Related references
Note: Only part of the references are listed.
Article
Computer Science, Theory & Methods
Zijun Li et al.
Summary: This article surveys and elaborates on the research domains in the serverless context, highlighting the key implications and limitations of works in each layer. It emphasizes the potential of the serverless computing model while pointing out the challenges for further research.
ACM COMPUTING SURVEYS
(2022)
Article
Computer Science, Theory & Methods
Hossein Shafiei et al.
Summary: The emerging serverless computing paradigm brings benefits such as less operational complexity, a pay-as-you-go pricing model, and an auto-scaling feature. It also opens up new opportunities and challenges for cloud application developers. The research areas related to serverless applications and the challenges and solutions in those areas are worth attention.
ACM COMPUTING SURVEYS
(2022)
Article
Computer Science, Software Engineering
David Bermbach et al.
Summary: The FaaS paradigm shows potential in fog computing environments, where a auction-inspired approach can be used to maximize revenue; simulations and experiments demonstrate that revenue for overloaded nodes can be optimized.
SOFTWARE-PRACTICE & EXPERIENCE
(2022)
Review
Computer Science, Theory & Methods
Gustavo Andre Setti Cassel et al.
Summary: Serverless computing, also known as Function as a Service (FaaS), is a research trend where applications are built and deployed as stateless functions. Originally proposed for the cloud, it is now being applied in Internet of Things (IoT) to reduce latency and energy consumption, integrating solutions across edge, fog, and cloud layers.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Computer Science, Information Systems
Dazhi Li et al.
Summary: As a coordination tool, workflow with a large number of interdependent tasks has become a new paradigm for orchestrating computationally intensive tasks. This paper proposes an elastic workflow scheduling framework called EWSF based on the serverless architecture, and introduces a serverless-oriented deadline-aware workflow scheduling algorithm called SoDa. Experimental results show that SoDa performs better in various scheduling environments, improving overall makespan and success rate.
WIRELESS COMMUNICATIONS & MOBILE COMPUTING
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Stefan Nastic et al.
Summary: Serverless computing is an emerging paradigm for developing cloud-native applications, offering advantages such as low management overhead, easy deployment, and optimized costs. It has expanded beyond traditional cloud environments to the Edge. This paper outlines the vision of the Serverless Computing Fabric (SCF) for the Edge-Cloud continuum, discussing opportunities, challenges, and design principles for achieving this vision.
2022 IEEE 4TH INTERNATIONAL CONFERENCE ON COGNITIVE MACHINE INTELLIGENCE, COGMI
(2022)
Proceedings Paper
Computer Science, Information Systems
Thomas Pusztai et al.
Summary: This paper presents Polaris Scheduler, an SLO-aware scheduler for Edge computing that optimizes the placement of microservices to improve the fulfillment of Service Level Objectives. The scheduler supports modeling of Edge topology and service dependencies, as well as network QoS requirements.
2022 IEEE/ACM 15TH INTERNATIONAL CONFERENCE ON UTILITY AND CLOUD COMPUTING, UCC
(2022)
Proceedings Paper
Computer Science, Information Systems
Philipp Raith et al.
Summary: This paper presents a mobility-aware framework for edge-cloud systems, which captures complex system behavior through the concept of pressure and combines the advantages of a centralized control-plane with a decentralized dataplane to handle distributed workloads and user mobility.
2022 IEEE/ACM 15TH INTERNATIONAL CONFERENCE ON UTILITY AND CLOUD COMPUTING, UCC
(2022)
Proceedings Paper
Computer Science, Information Systems
Joel Scheuner et al.
Summary: Serverless computing is a scalable event-based deployment paradigm. TriggerBench, a cross-provider benchmark based on distributed tracing, is presented to evaluate serverless function triggers. Results show that all triggers have long tail latency, and HTTP triggers are most suitable for interactive applications.
2022 IEEE INTERNATIONAL CONFERENCE ON CLOUD ENGINEERING (IC2E 2022)
(2022)
Proceedings Paper
Computer Science, Information Systems
Sebastian Werner et al.
