2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM)最新文献_第2页

Sequential Pluralization through Chromatic Operatives in Connected IoT Sensor's Topology 互联物联网传感器拓扑中通过色算子实现的顺序多元化

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/CCEM50674.2020.00020

Harish Bharti, R. Saxena, Sandeep Sukhija, Deepak Baiai

{"title":"Sequential Pluralization through Chromatic Operatives in Connected IoT Sensor's Topology","authors":"Harish Bharti, R. Saxena, Sandeep Sukhija, Deepak Baiai","doi":"10.1109/CCEM50674.2020.00020","DOIUrl":"https://doi.org/10.1109/CCEM50674.2020.00020","url":null,"abstract":"With the advent of the ubiquitous network, proliferation of sensors, wearables and other Internet of Things (IoT) devices are growing rapidly. All these devices generate a continual stream of data in different formats which is transmitted further over the network for processing. This ever-growing amount of data generated by different sensors need to be processed at a faster pace to maximize the benefits and take any corrective action if required. Although it is important to process data locally at the edge for rapid action, However, it is equally important to process data centrally to identify patterns across locations. Also, the gigantic time-series data captured by these IoT devices can't be stored and fully managed with the central system since it needs to be aggregated and analyzed within a short duration of time as produced by sensors. This paper provides a framework and mechanism that uses a chromatic value to represent sensor data for faster processing. Few bytes of data are passed by the framework with actual sensor data considering the standard Signal-to-Noise-interference for the receiver. These bytes are employed to represent pre-configured chromatic (‘coloring’ being used as a symbolic dimension) number that represents the different states/ranges of data associated which enables the integrator (logic/circuit) to parse this data quickly. This framework helps reduce the overhead cost and gain efficiency by mapping chromatic values with a range of values, which is a key contribution of this paper. This newly defined framework thus ensures optimal management of data at the edge of the device/sensor/network so that analytics on the central database can be performed in less duration.","PeriodicalId":160355,"journal":{"name":"2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM)","volume":"170 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126062677","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

Sponsors and Patrons 赞助商和赞助人

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/ccem50674.2020.00012

引用次数: 0

CCEM 2020: Message from the Steering Committee Chairs CCEM 2020:指导委员会主席的致辞

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/ccem50674.2020.00008

引用次数: 0

Machine Translation of English Language to Mizo Language 英语到米佐语的机器翻译

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/CCEM50674.2020.00028

Zaitinkhuma Thihlum, Vanlalmuansangi Khenglawt, Somen Debnath

引用次数: 7

Sub-Object Data Augmentation And Custom Loss Functions for Automated Anatomical Landmarks Localisation in 3D Brain MR Images 子目标数据增强和自定义损失函数在3D脑MR图像中的自动解剖地标定位

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/CCEM50674.2020.00016

Srinivasa Rao Kundeti, Md Afzal Ansari, A. Mv, Sumit Sharma, Jeevan Cr

{"title":"Sub-Object Data Augmentation And Custom Loss Functions for Automated Anatomical Landmarks Localisation in 3D Brain MR Images","authors":"Srinivasa Rao Kundeti, Md Afzal Ansari, A. Mv, Sumit Sharma, Jeevan Cr","doi":"10.1109/CCEM50674.2020.00016","DOIUrl":"https://doi.org/10.1109/CCEM50674.2020.00016","url":null,"abstract":"In medical image analysis, the landmarks detection helps to localize each anatomy for tasks like image registration. Deep learning models help to automate the landmark detection pipeline reducing medical technician efforts. The major challenge faced to apply deep learning in medical image analysis is the limited data available for training these models. In this work, we implemented several approaches for data augmentation to address this problem. We modified 2D Flat-net to 3D Flat-net landmark detection network, later used sub-object data augmentation and custom loss function - Masked loss and Soft-argmax loss. We trained and evaluated the combination of data augmentation and custom loss function. On the test set (n=63), using 3D Flat-net with sub-object augmentation, and masked loss function performed the best with Mean Absolute Error improved by 43 % (from 9.84 mm to 5.76 mm), Mean Euclidean Error improved by 34 % (19.94 mm to 13.16 mm). Root Mean Square Error (RMSE) improved by 39 %(from 11.52 mm to 7 mm) compared to Soft-argmax loss function and no augmentation. Data Augmentation of sub-objects and masked loss function improved 3D Flat-net performance for localizing landmarks on medical images.","PeriodicalId":160355,"journal":{"name":"2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123374554","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

