaling algorithm with both horizontal and vertical scaling to ensure inference SLOs and minimize GPU costs. The experiments demonstrated that compared to the mainstream serverless inference platform, HAS-GPU reduces function costs by an average of 10.8x with better SLO guarantees. Compared to state-of-the-art spatio-temporal GPU sharing serverless framework, HAS-GPU reduces function SLO violation by 4.8x and cost by 1.72x on average.

This paper presents the Serverless IoT Framework (SIF), a unified programming framework for IoT applications in the Edge Cloud Continuum based on Serverless Computing. It enables to bundle applications for the cloud, edge, and IoT devices, all written in a FaaS style where state-less functions subscribe to events that are created by various kinds of triggers. The execution of invocation can be optimized in space and time across federated FaaS platform and local execution resources of devices. Functions specify required resources, such as special sensors on device-level, as well as required data and the required response time. The execution of invocations is then optimized based on the given information. This paper introduces the SIF concepts and the implementation platforms on device and edge/cloud levels. We demonstrate the usage and available optimization within an example.

The proliferation of edge devices and the rising need for privacy-preserving machine learning underscore the importance of efficient federated learning (FL) systems in senior homecare-related applications. Traditional FL frameworks risk permanent failure and security risk due to a pre-defined single server implementation and often focus on a single performance metric. To address the challenge, a dynamic leader-based federated learning framework that optimizes multiple metrics such as accuracy, loss, energy consumption, CO2 emissions, and training time is proposed. It features four distinct algorithms, namely the Energy-Efficient Algorithm (EEA), the Carbon-Efficient Algorithm (CEA), the Accuracy-Friendly Algorithm (AFA), and the Eco-Friendly Algorithm (EFA). Experiments were carried out to evaluate the performance of SeniorFed's algorithms and the traditional FedAvg algorithm using sleep apnea detection models. The EEA reduced energy consumption by 5235%; the CEA reduced CO2 emissions by 60.07%; the AFA attained an accuracy improvement of its 3.56%; and the EFA minimized energy consumption by 46.93%, CO2 emissions by 57.45%, and training time by 35.55%.

Malleable MPI applications, which can dynamically adjust their resources during runtime, require efficient and flexible checkpointing systems to maintain fault tolerance. Traditional checkpointing methods often struggle to adapt to these changes, leading to performance bottlenecks. This paper presents iCheck, a resource-aware, application-level checkpointing system designed to support malleable applications. iCheck introduces dynamic pipelining to optimize checkpoint transfers by breaking data into smaller chunks, allowing efficient and scalable transfers. Combined with Remote Direct Memory Access (RDMA), iCheck reduces data transfer latency and minimizes memory overhead. The system adapts dynamically to resource changes by adjusting pipeline size and agent distribution, ensuring minimal disruption to application performance. The impact of pipelining on checkpointing performance is evaluated, demonstrating its ability to efficiently handle dynamic resource allocation in HPC environments. We also examine the system's behavior during component failures, where iCheck temporarily bypasses RDMA and writes checkpoints directly to the Parallel File System (PFS).

Although the performance of mobile devices (MDs) has been steadily improving, advanced applications might suffer long latency for complex functions as well as from draining the battery when continuously used. Function offloading to cloud servers is a proposed technique to solve these challenges. Recently edge computing enables low latency access to resources and, in combination with Function-as-a-service (FaaS), is an obvious target for offloading from MDs. In this paper we introduce a framework for FaaS applications on MDs and function offloading to a FaaS Edge implementation. The scheduling of function invocations to local or remote resources is determined by function-specific offloading policies. The policies are selected periodically according to historic data and function specific performance and energy models. The periodic planning considers the distribution of the application's energy budget across a number of periods and adapts the functionspecific policies for a period. This enables more time consuming planning and fast decision making for individual function invocations. Our results validate the efficiency of the proposed approach, showing about 33 % reduction in energy consumption.

As FPGAs gain popularity for on-demand application acceleration in data center computing, dynamic partial reconfiguration (DPR) has become an effective fine-grained sharing technique for FPGA multiplexing. However, current FPGA sharing encounters partial reconfiguration contention and task execution blocking problems introduced by the DPR, which significantly degrade application performance. In this paper, we propose VersaSlot, an efficient spatio-temporal FPGA sharing system with novel Big.Little slot architecture that can effectively resolve the contention and task blocking while improving resource utilization. For the heterogeneous Big.Little architecture, we introduce an efficient slot allocation and scheduling algorithm, along with a seamless cross-board switching and live migration mechanism, to maximize FPGA multiplexing across the cluster. We evaluate the VersaSlot system on an FPGA cluster composed of the latest Xilinx UltraScale+ FPGAs (ZCU216) and compare its performance against four existing scheduling algorithms. The results demonstrate that VersaSlot achieves up to 13.66x lower average response time than the traditional temporal FPGA multiplexing, and up to 2.19x average response time improvement over the state-of-the-art spatio-temporal sharing systems. Furthermore, VersaSlot enhances the LUT and FF resource utilization by 35% and 29% on average, respectively.

