{"title":"Session details: Topics in Edge Computing","authors":"N. Karamchandani","doi":"10.1145/3284815","DOIUrl":"https://doi.org/10.1145/3284815","url":null,"abstract":"","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"311 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115689650","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}
N. Makris, Virgilios Passas, T. Korakis, L. Tassiulas
5G network access is expected to deliver high performance with low-latency network connections for the end-users, suitable for a plethora of different applications, as well as add up to the network flexibility and manageability from the operator's perspective. In order to achieve low-latency, Multiple-access Edge Computing (MEC) is considered, whereas for achieving flexibility, the disaggregation of the base station elements and moving parts of their functionality to the Cloud is proposed. In this paper, we consider the case of disaggregated base stations based on the CU-DU paradigm, able to provision MEC functions in a per-packet and per-client basis, over real networks. We evaluate the placement of the MEC functions over the fronthaul interface or collocating them with the Core Network. We employ the OpenAirInterface platform and evaluate our MEC solution with dynamically adaptive video streams. Our results show significant gains for the service-to-UE path latency, complying with the requirements set for the 5G MEC operation.
{"title":"Employing MEC in the Cloud-RAN: An Experimental Analysis","authors":"N. Makris, Virgilios Passas, T. Korakis, L. Tassiulas","doi":"10.1145/3266276.3266281","DOIUrl":"https://doi.org/10.1145/3266276.3266281","url":null,"abstract":"5G network access is expected to deliver high performance with low-latency network connections for the end-users, suitable for a plethora of different applications, as well as add up to the network flexibility and manageability from the operator's perspective. In order to achieve low-latency, Multiple-access Edge Computing (MEC) is considered, whereas for achieving flexibility, the disaggregation of the base station elements and moving parts of their functionality to the Cloud is proposed. In this paper, we consider the case of disaggregated base stations based on the CU-DU paradigm, able to provision MEC functions in a per-packet and per-client basis, over real networks. We evaluate the placement of the MEC functions over the fronthaul interface or collocating them with the Core Network. We employ the OpenAirInterface platform and evaluate our MEC solution with dynamically adaptive video streams. Our results show significant gains for the service-to-UE path latency, complying with the requirements set for the 5G MEC operation.","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"213 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133986536","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}
We consider a content delivery system consisting of a central server and multiple end-users. The central server stores the entire catalog of contents on offer and can deliver the requested content to the end-users. In addition, the end-users are equipped with limited caching capabilities and have the ability to deliver content to each other via D2D communication. The system also allows a third mode of content delivery where the central server delivers content to some of the end-users who then relay it to the other users. Our goal is to determine which contents to cache at the end-users in order to minimize the cost of service. We characterize the optimal caching policy and evaluate the benefits of allowing the central server to use other end-users as relays to deliver content. The key takeaway from this work is that if end-users have caching capabilities, the benefits of the central server using end-users as relays is negligible. This is in contrast to the case where the end-users cannot cache content where using end-users as relays leads to significant improvement in system performance.
{"title":"Caching Policies for D2D-Assisted Content Delivery Systems","authors":"A. Mete, Sharayu Moharir","doi":"10.1145/3266276.3266278","DOIUrl":"https://doi.org/10.1145/3266276.3266278","url":null,"abstract":"We consider a content delivery system consisting of a central server and multiple end-users. The central server stores the entire catalog of contents on offer and can deliver the requested content to the end-users. In addition, the end-users are equipped with limited caching capabilities and have the ability to deliver content to each other via D2D communication. The system also allows a third mode of content delivery where the central server delivers content to some of the end-users who then relay it to the other users. Our goal is to determine which contents to cache at the end-users in order to minimize the cost of service. We characterize the optimal caching policy and evaluate the benefits of allowing the central server to use other end-users as relays to deliver content. The key takeaway from this work is that if end-users have caching capabilities, the benefits of the central server using end-users as relays is negligible. This is in contrast to the case where the end-users cannot cache content where using end-users as relays leads to significant improvement in system performance.","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130098454","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 2018 on Technologies for the Wireless Edge Workshop","authors":"","doi":"10.1145/3266276","DOIUrl":"https://doi.org/10.1145/3266276","url":null,"abstract":"","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124177988","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":"Session details: Caching Networks","authors":"Manjunath D.","doi":"10.1145/3284813","DOIUrl":"https://doi.org/10.1145/3284813","url":null,"abstract":"","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123759831","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}
Ranking has wide range of applications like social choiceciteSC, recommendation systemsciteRS, web searchciteWS, crowd sourcing citeCS etc. textttTeraSort is a distributed algorithm, commonly used in systems like Hadoop MapReduce, for sorting large datasets. However, in most applications of interest we do not desire complete ordering of data, rather only a few items which have the highest ranks. In this paper we propose Coded Partial Sort to obtain partially sorted data from large datasets using distributed computing systems. We intend to find texttttopK ordered elements of a dataset by optimally utilizing servers in distributed network. Coded Partial Sort modifies conventional textttTeraSort algorithm to remove data irrelevant for partial ordering and applies ideas of "coding" to improve run-time performance by significantly decreasing communication load of Uncoded Partial SortciteUs. We empirically evaluate the performance of tCoded and Uncoded Partial Sort on Amazon EC2 clusters for experimental settings of interest.
