2023 IEEE Cloud Summit最新文献_第2页

Some New Observations on SLO-aware Edge Stream Processing 关于慢速感知边缘流处理的一些新观察

2023 IEEE Cloud Summit Pub Date : 2023-07-01 DOI: 10.1109/CloudSummit57601.2023.00011

Amna Shahid, Peng Kang, P. Lama, S. Khan

引用次数: 0

Multi-Tenant Deep Learning Acceleration with Competitive GPU Resource Sharing

2023 IEEE Cloud Summit Pub Date : 2023-07-01 DOI: 10.1109/CloudSummit57601.2023.00014

Yongbo Yu, Xiang Chen

{"title":"Multi-Tenant Deep Learning Acceleration with Competitive GPU Resource Sharing","authors":"Yongbo Yu, Xiang Chen","doi":"10.1109/CloudSummit57601.2023.00014","DOIUrl":"https://doi.org/10.1109/CloudSummit57601.2023.00014","url":null,"abstract":"As Deep Learning (DL) continues to drive a variety of applications in edge and cloud data centers, co-locating multiple DL models onto the same GPU become widely deployed to improve resource utilization, and achieve acceleration. For example, a self-driving system hosts multiple tasks simultaneously (e.g., detection, classification, segmentation, etc.) and expects concurrent computing on one single device. However, our analysis demonstrates that, when deploying compound DNN models for multiple tenants on a GPU, certain issues arise: As different models’ structure heterogeneities and skewed data distributions, corresponding models cause highly imbalanced computing workloads. However, current GPU scheduling methods lack effective resource allocations. To address these issues, we propose a novel resource allocation method – competitive resource sharing, which is beneficial for parallel model executions, and the proposed concept of “virtual resource” could effectively characterize and guide the practical per-task resource utilization and allocation. Our experiments demonstrate that the DNN computing throughput could be significantly escalated by $2.16 times sim 2.80 times$ in various multitenant scenarios.","PeriodicalId":185645,"journal":{"name":"2023 IEEE Cloud Summit","volume":"8 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117336936","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}

引用次数: 0

Geofencing Implement For Self-Monitoring Wandering Behavior And Sharing Location In Realtime With Firebase 用于自我监控漫游行为和与Firebase实时共享位置的地理围栏实现

2023 IEEE Cloud Summit Pub Date : 2023-07-01 DOI: 10.1109/CloudSummit57601.2023.00020

Dennis Flynn, H. Kimm

{"title":"Geofencing Implement For Self-Monitoring Wandering Behavior And Sharing Location In Realtime With Firebase","authors":"Dennis Flynn, H. Kimm","doi":"10.1109/CloudSummit57601.2023.00020","DOIUrl":"https://doi.org/10.1109/CloudSummit57601.2023.00020","url":null,"abstract":"Geofencing is an advanced technique that collects geolocation data in order to test whether a path of travel has crossed into or out of a geographical area represented by a geometric shape. The Firebase Backend-as-a-Service (Baas) supports Android applications with data store and realtime share capabilities. These techniques are applied in location based services and areas of human activity recognition research, especially for problems in medicine like dementia/Alzheimer patients, HIV transmission prevention, and monitoring COVID-19 patients’ movements. In this paper, we present an Android application with a two-stage geofencing trigger used to initiate communication with the Firebase BaaS and trigger features that address wandering behavior in geriatric healthcare patients. The proposed tool uses the Android Location API, the Google Maps SDK and the Firebase BaaS. In order to integrate geofencing into our existing Firebase feature platform, this paper investigates use of the Android Location Library’s Geofence. Our toolchain and methods for analyzing geofence location data to address wandering behavior were preferred after review of related research and comparison with existing commercial solutions. Lastly, our geofence integration tool is developed, implemented and field-tested in order to calibrate user options for the primary geofence parameters. We also provide analysis and discussion of our geofencing optimizations in the context of geriatric healthcare.","PeriodicalId":185645,"journal":{"name":"2023 IEEE Cloud Summit","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121648725","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}

引用次数: 0

Providing Reliable Services for Hardware Cryptography Accelerator in Virtualization 为虚拟化中的硬件加密加速器提供可靠的服务

2023 IEEE Cloud Summit Pub Date : 2023-07-01 DOI: 10.1109/CloudSummit57601.2023.00018

Chen Yuli, Songwu Shen, Wang Yuan, Zhu Tong, Mannepalli Veerendranath

引用次数: 0

Distributed Scalable Edge Computing Infrastructure for Open Metaverse 面向开放元宇宙的分布式可扩展边缘计算基础设施

2023 IEEE Cloud Summit Pub Date : 2023-07-01 DOI: 10.1109/CloudSummit57601.2023.00007

Larry Zhou, Jordan Lambert, Yanyan Zheng, Zheng Li, Alan Yen, Sandra Liu, Vivian Ye, Maggie Zhou, D. Mahar, J. Gibbons, M. Satterlee

{"title":"Distributed Scalable Edge Computing Infrastructure for Open Metaverse","authors":"Larry Zhou, Jordan Lambert, Yanyan Zheng, Zheng Li, Alan Yen, Sandra Liu, Vivian Ye, Maggie Zhou, D. Mahar, J. Gibbons, M. Satterlee","doi":"10.1109/CloudSummit57601.2023.00007","DOIUrl":"https://doi.org/10.1109/CloudSummit57601.2023.00007","url":null,"abstract":"In order to create and distribute the immense computing power required for the Metaverse, the industry must make strategic investments. AT&T believes that networking and edge computing are among the most essential components of a successful Metaverse ecosystem. AT&T proposes an “end-toend distributed edge computing network infrastructure” using its innovative uCPE [1] Whitebox platform. This innovation includes GPU virtualization, low-latency graphic rendering at the edge, and real-time streaming of 3D immersive content to end users. The groundbreaking design leverages existing wireless/wired home networks to create an overlay network that offloads heavy computing from head-mounted displays (HMDs) to the edge uCPE and distributes the virtual computing power to every end user with high scalability and low costs. In addition, AT&T introduces hardware and software probes to proactively monitor the end-to-end network performance with fault detection and self-healing capabilities, optimizing network paths and providing an uninterrupted immersive experience to users.","PeriodicalId":185645,"journal":{"name":"2023 IEEE Cloud Summit","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129609852","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}

引用次数: 0

Quantum-based Distributed Algorithms for Edge Node Placement and Workload Allocation 基于量子的分布式边缘节点放置和工作负载分配算法

2023 IEEE Cloud Summit Pub Date : 2023-06-01 DOI: 10.1109/CloudSummit57601.2023.00015

Duong The Do, Ni Trieu, D. Nguyen

引用次数: 0