To protect large-scale computing environments necessary to meet increasing computing demand, cloud providers have implemented security measures to monitor Operations and Maintenance (O&M) activities and therefore prevent data loss and service interruption. Command interception systems are used to intercept, assess, and block dangerous Command-line Interface (CLI) commands before they can cause damage. Traditional solutions for command risk assessment include rule-based systems, which require expert knowledge and constant human revision to account for unseen commands. To overcome these limitations, several end-to-end learning systems have been proposed to classify CLI commands. These systems, however, have several other limitations, including the adoption of general-purpose text classifiers, which may not adapt to the language characteristics of scripting languages such as Bash or PowerShell, and may not recognize dangerous commands in the presence of an unbalanced class distribution. In this paper, we propose a transformer-based command risk classification system, which leverages the generalization power of Large Language Models (LLM) to provide accurate classification and the ability to identify rare dangerous commands effectively, by exploiting the power of transfer learning. We verify the effectiveness of our approach on a realistic dataset of production commands and show how to apply our model for other security-related tasks, such as dangerous command interception and auditing of existing rule-based systems.

Federated Learning (FL) is an emerging machine learning paradigm that enables the collaborative training of a shared global model across distributed clients while keeping the data decentralized. Recent works on designing systems for efficient FL have shown that utilizing serverless computing technologies, particularly Function-as-a-Service (FaaS) for FL, can enhance resource efficiency, reduce training costs, and alleviate the complex infrastructure management burden on data holders. However, existing serverless FL systems implicitly assume a uniform global model architecture across all participating clients during training. This assumption fails to address fundamental challenges in practical FL due to the resource and statistical data heterogeneity among FL clients. To address these challenges and enable heterogeneous client models in serverless FL, we utilize Knowledge Distillation (KD) in this paper. Towards this, we propose novel optimized serverless workflows for two popular conventional federated KD techniques, i.e., FedMD and FedDF. We implement these workflows by introducing several extensions to an open-source serverless FL system called FedLess. Moreover, we comprehensively evaluate the two strategies on multiple datasets across varying levels of client data heterogeneity using heterogeneous client models with respect to accuracy, fine-grained training times, and costs. Results from our experiments demonstrate that serverless FedDF is more robust to extreme non-IID data distributions, is faster, and leads to lower costs than serverless FedMD. In addition, compared to the original implementation, our optimizations for particular steps in FedMD and FedDF lead to an average speedup of 3.5x and 1.76x across all datasets.

Companies traditionally dependent on on-premise HPC clusters for simulations are increasingly migrating workloads to the cloud. Cloud computing offers greater flexibility in selecting processors, memory, network bandwidth, along with enhanced resource availability and scalability. Automotive companies rely on computationally intensive numerical simulation tools for CAE (Computer-Aided Engineering), particularly with the growing demand for generative design, which utilizes algorithms to automatically explore a large solution space. This work addresses the gap between the growing runtime demands of these simulations and the limitations of static HPC infrastructure by representing iterative workflows as Directed Acyclic Graphs (DAGs) and optimizing their scheduling. We propose a unified hybrid infrastructure that leverages the elasticity of cloud resources along with existing HPC clusters to maximize computational efficiency, ensure timely completion of simulations, and optimize resource utilization and costs.

With the advent of AWS Lambda in 2014, Serverless Computing, particularly Function-as-a-Service (FaaS), has witnessed growing popularity across various application domains. FaaS enables an application to be decomposed into fine-grained functions that are executed on a FaaS platform. It offers several advantages such as no infrastructure management, a pay-per-use billing policy, and on-demand fine-grained autoscaling. However, despite its advantages, developers today encounter various challenges while adopting FaaS solutions that reduce productivity. These include FaaS platform lock-in, support for diverse function deployment parameters, and diverse interfaces for interacting with FaaS platforms. To address these challenges, we present gFaaS, a novel framework that facilitates the holistic development and management of functions across diverse FaaS platforms. Our framework enables the development of generic functions in multiple programming languages that can be seamlessly deployed across different platforms without modifications. Results from our experiments demonstrate that gFaaS functions perform similarly to native platform-specific functions across various scenarios. A video demonstrating the functioning of gFaaS is available from https://youtu.be/STbb6ykJFf0.

Dynamic resource management and application malleability present numerous opportunities in High-Performance Computing (HPC), enhancing both system-level services and application performance. Recent trends in malleability research, encompassing both application and system dynamism, are building a new era in HPC. Dynamic applications, particularly those leveraging malleable resources, require adaptive checkpointing systems to enhance performance and resource use. Applications can significantly benefit from checkpointing systems becoming dynamic, especially in handling data redistribution during resource changes. Consequently, checkpointing services should also become malleable (or adaptive). Therefore, we propose iCheck, an adaptive application-level checkpoint management system that caters to malleable MPI applications. iCheck aids these applications by dynamically reconfiguring checkpointing resources and offering robust checkpointing and data redistribution services. By leveraging Remote Direct Memory Access (RDMA) to support malleable applications, iCheck facilitates faster data transfers (up to 40 times improvement over the PFS-based solution), ensuring efficient performance. The system can dynamically adjust checkpointing processes based on metrics such as available memory, checkpoint frequency, and number of processes, maintaining or improving checkpoint performance amid resource changes. This adaptive approach supports fault tolerance as well as simplifies the development of malleable applications by effectively managing resource redistribution during resource changes.