Develop CSR Themes using Text-Mining and Topic Modelling Techniques 使用文本挖掘和主题建模技术开发CSR主题

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/CCEM50674.2020.00024

Niveditha R, N. Ns, Mastan Rao Parimi, Aaditya Raam S, S. Babu

引用次数: 0

Energy Efficiency for HDFS HDFS的能源效率

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/CCEM50674.2020.00014

Abhishek Das, Bhargav Snv, N. S. Reddy, Phalachandra H L

{"title":"Energy Efficiency for HDFS","authors":"Abhishek Das, Bhargav Snv, N. S. Reddy, Phalachandra H L","doi":"10.1109/CCEM50674.2020.00014","DOIUrl":"https://doi.org/10.1109/CCEM50674.2020.00014","url":null,"abstract":"The rapidly increasing growth of big data is leading to increased usage of distributed computing and clusters as this can enable tasks to be split and processed in parallel resulting in higher computational efficiency. Environments like Apache Hadoop use Distributed file systems like the HDFS to support large clusters of typically commodity hardware to achieve the same. Commodity hardware present in HDFS consumes significant energy even when not in an active state, and considering our fragile ecosystem, reduction of the carbon footprint and conservation of energy in these environments is the focus of our work. We have evolved on a dynamic approach to Hadoop Distributed File Systems (HDFS) that provides an energy efficient paradigm for Hadoop clusters by segregating the servers in the cluster to two “Zones”, a “Hot-Zone”, and a “Cold-Zone”. We use energy efficient storage devices (SSDs), to store data that is frequently used for the hot zone. The Cold-Zone would have lesser energy efficient devices (HDDs) and would be used to store data that is not frequently used. The Hot-Zone would always be in an active state to compute while the cold zone would have the HDDs to be in a low power consumption state. We have also enhanced the block placement policy for HDFS to support dynamic block transfer among the two zones.","PeriodicalId":160355,"journal":{"name":"2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132958447","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

Intermediate Caching Technique via Fog Layer to Achieve Low-latency Video Analytic Solutions on Cloud 基于雾层的中间缓存技术实现云上低延迟视频分析解决方案

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/CCEM50674.2020.00018

Pankaj M Thakur, S. K. G.

引用次数: 0

ECOSTAR - Energy Conservation in Swift through Tiered Storage Architecture ECOSTAR -通过分级存储架构实现Swift节能

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/CCEM50674.2020.00015

Sanjay K A, S. Sivakumar, S. S. Shreya, Phalachandra H L

{"title":"ECOSTAR - Energy Conservation in Swift through Tiered Storage Architecture","authors":"Sanjay K A, S. Sivakumar, S. S. Shreya, Phalachandra H L","doi":"10.1109/CCEM50674.2020.00015","DOIUrl":"https://doi.org/10.1109/CCEM50674.2020.00015","url":null,"abstract":"The emergence of modern social media that rapidly deploy web-scale applications combined with growing IoT businesses, is leading to an astronomical increase in unstructured data on the Cloud. This is leading to an increase in the number and size of data centers that consume significant power. Thus, creating energy-efficient storage systems, that reduce the carbon footprint of enterprise data centers is an area of interest and need. We propose ECOSTAR, a cost-effective, tiered object storage architecture modeled for OpenStack Swift, that is aimed at significantly optimizing the energy consumption in data centers while keeping availability constraints in check. There has been work towards using Solid State Drives (SSDs) as a staging layer on top of traditional HDDs, to improve request performance and provide greater energy efficiency. In the proposed hierarchical architecture, we extend the tiering approach by introducing a layer of Tape Drives that will be used to store replicas of partitions that are infrequently accessed, providing a noteworthy reduction in energy consumption. Tape Drives also provide increased durability while significantly reducing the architectural setup costs. Tapes, being a means of offline storage, also increases security against cyber attacks. The simulation results using a Dropbox-like workload show that ECOSTAR was able to reduce total disk energy consumption by 25.83 % and improve performance by 27.34 % when compared to traditional Swift.","PeriodicalId":160355,"journal":{"name":"2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128967064","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

Biometric Authentication Security: An Overview 生物识别认证安全:概述

2020 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) Pub Date : 2020-11-01 DOI: 10.1109/CCEM50674.2020.00027

Saleh Alwahaishi, J. Zdralek

引用次数: 2