arXiv subjects

Mohsen Amini Salehi

Publications and source records attributed to Mohsen Amini Salehi.

At least 19 recordsLinked to original sources

arXiv2026

Audo-Sight: AI-driven Ambient Perception Across Edge-Cloud for Blind and Low Vision Users

Despite advances in assistive technologies, Blind and Low-Vision (BLV) individuals continue to face challenges in understanding their surroundings. Delivering concise, useful, and timely scene descriptions for ambient perception remains a long-standing problem in accessibility. Existing solutions often fail to identify user expectations for real-time and accessible responses. Moreover, for a given task, they either rely on cloud offloading, which imposes a significant delay, or edge-based AI, which often sacrifices accuracy. To address this, we present Audo-Sight, an AI-driven assistive system that spans across Edge-Cloud continuum and enables BLV individuals to perceive their surroundings through voice-based conversation. Audo-Sight provides low-latency, accurate, and human-friendly responses through a novel mechanism that seamlessly fuses Edge and Cloud responses. The system also addresses challenges in catering to BLV users through response editing informed by BLV needs. Audo-Sight orchestrates a set of AI models based on user query contextual analysis to infer intent and adjust for a variety of situations. In urgent cases where users require fast responses, Audo-Sight leverages parallel Edge and Cloud pipelines and seamlessly combines responses through its Response Fusion Engine. Systematic evaluation shows that Audo-Sight delivers speech output around 80% faster for urgent tasks and generates complete responses approximately 50% faster across all tasks compared to a commercial cloud-based solution---highlighting the need for customized AI-based solutions. Human evaluation of Audo-Sight shows that it is the preferred choice over GPT-5 for 62% of BLV participants with another 23% stating both perform comparably. Speed and interruption evaluations demonstrate that in most situations, the system can seamlessly respond at a rapid pace to keep up with BLV expectations.

cs.DC↗

arXiv2026

Empirical Analysis of Cloud-Edge Infrastructure Complexity: Practitioner Pain Points and Architectural Directions

The proliferation of cloud, edge, and Internet of Things (IoT) computing has created unprecedented opportunities for distributed applications. However, this architectural shift introduces profound infrastructural complexity, acting as a significant barrier to developer productivity and innovation. In this paper, we present an empirical analysis based on 101 semi-structured interviews across 86 organizations to investigate the state of cloud-native development practices, pain points, and expectations. Our findings quantitatively validate that deployment complexity (38.6%) and onboarding difficulty (35.6%) are the dominant operational bottlenecks, while developers heavily prioritize productivity (53.5%) and automation (44.6%) over raw performance optimization. Based on these empirical insights, we examine four architectural directions that address the validated pain points: unified object abstractions (Object-as-a-Service), platform engineering via Internal Developer Platforms, declarative AI/ML serving pipelines, and lightweight edge runtimes based on WebAssembly. Furthermore, we detail the multi-stakeholder ecosystem required for adopting novel infrastructure paradigms, emphasizing that security, operational integration, and strict multi-tenant isolation are prerequisites for production readiness. Our results demonstrate that the primary barrier to distributed computing adoption is not execution performance but infrastructural complexity, and that declaratively governed, higher-level abstractions across multiple paradigms offer viable architectural paths toward alleviating it.

cs.DC↗

arXiv2026

A Periodic Space of Distributed Computing: Vision & Framework

Advances in networking and computing technologies throughout the early decades of the 21st century have transformed long-standing dreams of pervasive communication and computation into reality. These technologies now form a rapidly evolving and increasingly complex global infrastructure that will underpin the next aspiration of computing: supporting intelligent systems with human-level or even superhuman capabilities. We examine how today's distributed computing landscape can evolve to meet the demands of future users, intelligent systems, and emerging application domains. We propose a "periodic framework" for characterizing the distributed computing landscape, inspired by the systematic structure and explanatory power of the "periodic table" in chemistry. This framework provides a structured way to describe, compare, and reason about the behaviors and design choices of different distributed computing solutions. Using this framework, we can identify patterns in key system properties, such as responsiveness and availability, across the distributed computing landscape. We also explain how the framework can help in predicting future trajectories in the field. Lastly, we synthesize insights from leading researchers worldwide regarding the desired properties, design principles, and implications of emerging areas in the forthcoming distributed computing landscape and in relation to the periodic framework. Together, these perspectives shed light on the considerations that will shape the distributed computing landscape underpinning future intelligent systems.