Summary: This paper presents an initial design and implementation of HARDLESS, an extensible and generalized serverless computing architecture that supports workloads for arbitrary hardware accelerators. It demonstrates how HARDLESS can scale across different commodity hardware accelerators and support a variety of workloads using the same execution and programming model common in serverless computing today.
2022 IEEE INTERNATIONAL CONFERENCE ON CLOUD ENGINEERING (IC2E 2022)
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Yaodong Huang et al.
Summary: Serverless Computing and Function-as-a-Service offer convenient services to developers and users. This paper proposes a deviceless edge computing system targeting the mobility of end users, focusing on the migration of virtual functions to provide uninterrupted services. Extensive experiments and simulations demonstrate the system's ability to support high-mobility and delay-sensitive application scenarios.
2022 IEEE 42ND INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS (ICDCS 2022)
(2022)
Proceedings Paper
Computer Science, Theory & Methods
Tom Goethals et al.
Summary: The rise of edge computing and FaaS has revolutionized software virtualization, improving performance and security. This paper benchmarks lightweight virtualization platforms, such as microVMs and containers, in the context of edge microservices. The results show that standard Docker containers have reliable performance and low memory usage, microVM-based solutions provide better isolation, OSv unikernels have extremely low boot times and significantly better performance than Docker containers, and gVisor offers improved security and compatibility but lower performance compared to Docker.
2022 IEEE INTERNATIONAL CONFERENCE ON EDGE COMPUTING & COMMUNICATIONS (IEEE EDGE 2022)
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Xue Li et al.
Summary: This paper presents an adaptive auto-scaler, KneeScale, for serverless functions in edge computing. Experimental results demonstrate that KneeScale outperforms other competing techniques, achieving higher performance and resource efficiency.
2022 22ND IEEE/ACM INTERNATIONAL SYMPOSIUM ON CLUSTER, CLOUD AND INTERNET COMPUTING (CCGRID 2022)
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Philipp Gackstatter et al.
Summary: Serverless computing is a popular part of cloud computing, allowing developers to write auto-scalable functions in a polyglot environment. However, the cold-start latency and resource limitations in edge computing hinder the support for latency-critical IoT services. This research explores the use of WebAssembly as a lightweight alternative to container technologies for serverless computing, specifically in edge computing settings. The prototype implementation of WebAssembly in Apache OpenWhisk shows significant improvements in cold-start latency, memory consumption, and function execution throughput compared to the standard Docker-based container runtime.
2022 22ND IEEE/ACM INTERNATIONAL SYMPOSIUM ON CLUSTER, CLOUD AND INTERNET COMPUTING (CCGRID 2022)
(2022)
Proceedings Paper
Automation & Control Systems
Luciano Baresi et al.
Summary: This paper presents NEPTUNE, a serverless-based framework for managing complex MEC solutions. NEPTUNE reduces latency and resource consumption by deploying functions on edge nodes according to user locations, avoiding node saturation, utilizing GPUs, and dynamically allocating resources.
2022 17TH INTERNATIONAL SYMPOSIUM ON SOFTWARE ENGINEERING FOR ADAPTIVE AND SELF-MANAGING SYSTEMS (SEAMS)
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Christopher Peter Smith et al.
Summary: This study proposes a tool called FaDO that addresses the current flaw of serverless platforms not considering data placement when scheduling functions. FaDO allows for data-aware function scheduling and load balancing through load-balancing algorithms. It also provides an abstraction of the serverless compute cluster's storage and supports automatic and policy-aware data replication.
6TH IEEE INTERNATIONAL CONFERENCE ON FOG AND EDGE COMPUTING (ICFEC 2022)
(2022)
Proceedings Paper
Computer Science, Information Systems
Priscilla Benedetti et al.
Summary: Serverless computing is an alternative deployment paradigm for cloud computing platforms that aims to provide scalability and cost reduction. Open-source serverless computing platforms typically rely on workload-based or resource-based auto-scaling approaches. This study investigates the feasibility of using a reinforcement-based approach for resource-based auto-scaling in the widely used open-source serverless platform, OpenFaaS. The experimental analysis conducted on constrained Kubernetes-based nodes demonstrates that the proposed model successfully learns an effective scaling policy based on CPU utilization to minimize service latency within a limited number of iterations.
