The EDGELESS project is set to efficiently operate serverless computing in extremely diverse computing environments, from resource-constrained edge devices to highly-virtualized cloud platforms. Automatic deployment and reconfiguration will leverage AI/ML techniques, resulting in a flexible horizontally-scalable computation solution able to fully use heterogeneous edge resources while preserving vertical integration with the cloud and the benefits of serverless and its companion programming model, i.e., Function-as-a-Service (FaaS). The system under design will be environmentally sustainable, as it will dynamically concentrate resources physically (e.g., by temporarily switching off far-edge devices) or logically (e.g., by dispatching tasks towards a specific set of nodes) at the expense of performance-tolerant applications.
{"title":"EDGELESS Project: On the Road to Serverless Edge AI","authors":"C. Cicconetti, E. Carlini, Antonio Paradell","doi":"10.1145/3589010.3594890","DOIUrl":"https://doi.org/10.1145/3589010.3594890","url":null,"abstract":"The EDGELESS project is set to efficiently operate serverless computing in extremely diverse computing environments, from resource-constrained edge devices to highly-virtualized cloud platforms. Automatic deployment and reconfiguration will leverage AI/ML techniques, resulting in a flexible horizontally-scalable computation solution able to fully use heterogeneous edge resources while preserving vertical integration with the cloud and the benefits of serverless and its companion programming model, i.e., Function-as-a-Service (FaaS). The system under design will be environmentally sustainable, as it will dynamically concentrate resources physically (e.g., by temporarily switching off far-edge devices) or logically (e.g., by dispatching tasks towards a specific set of nodes) at the expense of performance-tolerant applications.","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129444029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ioannis Korontanis, Antonios Makris, T. Theodoropoulos, K. Tserpes
As Edge and Cloud platforms increasingly become the basis for hosting applications, monitoring systems need to be able to monitor both the hosts and the deployed applications. Ensuring seamless monitoring of Edge hosts is the most crucial factor, regardless of whether they are virtual machines, Raspberry Pis, or even typical personal computers. The application orchestration process could benefit greatly from monitoring systems, as they could serve as the main source of characterizing hosts and reporting metrics related to users and platforms. This paper introduces a Prometheus-based monitoring stack that is able to monitor both applications and resources on Kubernetes clusters. This monitoring stack has been tested and is already in use within the ACCORDION platform to efficiently monitor applications, regardless of their development unit and host types.
{"title":"Real-time Monitoring and Analysis of Edge and Cloud Resources","authors":"Ioannis Korontanis, Antonios Makris, T. Theodoropoulos, K. Tserpes","doi":"10.1145/3589010.3594892","DOIUrl":"https://doi.org/10.1145/3589010.3594892","url":null,"abstract":"As Edge and Cloud platforms increasingly become the basis for hosting applications, monitoring systems need to be able to monitor both the hosts and the deployed applications. Ensuring seamless monitoring of Edge hosts is the most crucial factor, regardless of whether they are virtual machines, Raspberry Pis, or even typical personal computers. The application orchestration process could benefit greatly from monitoring systems, as they could serve as the main source of characterizing hosts and reporting metrics related to users and platforms. This paper introduces a Prometheus-based monitoring stack that is able to monitor both applications and resources on Kubernetes clusters. This monitoring stack has been tested and is already in use within the ACCORDION platform to efficiently monitor applications, regardless of their development unit and host types.","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125244500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The compute continuum is considered to be just an aggregation of computational entities encompassing a diverse array of resources, ranging from public clouds to edge devices. To facilitate the development of next-generation intelligent applications, we must transcend this restrictive perspective. In this keynote, we present the notion of an "intelligent meta-continuum" - a purpose-driven, adaptable continuum capable of effortlessly amalgamating distributed resources to execute applications in an optimized and scalable manner. The meta-continuum integrates pivotal technological and architectural advances, such as software-defined infrastructure, workload orchestration, microservices, virtualization, and edge computing. Harnessing these enablers, the intelligent meta-continuum offers a flexible foundation for contemporary distributed applications, allowing them to dynamically adapt to highly variable conditions by optimizing their placement and resource utilization based on factors such as bandwidth, latency, cost, and availability. Innovative programming and organizational models, such as the actor model, play a crucial role in actualizing this vision. These models enable the decomposition of applications into discrete, stateful components that intelligently interact across the meta-continuum. The implementation of such models through technologies like containers and serverless computing allows applications to glide effortlessly across infrastructures. The meta-continuum holds the promise of substantial benefits for next-generation applications in domains such as smart cities, industrial automation, healthcare, and autonomous systems. By seamlessly amalgamating resources from cloud to edge, applications can achieve low latency, extensive distribution, mobility, scalability, and cost-efficiency. To fully unlock its potential, however, we must confront prevailing challenges related to interoperability, security, skill development, and organizational transformation.