cs.DC↗

arXiv2026

Downsides of Smartness Across Edge-Cloud Continuum in Modern Industry

The fast pace of modern AI is rapidly transforming traditional industrial systems into vast, intelligent and potentially unmanned autonomous operational environments driven by AI-based solutions. These solutions leverage various forms of machine learning, reinforcement learning, and generative AI. The introduction of such smart capabilities has pushed the envelope in multiple industrial domains, enabling predictive maintenance, optimized performance, and streamlined workflows. These solutions are often deployed across the Industrial Internet of Things (IIoT) and supported by the Edge-Fog-Cloud computing continuum to enable urgent (i.e., real-time or near real-time) decision-making. Despite the current trend of aggressively adopting these smart industrial solutions to increase profit, quality, and efficiency, large-scale integration and deployment also bring serious hazards that if ignored can undermine the benefits of smart industries. These hazards include unforeseen interoperability side-effects and heightened vulnerability to cyber threats, particularly in environments operating with a plethora of heterogeneous IIoT systems. The goal of this study is to shed light on the potential consequences of industrial smartness, with a particular focus on security implications, including vulnerabilities, side effects, and cyber threats. We distinguish software-level downsides stemming from both traditional AI solutions and generative AI from those originating in the infrastructure layer, namely IIoT and the Edge-Cloud continuum. At each level, we investigate potential vulnerabilities, cyber threats, and unintended side effects. As industries continue to become smarter, understanding and addressing these downsides will be crucial to ensure secure and sustainable development of smart industrial systems.

cs.CR↗

arXiv2026

Benchmarking Message Brokers for IoT Edge Computing: A Comprehensive Performance Study

Asynchronous messaging is a cornerstone of modern distributed systems, enabling decoupled communication for scalable and resilient applications. Today's message queue (MQ) ecosystem spans a wide range of designs, from high-throughput streaming platforms to lightweight protocols tailored for edge and IoT environments. Despite this diversity, choosing an appropriate MQ system remains difficult. Existing evaluations largely focus on throughput and latency on fixed hardware, while overlooking CPU and memory footprint and the effects of resource constraints, factors that are critical for edge and IoT deployments. In this paper, we present a systematic performance study of eight prominent message brokers: Mosquitto, EMQX, HiveMQ, RabbitMQ, ActiveMQ Artemis, NATS Server, Redis (Pub/Sub), and Zenoh Router. We introduce mq-bench, a unified benchmarking framework to evaluate these systems under identical conditions, scaling up to 10,000 concurrent client pairs across three VM configurations representative of edge hardware. This study reveals several interesting and sometimes counter-intuitive insights. Lightweight native brokers achieve sub-millisecond latency, while feature-rich enterprise platforms incur 2-3X higher overhead. Under high connection loads, multi-threaded brokers like NATS and Zenoh scale efficiently, whereas the widely-deployed Mosquitto saturates earlier due to its single-threaded architecture. We also find that Java-based brokers consume significantly more memory than native implementations, which has important implications for memory-constrained edge deployments. Based on these findings, we provide practical deployment guidelines that map workload requirements and resource constraints to appropriate broker choices for telemetry, streaming analytics, and IoT use cases.

cs.DC↗

arXiv2026

Exploring Challenges in Developing Edge-Cloud-Native Applications Across Multiple Business Domains