{"title":"TopK Ordering on Distributed Systems","authors":"Prarthana, N. Karamchandani","doi":"10.1145/3266276.3266280","DOIUrl":"https://doi.org/10.1145/3266276.3266280","url":null,"abstract":"Ranking has wide range of applications like social choiceciteSC, recommendation systemsciteRS, web searchciteWS, crowd sourcing citeCS etc. textttTeraSort is a distributed algorithm, commonly used in systems like Hadoop MapReduce, for sorting large datasets. However, in most applications of interest we do not desire complete ordering of data, rather only a few items which have the highest ranks. In this paper we propose Coded Partial Sort to obtain partially sorted data from large datasets using distributed computing systems. We intend to find texttttopK ordered elements of a dataset by optimally utilizing servers in distributed network. Coded Partial Sort modifies conventional textttTeraSort algorithm to remove data irrelevant for partial ordering and applies ideas of \"coding\" to improve run-time performance by significantly decreasing communication load of Uncoded Partial SortciteUs. We empirically evaluate the performance of tCoded and Uncoded Partial Sort on Amazon EC2 clusters for experimental settings of interest.","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129321170","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}
Aravindh Raman, Nishanth R. Sastry, N. Mokari, Mostafa Salehi, Tooba Faisal, Andrew Secker, Jigna Chandaria
The exponential growth in online content consumption is a key concern for designing future generation network architectures. In this paper, we use content access patterns from a large trace of content accesses comprising about half the population of United Kingdom to make the case that a large portion of the backhaul load can be mitigated by content sharing amongst edge devices. We explore various models for edge devices to store and share content amongst each other, ranging from reactive opportunistic sharing to predicting future content access and speculatively placing content on strategic devices prior to request. We analyse the performance of each of these models in terms of content placement and traffic savings, which are constrained by the storage available on edge devices, the performance of the speculation engine and the wireless channel conditions. We formulate and solve at scale an optimisation problem for strategically placing content for sharing within a geographically localised cell to show such an approach can save up to 47% of the traffic generated from a small cell.
{"title":"Care to Share?: An Empirical Analysis of Capacity Enhancement by Sharing at the Edge","authors":"Aravindh Raman, Nishanth R. Sastry, N. Mokari, Mostafa Salehi, Tooba Faisal, Andrew Secker, Jigna Chandaria","doi":"10.1145/3266276.3266279","DOIUrl":"https://doi.org/10.1145/3266276.3266279","url":null,"abstract":"The exponential growth in online content consumption is a key concern for designing future generation network architectures. In this paper, we use content access patterns from a large trace of content accesses comprising about half the population of United Kingdom to make the case that a large portion of the backhaul load can be mitigated by content sharing amongst edge devices. We explore various models for edge devices to store and share content amongst each other, ranging from reactive opportunistic sharing to predicting future content access and speculatively placing content on strategic devices prior to request. We analyse the performance of each of these models in terms of content placement and traffic savings, which are constrained by the storage available on edge devices, the performance of the speculation engine and the wireless channel conditions. We formulate and solve at scale an optimisation problem for strategically placing content for sharing within a geographically localised cell to show such an approach can save up to 47% of the traffic generated from a small cell.","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114618512","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":"Session details: Mobile Edge Computing","authors":"L. Tassiulas","doi":"10.1145/3284814","DOIUrl":"https://doi.org/10.1145/3284814","url":null,"abstract":"","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123421539","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 Platforms for Advanced Wireless Research (PAWR) program [1] aims to enable experimental wireless communications research across devices, communication techniques, networks, systems, and services conceived by the US academic and industrial wireless research community and deployed in partnership with local communities. PAWR seeks to accelerate the wireless innovation ecosystem, thereby enhancing broadband connectivity; enabling the emerging Internet of Things (IoT), edge computing and heterogeneous wireless connectivity technologies. Each research platform conceived under the PAWR program will enable at-scale experimentation by supporting the geographic size, technical diversity, and user density representative of a small city/community. From chipmakers to networking companies to software companies to application developers to vertical technology providers and users, the industry is devoting significant efforts to "moving past science experiments" into developing use cases for edge computing technologies. This calls for fundamental rethinking of computing and networking architectures that can disrupt existing business models and reshape industry landscapes. This talk details the edge computing ecosystem developed by the first two platforms; COSMOS [2] and POWDER [3]. We present the system architecture and components from radio clients, transport X-Haul, near edge cloud, and core cloud to rapidly develop and test Use-cases such as IoT Security via Edge AI, Smart City and Machine Vision, AR/VR and Automotive Edge (safety, navigation, automation + infotainment) on PAWR Platforms.