2022 IEEE INTERNATIONAL CONFERENCE ON PERVASIVE COMPUTING AND COMMUNICATIONS WORKSHOPS AND OTHER AFFILIATED EVENTS (PERCOM WORKSHOPS)
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Zijun Li et al.
Summary: Serverless computing provides fine-grain resource sharing, but the traditional workflow execution architecture performs poorly in this context. We propose a worker-side schedule pattern for serverless workflows to enable efficient execution. Experimental results show significant reductions in workflow scheduling and data transmission overheads.
ASPLOS '22: PROCEEDINGS OF THE 27TH ACM INTERNATIONAL CONFERENCE ON ARCHITECTURAL SUPPORT FOR PROGRAMMING LANGUAGES AND OPERATING SYSTEMS
(2022)
Article
Computer Science, Information Systems
Sebastian Burckhardt et al.
Summary: Serverless computing is a popular choice for cloud service architects because it offers scalability and load-based billing with minimal developer effort. New frameworks like Durable Functions allow for the addition of stateful abstractions in serverless applications. Netherite is a novel architecture that improves latency and throughput of serverless workflows by optimizing partitioning and state persistence.
PROCEEDINGS OF THE VLDB ENDOWMENT
(2022)
Article
Computer Science, Theory & Methods
Thomas Rausch et al.
Summary: This paper introduces a container scheduling system that enables serverless platforms to efficiently utilize edge infrastructures, and proposes an automatic fine-tuning method for scheduling constraints to optimize high-level operational objectives. The research results show that the scheduler significantly improves the quality of task placement compared to Kubernetes' latest scheduler, and the fine-tuning method for scheduling parameters is beneficial in meeting operational goals.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Review
Computer Science, Interdisciplinary Applications
Vojdan Kjorveziroski et al.
Summary: Serverless computing is a new concept that allows developers to focus on core functionalities without worrying about infrastructure. To address the challenges posed by the growing number of IoT devices, research is actively looking into moving serverless computing to the edge of the network for better bandwidth utilization, reduced latency, and compliance with privacy regulations.
Proceedings Paper
Computer Science, Theory & Methods
Francesc Lordan et al.
Summary: This paper proposes a solution to organize smart devices in the Cloud-Edge Continuum, by treating each device as an autonomous individual -Agent- that processes events/data and offers computing capacity to the infrastructure. The approach converts the logic of functions into task-based workflows, offloading workload to other agents to improve service performance.
EURO-PAR 2021: PARALLEL PROCESSING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Philipp Raith et al.
Summary: Edge Intelligence is a new umbrella term for applications driven by the Internet of Things and Edge Computing, aimed at improving data processing efficiency, achieving low latency, and protecting privacy. However, the development is still in its early stages, requiring solutions to main tasks encountered in developing AI applications and issues related to Edge Intelligence.
2021 IEEE INTERNATIONAL CONFERENCE ON SERVICES COMPUTING (SCC 2021)
(2021)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Thomas Pusztai et al.
Summary: Service Level Objectives (SLOs) define expected performance of cloud services, allowing service providers and consumers to have a clear understanding of guarantees and expectations. SLO Script is a language and framework designed for defining high-level SLOs, featuring type safety and decoupling of SLOs from elasticity strategies.
2021 IEEE INTERNATIONAL CONFERENCE ON WEB SERVICES, ICWS 2021
(2021)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Samuel H. Christie et al.
Summary: A decentralized application involves multiple autonomous principals, and requires a protocol to specify interactions and a programming model that enables each principal to independently construct agents.
2021 IEEE INTERNATIONAL CONFERENCE ON WEB SERVICES, ICWS 2021
(2021)
Proceedings Paper
Computer Science, Hardware & Architecture
Thomas Pusztai et al.