As the convergence of cloud computing and advanced networking continues to reshape modern software development, edge-cloud-native paradigms have become essential for enabling scalable, resilient, and agile digital services that depend on high-performance, low-latency, and reliable communication. This study investigates the practical challenges of developing, deploying, and maintaining edge-cloud-native applications through in-depth interviews with professionals from diverse domains, including IT, finance, healthcare, education, and industry. Despite significant advancements in cloud technologies, practitioners, particularly those from non-technical backgrounds-continue to encounter substantial complexity stemming from fragmented toolchains, steep learning curves, and operational overhead of managing distributed networking and computing, ensuring consistent performance across hybrid environments, and navigating steep learning curves at the cloud-network boundary. Across sectors, participants consistently prioritized productivity, Quality of Service, and usability over conventional concerns such as cost or migration. These findings highlight the need for operationally simplified, SLA-aware, and developer-friendly platforms that streamline the full application lifecycle. This study contributes a practice-informed perspective to support the alignment of edge-cloud-native systems with the realities and needs of modern enterprises, offering critical insights for the advancement of seamless cloud-network convergence.

cs.DC↗

arXiv2026

EdgeWeaver: Accelerating IoT Application Development Across Edge-Cloud Continuum

The rise of complex, latency-sensitive IoT applications across the Edge-Cloud continuum exposes the limitations of current Function-as-a-Service (FaaS) platforms in seamlessly addressing the complexity, heterogeneity, and intermittent connectivity of Edge-Cloud environments. Developers are left to manage integration and Quality of Service (QoS) enforcement manually, rendering application development complicated and costly. To overcome these limitations, we introduce the EdgeWeaver platform that offers a unified "object" abstraction that is seamlessly distributed across the continuum to encapsulate application logic, state, and QoS. EdgeWeaver automates "class" deployment across edge and cloud by composing established distributed algorithms (e.g., Raft, CRDTs)-enabling developers to declaratively express QoS (e.g., availability and consistency) desires that, in turn, guide internal resource allocation, function placement, and runtime adaptation to fulfill them. We implement a prototype of EdgeWeaver and evaluate it under diverse settings and using human subjects. Results show that EdgeWeaver boosts development productivity by 31%, while declaratively enforcing strong consistency and achieving 9 nines availability, 10,000X higher than the current standard, with negligible performance impact.

cs.DC↗

arXiv2026

Foundation CAN LM: A Pretrained Language Model For Automotive CAN Data

The Controller Area Network (CAN) bus provides a rich source of vehicular signals increasingly leveraged for applications in automotive and auto insurance domains, including collision detection, predictive maintenance, and driver risk modeling. Despite this potential, existing pipelines largely train isolated task-specific models on raw CAN data, with only limited efforts exploring decoded signals. Such fragmentation prevents shared representation learning and limits cross-task generalization. By contrast, natural language processing (NLP) and computer vision (CV) have been transformed by the foundation model paradigm: large-scale pretraining followed by task-specific adaptation. In this work, we introduce the foundation CAN model that demonstrates multi-objective downstream generalization using a single pretrained backbone. Our approach treats CAN data as a language: we pretrain on large-scale, unlabeled decoded CAN signals and fine-tune across heterogeneous auto insurance tasks. To enable this, we propose a unified tokenization scheme for mixed discrete-continuous signals and address challenges of temporal complexity and trip-specific variability. Our results show that one pretrained CAN model can adapt effectively to diverse predictive tasks, validating that the foundation modeling paradigm, proven in NLP and CV, also holds for CAN data. This establishes a new direction for generalizable representation learning in automotive AI.

cs.AI↗

arXiv2025

Object as a Service: Simplifying Cloud-Native Development through Serverless Object Abstraction

The function-as-a-service (FaaS) paradigm is envisioned as the next generation of cloud computing systems that mitigate the burden for cloud-native application developers by abstracting them from cloud resource management. However, it does not deal with the application data aspects. As such, developers have to intervene and undergo the burden of managing the application data, often via separate cloud storage services. To further streamline cloud-native application development, in this work, we propose a new paradigm, known as Object as a Service (OaaS) that encapsulates application data and functions into the cloud object abstraction. OaaS relieves developers from resource and data management burden while offering built-in optimization features. Inspired by OOP, OaaS incorporates access modifiers and inheritance into the serverless paradigm that: (a) prevents developers from compromising the system via accidentally accessing underlying data; and (b) enables software reuse in cloud-native application development. Furthermore, OaaS natively supports dataflow semantics. It enables developers to define function workflows while transparently handling data navigation, synchronization, and parallelism issues. To establish the OaaS paradigm, we develop a platform named Oparaca that offers state abstraction for structured and unstructured data with consistency and fault-tolerant guarantees. We evaluated Oparaca under real-world settings against state-of-the-art platforms with respect to the imposed overhead, scalability, and ease of use. The results demonstrate that the object abstraction provided by OaaS can streamline flexible and scalable cloud-native application development with an insignificant overhead on the underlying serverless system.