The advent of generative design in the automotive sector, characterised by the automatic and iterative exploration of expansive solution spaces to discover optimal design configurations, has significantly increased the demand for computational resources to run intensive computer-aided engineering (CAE) simulations within constrained time frames. The inherent limitations of static high-performance computing (HPC) clusters have necessitated the adoption of cloud resources due to their flexible and elastic nature, thereby enhancing the capacity to accommodate the computational demands of these iterative workflows. These workflows, represented as Directed Acyclic Graphs (DAGs), involve the serial and parallel execution of tasks, which can dynamically share resources with other workflows during idle periods. In this paper, we propose an economy-based approach to exploit the gaps generated by these idle periods through a bidding system, thereby enabling more efficient resource utilisation and reducing the average wait time, makespan, cost and deadline miss by more than 40%, 6%, 13% and 45%respectively against certain infrastructures and baselines. Furthermore, we explore the potential for generating revenue by renting out idle resources in a hybrid cloud setup. This approach not only aims to optimise the use of computational resources but also seeks to provide cost-effective solutions to meet the escalating demands of generative design in the automotive sector.

Federated Learning (FL) is an emerging machine learning paradigm that enables the collaborative training of a shared global model across distributed clients while keeping the data decentralized. Recent works on designing systems for efficient FL have shown that utilizing serverless computing technologies, particularly Function-as-a-Service (FaaS) for FL, can enhance resource efficiency, reduce training costs, and alleviate the complex infrastructure management burden on data holders. However, current serverless FL systems still suffer from the presence of stragglers, i.e., slow clients that impede the collaborative training process. While strategies aimed at mitigating stragglers in these systems have been proposed, they overlook the diverse hardware resource configurations among FL clients. To this end, we present Apodotiko, a novel asynchronous training strategy designed for serverless FL. Our strategy incorporates a scoring mechanism that evaluates each client's hardware capacity and dataset size to intelligently prioritize and select clients for each training round, thereby minimizing the effects of stragglers on system performance. We comprehensively evaluate Apodotiko across diverse datasets, considering a mix of CPU and GPU clients, and compare its performance against five other FL training strategies. Results from our experiments demonstrate that Apodotiko outperforms other FL training strategies, achieving an average speedup of 2.75x and a maximum speedup of 7.03x. Furthermore, our strategy significantly reduces cold starts by a factor of four on average, demonstrating suitability in serverless environments.

Tackling climate change by reducing and eventually eliminating carbon emissions is a significant milestone on the path toward establishing an environmentally sustainable society. As we transition into the exascale era, marked by an increasing demand and scale of HPC resources, the HPC community must embrace the challenge of reducing carbon emissions from designing and operating modern HPC systems. In this position paper, we describe challenges and highlight different opportunities that can aid HPC sites in reducing the carbon footprint of modern HPC systems.

Emotions not only remained as an instrument of choice to unite humans but also to establish human-animal bonds. Identifying emotions from two distinct binaries among the livelihood animals, the home and wild livelihood animals, mandated specialized knowledge or skilled mahouts in the past. This article proposed an Elephant Emotion and Distraction Detection (E2D2) framework using YOLOv5-based deep learning algorithms to detect the emotions of elephants such as fear, anger, joy, and neutrality, in two different approaches - Elephant Emotion Detection (EED) approach and Elephant Emotional Feature Detection (EEFD) approach. E2D2 captures the distracting moments of elephants while working on gargantuan tasks such as carrying wooden logs. The article manifests the importance of the proposed E2D2 framework when experimenting on edge-enabled heterogeneous devices, including Jetson Nano. Our approach achieved an accuracy of 83.28% to 88.89% when experimented with elephant videos in real time. The framework can benefit people while approaching elephants; it can prevent tourists from elephant attacks; and, it can enhance the working abilities of trained elephant-based logistics by establishing strong social bonds with mahouts.

Accurate, timely Root Cause Analysis (RCA) is essential to successful IT operations as a primary step to incident remediation. RCA automation using data mining techniques in large heterogeneous systems is, however, a challenging task, because it requires correlating multimodal information across various data sources. An increasing number of services are migrating to structured logging to enable automated monitoring and debugging of complex large-scale systems. In this paper, we leverage structured logs and association rule mining (ARM) to automate RCA. We propose the LogRule algorithm, which automatically analyzes structured logs to generate a list of explanations for an event of interest. It achieves 0.921 F1-score for the diagnosis task, while computing results 37x faster compared to the state-of-the-art solution based on FP-growth, making it a time-efficient, accurate, and interpretable ARM-based RCA algorithm. Evaluation results show that LogRule enables RCA in complex multidimensional datasets, where the execution time of the current state-of-the-art algorithm is prohibitively large.