Summary: Service Level Objectives (SLOs) are critical in guiding the elasticity of cloud applications, which requires metrics for evaluation and a flexible middleware for implementation. The paper presents a middleware that provides an orchestrator-independent SLO controller and evaluates its performance through a motivating use case.
2021 IEEE 14TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING (CLOUD 2021)
(2021)
Proceedings Paper
Computer Science, Hardware & Architecture
Stefan Nastic et al.
Summary: In recent years, extensive research has been conducted on application workload scheduling in hybrid Cloud-Edge-IoT infrastructures, leading to the development of various solutions. However, current approaches are lacking in sensitivity to edge properties and Service Level Objective (SLO) awareness, highlighting the need for more advanced scheduling solutions. This paper introduces a novel scheduling framework, Polaris Scheduler, aimed at achieving edge-sensitive and SLO-aware scheduling in the Cloud-Edge-IoT Continuum.
2021 IEEE 14TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING (CLOUD 2021)
(2021)
Article
Green & Sustainable Science & Technology
Luciano Baresi et al.
Summary: Edge computing infrastructures are utilized for running low-latency applications. PAPS framework serves as a serverless platform for efficient management of large-scale Edge topologies, utilizing containers and automated resource allocation. The assessment demonstrates feasibility and efficiency in managing serverless functions and handling workload variations.
FRONTIERS IN SUSTAINABLE CITIES
(2021)
Article
Computer Science, Software Engineering
Sebastian Burckhardt et al.
Summary: Serverless, or Functions-as-a-Service (FaaS), is a popular paradigm for application development, but faces challenges when developing applications that require persistent state, reliable progress, or synchronization. A new generation of serverless frameworks, such as Azure's Durable Functions (DF), address these challenges by providing stateful abstractions and combining task and actor parallelism. This makes DF suitable for a wide range of serverless applications.
PROCEEDINGS OF THE ACM ON PROGRAMMING LANGUAGES-PACMPL
(2021)
Article
Computer Science, Theory & Methods
Diana M. Naranjo et al.
JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING
(2020)
Article
Computer Science, Information Systems
Shuiguang Deng et al.
IEEE INTERNET OF THINGS JOURNAL
(2020)
Proceedings Paper
Computer Science, Software Engineering
Phani Kishore Gadepalli et al.
PROCEEDINGS OF THE 2020 21ST INTERNATIONAL MIDDLEWARE CONFERENCE (MIDDLEWARE '20)
(2020)
Proceedings Paper
Computer Science, Information Systems
Anirban Das et al.
2020 20TH IEEE/ACM INTERNATIONAL SYMPOSIUM ON CLUSTER, CLOUD AND INTERNET COMPUTING (CCGRID 2020)
(2020)
Proceedings Paper
Computer Science, Theory & Methods
Vikram Sreekanti et al.
PROCEEDINGS OF THE FIFTEENTH EUROPEAN CONFERENCE ON COMPUTER SYSTEMS (EUROSYS'20)
(2020)
Proceedings Paper
Computer Science, Theory & Methods
Tobias Pfandzelter et al.
2020 IEEE INTERNATIONAL CONFERENCE ON FOG COMPUTING (ICFC 2020)
(2020)
Article
Computer Science, Information Systems
Vigor Prokhorenko et al.
Article
Computer Science, Information Systems
Carlos de Alfonso et al.
JOURNAL OF GRID COMPUTING
(2019)
Proceedings Paper
Computer Science, Information Systems
Pekka Karhula et al.
PROCEEDINGS OF THE 2ND ACM INTERNATIONAL WORKSHOP ON EDGE SYSTEMS, ANALYTICS AND NETWORKING (EDGESYS '19)
(2019)
Proceedings Paper
Computer Science, Hardware & Architecture
Alfonso Perez et al.
2019 IEEE 12TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING (IEEE CLOUD 2019)
(2019)
Article
Computer Science, Software Engineering
Abhinav Jangda et al.
PROCEEDINGS OF THE ACM ON PROGRAMMING LANGUAGES-PACMPL
(2019)