cs.DC↗

arXiv2025

Action Engine: Automatic Workflow Generation in FaaS

Function as a Service (FaaS) is poised to become the foundation of the next generation of cloud systems due to its inherent advantages in scalability, cost-efficiency, and ease of use. However, challenges such as the need for specialized knowledge, platform dependence, and difficulty in scalability in building functional workflows persist for cloud-native application developers. To overcome these challenges and mitigate the burden of developing FaaS-based applications, in this paper, we propose a mechanism called Action Engine, that makes use of tool-augmented large language models (LLMs) at its kernel to interpret human language queries and automates FaaS workflow generation, thereby, reducing the need for specialized expertise and manual design. Action Engine includes modules to identify relevant functions from the FaaS repository and seamlessly manage the data dependency between them, ensuring the developer's query is processed and resolved. Beyond that, Action Engine can execute the generated workflow by injecting the user-provided arguments. On another front, this work addresses a gap in tool-augmented LLM research via adopting an Automatic FaaS Workflow Generation perspective to systematically evaluate methodologies across four fundamental sub-processes. Through benchmarking various parameters, this research provides critical insights into streamlining workflow automation for real-world applications, specifically in the FaaS continuum. Our evaluations demonstrate that the Action Engine achieves comparable performance to the few-shot learning approach while maintaining platform- and language-agnosticism, thereby, mitigating provider-specific dependencies in workflow generation. We notice that Action Engine can unlock FaaS workflow generation for non-cloud-savvy developers and expedite the development cycles of cloud-native applications.

cs.DC↗

arXiv2025

Benchmarking Different Application Types across Heterogeneous Cloud Compute Services

Infrastructure as a Service (IaaS) clouds have become the predominant underlying infrastructure for the operation of modern and smart technology. IaaS clouds have proven to be useful for multiple reasons such as reduced costs, increased speed and efficiency, and better reliability and scalability. Compute services offered by such clouds are heterogeneous -- they offer a set of architecturally diverse machines that fit efficiently executing different workloads. However, there has been little study to shed light on the performance of popular application types on these heterogeneous compute servers across different clouds. Such a study can help organizations to optimally (in terms of cost, latency, throughput, consumed energy, carbon footprint, etc.) employ cloud compute services. At HPCC lab, we have focused on such benchmarks in different research projects and, in this report, we curate those benchmarks in a single document to help other researchers in the community using them. Specifically, we introduce our benchmarks datasets for three application types in three different domains, namely: Deep Neural Networks (DNN) Inference for industrial applications, Machine Learning (ML) Inference for assistive technology applications, and video transcoding for multimedia use cases.

cs.DC↗

arXiv2024

HE2C: A Holistic Approach for Allocating Latency-Sensitive AI Tasks across Edge-Cloud

The high computational, memory, and energy demands of Deep Learning (DL) applications often exceed the capabilities of battery-powered edge devices, creating difficulties in meeting task deadlines and accuracy requirements. Unlike previous solutions that optimize a single metric (e.g., accuracy or energy efficiency), HE2C framework is designed to holistically address the latency, memory, accuracy, throughput, and energy demands of DL applications across edge-cloud continuum, thereby, delivering a more comprehensive and effective user experience. HE2C comprises three key modules: (a) a "feasibility-check module that evaluates the likelihood of meeting deadlines across both edge and cloud resources; (b) a "resource allocation strategy" that maximizes energy efficiency without sacrificing the inference accuracy; and (c) a "rescue module" that enhances throughput by leveraging approximate computing to trade accuracy for latency when necessary. Our primary objective is to maximize system prolong battery lifespan, throughput, and accuracy while adhering to strict latency constraints. Experimental evaluations in the context of wearable technologies for blind and visually impaired users demonstrate that HE2C significantly improves task throughput via completing a larger number of tasks within their specified deadlines, while preserving edge device battery and maintaining prediction accuracy with minimal latency impact. These results underscore HE2C's potential as a robust solution for resource management in latency-sensitive, energy-constrained edge-to-cloud environments.