{"title":"Platforms for Advanced Wireless Research: Helping Define a New Edge Computing Paradigm","authors":"A. Gosain","doi":"10.1145/3266276.3266283","DOIUrl":"https://doi.org/10.1145/3266276.3266283","url":null,"abstract":"The Platforms for Advanced Wireless Research (PAWR) program [1] aims to enable experimental wireless communications research across devices, communication techniques, networks, systems, and services conceived by the US academic and industrial wireless research community and deployed in partnership with local communities. PAWR seeks to accelerate the wireless innovation ecosystem, thereby enhancing broadband connectivity; enabling the emerging Internet of Things (IoT), edge computing and heterogeneous wireless connectivity technologies. Each research platform conceived under the PAWR program will enable at-scale experimentation by supporting the geographic size, technical diversity, and user density representative of a small city/community. From chipmakers to networking companies to software companies to application developers to vertical technology providers and users, the industry is devoting significant efforts to \"moving past science experiments\" into developing use cases for edge computing technologies. This calls for fundamental rethinking of computing and networking architectures that can disrupt existing business models and reshape industry landscapes. This talk details the edge computing ecosystem developed by the first two platforms; COSMOS [2] and POWDER [3]. We present the system architecture and components from radio clients, transport X-Haul, near edge cloud, and core cloud to rapidly develop and test Use-cases such as IoT Security via Edge AI, Smart City and Machine Vision, AR/VR and Automotive Edge (safety, navigation, automation + infotainment) on PAWR Platforms.","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125671450","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}
Faheem Zafari, Jian Li, K. Leung, D. Towsley, A. Swami
Mobile edge computing seeks to provide resources to different delay-sensitive applications. However, allocating the limited edge resources to a number of applications is a challenging problem. To alleviate the resource scarcity problem, we propose sharing of resources among multiple edge computing service providers where each service provider has a particular utility to optimize. We model the resource allocation and sharing problem as a multi-objective optimization problem and present a Cooperative Game Theory (CGT) based framework, where each edge service provider first satisfies its native applications and then shares its remaining resources (if available) with users of other providers. Furthermore, we propose an ~O (N) algorithm that provides allocation decisions from the core, hence the obtained allocations are Pareto optimal and the grand coalition of all the service providers is stable. Experimental results show that our proposed resource allocation and sharing framework improves the utility of all the service providers compared with the case where the service providers are working alone (no resource sharing). Our ~O (N) algorithm reduces the time complexity of obtaining a solution from the core by as much as 71.67% when compared with the Shapley value.
{"title":"A Game-Theoretic Approach to Multi-Objective Resource Sharing and Allocation in Mobile Edge","authors":"Faheem Zafari, Jian Li, K. Leung, D. Towsley, A. Swami","doi":"10.1145/3266276.3266277","DOIUrl":"https://doi.org/10.1145/3266276.3266277","url":null,"abstract":"Mobile edge computing seeks to provide resources to different delay-sensitive applications. However, allocating the limited edge resources to a number of applications is a challenging problem. To alleviate the resource scarcity problem, we propose sharing of resources among multiple edge computing service providers where each service provider has a particular utility to optimize. We model the resource allocation and sharing problem as a multi-objective optimization problem and present a Cooperative Game Theory (CGT) based framework, where each edge service provider first satisfies its native applications and then shares its remaining resources (if available) with users of other providers. Furthermore, we propose an ~O (N) algorithm that provides allocation decisions from the core, hence the obtained allocations are Pareto optimal and the grand coalition of all the service providers is stable. Experimental results show that our proposed resource allocation and sharing framework improves the utility of all the service providers compared with the case where the service providers are working alone (no resource sharing). Our ~O (N) algorithm reduces the time complexity of obtaining a solution from the core by as much as 71.67% when compared with the Shapley value.","PeriodicalId":365026,"journal":{"name":"Proceedings of the 2018 on Technologies for the Wireless Edge Workshop","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121807298","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：