{"title":"An Intelligent Continuum for Intelligent Applications","authors":"Patrizio Dazzi","doi":"10.1145/3589010.3594891","DOIUrl":"https://doi.org/10.1145/3589010.3594891","url":null,"abstract":"The compute continuum is considered to be just an aggregation of computational entities encompassing a diverse array of resources, ranging from public clouds to edge devices. To facilitate the development of next-generation intelligent applications, we must transcend this restrictive perspective. In this keynote, we present the notion of an \"intelligent meta-continuum\" - a purpose-driven, adaptable continuum capable of effortlessly amalgamating distributed resources to execute applications in an optimized and scalable manner. The meta-continuum integrates pivotal technological and architectural advances, such as software-defined infrastructure, workload orchestration, microservices, virtualization, and edge computing. Harnessing these enablers, the intelligent meta-continuum offers a flexible foundation for contemporary distributed applications, allowing them to dynamically adapt to highly variable conditions by optimizing their placement and resource utilization based on factors such as bandwidth, latency, cost, and availability. Innovative programming and organizational models, such as the actor model, play a crucial role in actualizing this vision. These models enable the decomposition of applications into discrete, stateful components that intelligently interact across the meta-continuum. The implementation of such models through technologies like containers and serverless computing allows applications to glide effortlessly across infrastructures. The meta-continuum holds the promise of substantial benefits for next-generation applications in domains such as smart cities, industrial automation, healthcare, and autonomous systems. By seamlessly amalgamating resources from cloud to edge, applications can achieve low latency, extensive distribution, mobility, scalability, and cost-efficiency. To fully unlock its potential, however, we must confront prevailing challenges related to interoperability, security, skill development, and organizational transformation.","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"49 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116319260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
E. Carlini, M. Coppola, Patrizio Dazzi, L. Ferrucci, Hanna Kavalionak, M. Mordacchini
This paper presents a dataset on edge devices and mobility patterns to comprehensively understand user behaviour and devices workload in Edge computing environments. The dataset is built on top of a publicly available dataset of cellular tower locations to simulate Edge devices, and on user mobility trajectories generated by a state-of-the-art simulator based on real location maps in the area of the city of Pisa, Italy. The resulting dataset reports the amount of vehicles in the range of about 200 Edge devices for each step of the simulation. The dataset can be used for various applications in edge computing and mobility, most notably for assessing results on resource and application management solutions at the edge in a realistic environment.