cs.DC↗

arXiv2024

Streamlining Cloud-Native Application Development and Deployment with Robust Encapsulation

Current Serverless abstractions (e.g., FaaS) poorly support non-functional requirements (e.g., QoS and constraints), are provider-dependent, and are incompatible with other cloud abstractions (e.g., databases). As a result, application developers have to undergo numerous rounds of development and manual deployment refinements to finally achieve their desired quality and efficiency. In this paper, we present Object-as-a-Service (OaaS) -- a novel serverless paradigm that borrows the object-oriented programming concepts to encapsulate business logic, data, and non-functional requirements into a single deployment package, thereby streamlining provider-agnostic cloud-native application development. We also propose a declarative interface for the non-functional requirements of applications that relieves developers from daunting refinements to meet their desired QoS and deployment constraint targets. We realized the OaaS paradigm through a platform called Oparaca and evaluated it against various real-world applications and scenarios. The evaluation results demonstrate that Oparaca can enhance application performance by 60X and improve reliability by 50X through latency, throughput, and availability enforcement -- all with remarkably less development and deployment time and effort.

cs.DC↗

arXiv2024

A Multi-Level Approach for Class Imbalance Problem in Federated Learning for Remote Industry 4.0 Applications

Deep neural network (DNN) models are effective solutions for industry 4.0 applications (\eg oil spill detection, fire detection, anomaly detection). However, training a DNN network model needs a considerable amount of data collected from various sources and transferred to the central cloud server that can be expensive and sensitive to privacy. For instance, in the remote offshore oil field where network connectivity is vulnerable, a federated fog environment can be a potential computing platform. Hence it is feasible to perform computation within the federation. On the contrary, performing a DNN model training using fog systems poses a security issue that the federated learning (FL) technique can resolve. In this case, the new challenge is the class imbalance problem that can be inherited in local data sets and can degrade the performance of the global model. Therefore, FL training needs to be performed considering the class imbalance problem locally. In addition, an efficient technique to select the relevant worker model needs to be adopted at the global level to increase the robustness of the global model. Accordingly, we utilize one of the suitable loss functions addressing the class imbalance in workers at the local level. In addition, we employ a dynamic threshold mechanism with user-defined worker's weight to efficiently select workers for aggregation that improve the global model's robustness. Finally, we perform an extensive empirical evaluation to explore the benefits of our solution and find up to 3-5% performance improvement than baseline federated learning methods.

cs.DC↗

arXiv2024

Tutorial: Object as a Service (OaaS) Serverless Cloud Computing Paradigm

While the first generation of cloud computing systems mitigated the job of system administrators, the next generation of cloud computing systems is emerging to mitigate the burden for cloud developers -- facilitating the development of cloud-native applications. This paradigm shift is primarily happening by offering higher-level serverless abstractions, such as Function as a Service (FaaS). Although FaaS has successfully abstracted developers from the cloud resource management details, it falls short in abstracting the management of both data (i.e., state) and the non-functional aspects, such as Quality of Service (QoS) requirements. The lack of such abstractions implies developer intervention and is counterproductive to the objective of mitigating the burden of cloud-native application development. To further streamline cloud-native application development, we present Object-as-a-Service (OaaS) -- a serverless paradigm that borrows the object-oriented programming concepts to encapsulate application logic and data in addition to non-functional requirements into a single deployment package, thereby streamlining provider-agnostic cloud-native application development. We realized the OaaS paradigm through the development of an open-source platform called Oparaca. In this tutorial, we will present the concept and design of the OaaS paradigm and its implementation -- the Oparaca platform. Then, we give a tutorial on developing and deploying the application on the Oparaca platform and discuss its benefits and its optimal configurations to avoid potential overheads.