{"title":"DATA7: A Dataset for Assessing Resource and Application Management Solutions at the Edge","authors":"E. Carlini, M. Coppola, Patrizio Dazzi, L. Ferrucci, Hanna Kavalionak, M. Mordacchini","doi":"10.1145/3589010.3595652","DOIUrl":"https://doi.org/10.1145/3589010.3595652","url":null,"abstract":"This paper presents a dataset on edge devices and mobility patterns to comprehensively understand user behaviour and devices workload in Edge computing environments. The dataset is built on top of a publicly available dataset of cellular tower locations to simulate Edge devices, and on user mobility trajectories generated by a state-of-the-art simulator based on real location maps in the area of the city of Pisa, Italy. The resulting dataset reports the amount of vehicles in the range of about 200 Edge devices for each step of the simulation. The dataset can be used for various applications in edge computing and mobility, most notably for assessing results on resource and application management solutions at the edge in a realistic environment.","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123311065","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
E. Carlini, Patrizio Dazzi, K. Tserpes, Lorenzo Blasi, Marco Di Girolamo, Dariusz Dober
The seamless utilization of resources in the cloud-edge spectrum is a key driver for innovation in the ICT sector, as it supports economic growth and strengthens the industry's competitiveness while making next-application services possible with minimal investments and disruption. In this context, the EU project ACCORDION provides an innovative three-layered architecture designed as a comprehensive solution dedicated to latency-aware applications. This paper summarizes the key technological innovations of ACCORDION, highlighting their alignment with the European agenda of the ICT sector.
{"title":"Innovation Potential of the ACCORDION Platform","authors":"E. Carlini, Patrizio Dazzi, K. Tserpes, Lorenzo Blasi, Marco Di Girolamo, Dariusz Dober","doi":"10.1145/3589010.3594887","DOIUrl":"https://doi.org/10.1145/3589010.3594887","url":null,"abstract":"The seamless utilization of resources in the cloud-edge spectrum is a key driver for innovation in the ICT sector, as it supports economic growth and strengthens the industry's competitiveness while making next-application services possible with minimal investments and disruption. In this context, the EU project ACCORDION provides an innovative three-layered architecture designed as a comprehensive solution dedicated to latency-aware applications. This paper summarizes the key technological innovations of ACCORDION, highlighting their alignment with the European agenda of the ICT sector.","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114411330","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Bacciu, K. Tserpes, M. Coppola, Georg Macher, C. Gallicchio, O. Veledar, Anna Maria Anaxagorou, Patrizio Dazzi
TEACHING proposes a distributed, trustworthy AI integrating continuous human feedback, supporting CPSoS application design and deployment. TEACHING envisions an intelligent environment, empowering humans through cybernetic assistance. It advances autonomous safety-critical systems, improving safety, reliability and acceptability through human-centred design and formal validation crossing paradigms. TEACHING brings humans and AI together, enabling participatory development, optimisation and oversight.
{"title":"TEACHING: A Computing Toolkit for Building Efficient Autonomous appliCations Leveraging Humanistic INtelliGence","authors":"D. Bacciu, K. Tserpes, M. Coppola, Georg Macher, C. Gallicchio, O. Veledar, Anna Maria Anaxagorou, Patrizio Dazzi","doi":"10.1145/3589010.3594886","DOIUrl":"https://doi.org/10.1145/3589010.3594886","url":null,"abstract":"TEACHING proposes a distributed, trustworthy AI integrating continuous human feedback, supporting CPSoS application design and deployment. TEACHING envisions an intelligent environment, empowering humans through cybernetic assistance. It advances autonomous safety-critical systems, improving safety, reliability and acceptability through human-centred design and formal validation crossing paradigms. TEACHING brings humans and AI together, enabling participatory development, optimisation and oversight.","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122290151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alessio Matricardi, A. Bocci, Stefano Forti, Antonio Brogi
Deploying Function-as-a-Service (FaaS) applications to resources in the Cloud-Edge continuum calls for suitable simulation environments to assess new proposals for managing those applications while accounting for their specificity, i.e. ephemeral deployment and orchestrated execution. Indeed, exploiting real infrastructures for extensive experiments, while considering heterogeneous nodes and different applications, is costly and time-consuming. In this article, we propose a novel open-source discrete-time simulator for the management (i.e. deployment and migration) of orchestrated FaaS in the Cloud-Edge continuum. It enables assessing customised management policies against varying infrastructure conditions. It collects data about execution time, energy consumption, and successes/failures of management operations. We illustrate the prototype at work over a lifelike case study.
{"title":"Simulating FaaS Orchestrations In The Cloud-Edge Continuum","authors":"Alessio Matricardi, A. Bocci, Stefano Forti, Antonio Brogi","doi":"10.1145/3589010.3594893","DOIUrl":"https://doi.org/10.1145/3589010.3594893","url":null,"abstract":"Deploying Function-as-a-Service (FaaS) applications to resources in the Cloud-Edge continuum calls for suitable simulation environments to assess new proposals for managing those applications while accounting for their specificity, i.e. ephemeral deployment and orchestrated execution. Indeed, exploiting real infrastructures for extensive experiments, while considering heterogeneous nodes and different applications, is costly and time-consuming. In this article, we propose a novel open-source discrete-time simulator for the management (i.e. deployment and migration) of orchestrated FaaS in the Cloud-Edge continuum. It enables assessing customised management policies against varying infrastructure conditions. It collects data about execution time, energy consumption, and successes/failures of management operations. We illustrate the prototype at work over a lifelike case study.","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129545209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ismet Dagli, Andrew Depke, Andrew Mueller, M. S. Hassan, A. Akoglu, M. Belviranli
Diversely Heterogeneous System-on-Chips (DH-SoC) are increasingly popular computing platforms in many fields, such as autonomous driving and AR/VR applications, due to their ability to effectively balance performance and energy efficiency. Having multiple target accelerators for multiple concurrent workloads requires a careful runtime analysis of scheduling. In this study, we examine a scenario that mandates several concerns to be carefully addressed: 1) exploring the mapping of various workloads to heterogeneous accelerators to optimize the system for better performance, 2) analyzing data from the physical world in runtime to minimize the response time of the system 3) accurately estimating the resource contention by workloads during runtime since there will be con- current operations running under the same die, and 4) deferring the operation to the cloud for computationally more demanding operations such as continuous learning or real-time rendering, de- pending on the complexity of the computation. We demonstrate our analysis and approach on a VR project as a case study by using NVIDIA Xavier NX Edge DH-SoC and a server equipped with NVIDIA GeForce RTX 3080 GPU and AMD EPYC 7402 CPU.
{"title":"Contention-aware Performance Modeling for Heterogeneous Edge and Cloud Systems","authors":"Ismet Dagli, Andrew Depke, Andrew Mueller, M. S. Hassan, A. Akoglu, M. Belviranli","doi":"10.1145/3589010.3594889","DOIUrl":"https://doi.org/10.1145/3589010.3594889","url":null,"abstract":"Diversely Heterogeneous System-on-Chips (DH-SoC) are increasingly popular computing platforms in many fields, such as autonomous driving and AR/VR applications, due to their ability to effectively balance performance and energy efficiency. Having multiple target accelerators for multiple concurrent workloads requires a careful runtime analysis of scheduling. In this study, we examine a scenario that mandates several concerns to be carefully addressed: 1) exploring the mapping of various workloads to heterogeneous accelerators to optimize the system for better performance, 2) analyzing data from the physical world in runtime to minimize the response time of the system 3) accurately estimating the resource contention by workloads during runtime since there will be con- current operations running under the same die, and 4) deferring the operation to the cloud for computationally more demanding operations such as continuous learning or real-time rendering, de- pending on the complexity of the computation. We demonstrate our analysis and approach on a VR project as a case study by using NVIDIA Xavier NX Edge DH-SoC and a server equipped with NVIDIA GeForce RTX 3080 GPU and AMD EPYC 7402 CPU.","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121853994","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","authors":"","doi":"10.1145/3589010","DOIUrl":"https://doi.org/10.1145/3589010","url":null,"abstract":"","PeriodicalId":325857,"journal":{"name":"Proceedings of the 3rd Workshop on Flexible Resource and Application Management on the Edge","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128035022","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：