cs.DC↗

arXiv2024

FastMig: Leveraging FastFreeze to Establish Robust Service Liquidity in Cloud 2.0

Service liquidity across edge-to-cloud or multi-cloud will serve as the cornerstone of the next generation of cloud computing systems (Cloud 2.0). Provided that cloud-based services are predominantly containerized, an efficient and robust live container migration solution is required to accomplish service liquidity. In a nod to this growing requirement, in this research, we leverage FastFreeze, a popular platform for process checkpoint/restore within a container, and promote it to be a robust solution for end-to-end live migration of containerized services. In particular, we develop a new platform, called FastMig that proactively controls the checkpoint/restore operations of FastFreeze, thereby, allowing for robust live migration of containerized services via standard HTTP interfaces. The proposed platform introduces post-checkpointing and pre-restoration operations to enhance migration robustness. Notably, the pre-restoration operation includes containerized service startup options, enabling warm restoration and reducing the migration downtime. In addition, we develop a method to make FastFreeze robust against failures that commonly happen during the migration and even during the normal operation of a containerized service. Experimental results under real-world settings show that the migration downtime of a containerized service can be reduced by 30X compared to the situation where the original FastFreeze was deployed for the migration. Moreover, we demonstrate that FastMig and warm restoration method together can significantly mitigate the container startup overhead. Importantly, these improvements are achieved without any significant performance reduction and only incurs a small resource usage overhead, compared to the bare (\ie non-FastFreeze) containerized services.

cs.DC↗

arXiv2024

Resource Allocation of Industry 4.0 Micro-Service Applications across Serverless Fog Federation

The Industry 4.0 revolution has been made possible via AI-based applications (e.g., for automation and maintenance) deployed on the serverless edge (aka fog) computing platforms at the industrial sites -- where the data is generated. Nevertheless, fulfilling the fault-intolerant and real-time constraints of Industry 4.0 applications on resource-limited fog systems in remote industrial sites (e.g., offshore oil fields) that are uncertain, disaster-prone, and have no cloud access is challenging. It is this challenge that our research aims at addressing. We consider the inelastic nature of the fog systems, software architecture of the industrial applications (micro-service-based versus monolithic), and scarcity of human experts in remote sites. To enable cloud-like elasticity, our approach is to dynamically and seamlessly (i.e., without human intervention) federate nearby fog systems. Then, we develop serverless resource allocation solutions that are cognizant of the applications' software architecture, their latency requirements, and distributed nature of the underlying infrastructure. We propose methods to seamlessly and optimally partition micro-service-based application across the federated fog. Our experimental evaluation express that not only the elasticity is overcome in a serverless manner, but also our developed application partitioning method can serve around 20% more tasks on-time than the existing methods in the literature.

cs.DC↗

arXiv2023

HEET: A Heterogeneity Measure to Quantify the Difference across Distributed Computing Systems

Although system heterogeneity has been extensively studied in the past, there is yet to be a study on measuring the impact of heterogeneity on system performance. For this purpose, we propose a heterogeneity measure that can characterize the impact of the heterogeneity of a system on its performance behavior in terms of throughput or makespan. We develop a mathematical model to characterize a heterogeneous system in terms of its task and machine heterogeneity dimensions and then reduce it to a single value, called Homogeneous Equivalent Execution Time (HEET), which represents the execution time behavior of the entire system. We used AWS EC2 instances to implement a real-world machine learning inference system. Performance evaluation of the HEET score across different heterogeneous system configurations demonstrates that HEET can accurately characterize the performance behavior of these systems. In particular, the results show that our proposed method is capable of predicting the true makespan of heterogeneous systems without online evaluations with an average precision of 84%. This heterogeneity measure is instrumental for solution architects to configure their systems proactively to be sufficiently heterogeneous to meet their desired performance objectives.

cs.DC↗