In recent years, cloud service has been used in many field. In particular, a micro datacenter based on edge computing has been used to overcome the limitations of latency and bandwidth of cloud network architecture. Micro datacentre is small-scaled to reduce environmental constraints and minimize the complexity of fieldwork. Since there are not enough computing resources when compared with existing datacentre, efficient operation methods of services are being researched. Migration is used as a way to duplicate services and load balance services. However, migration is executed when the workload was overloading, delay may occur. This paper propose novel method MSFL(Migration Selection Method using two Factors based LSTM) that provides service availability by predicting power values based on past resource utilization rates and Long Short-Term Memory(LSTM) and pre-selecting migration target based on the relationship between servers and workloads, and values predicted. This method calculates the power consumption of servers and workloads by collecting the utilization of workload resource(CPU, Memory, Network, Disk I/O). The collected power consumption is an input value of LSTM to predict future workload is defined by analyzing the resource utilization of the workload and the server, the weight of the LSTM result is calculated using the two factors. Finally, the target of migration is selected.
近年来，云服务在许多领域得到了应用。特别是基于边缘计算的微数据中心，克服了云网络架构的延迟和带宽限制。微型数据中心规模较小，以减少环境限制并最大限度地降低现场工作的复杂性。由于与现有的数据中心相比，计算资源不足，因此正在研究有效的业务操作方法。迁移被用作复制服务和负载平衡服务的一种方式。但是，迁移是在工作负载过载时执行的，可能会出现延迟。本文提出了一种新的迁移选择方法MSFL(Migration Selection method using two Factors based LSTM)，该方法基于过去的资源利用率和LSTM预测功率值，并根据服务器和工作负载之间的关系以及预测值预先选择迁移目标，从而提供服务可用性。该方法通过收集工作负载资源(CPU、内存、网络、磁盘I/O)的利用率来计算服务器和工作负载的功耗。收集到的功耗是LSTM预测未来工作负载的输入值，通过分析工作负载和服务器的资源利用率来定义，使用这两个因素计算LSTM结果的权重。最后，选择迁移目标。
{"title":"Migration Selection Method using two Factors based LSTM in Micro DataCenter","authors":"Su June Lee, J. An, Younghwan Kim","doi":"10.1145/3400286.3418280","DOIUrl":"https://doi.org/10.1145/3400286.3418280","url":null,"abstract":"In recent years, cloud service has been used in many field. In particular, a micro datacenter based on edge computing has been used to overcome the limitations of latency and bandwidth of cloud network architecture. Micro datacentre is small-scaled to reduce environmental constraints and minimize the complexity of fieldwork. Since there are not enough computing resources when compared with existing datacentre, efficient operation methods of services are being researched. Migration is used as a way to duplicate services and load balance services. However, migration is executed when the workload was overloading, delay may occur. This paper propose novel method MSFL(Migration Selection Method using two Factors based LSTM) that provides service availability by predicting power values based on past resource utilization rates and Long Short-Term Memory(LSTM) and pre-selecting migration target based on the relationship between servers and workloads, and values predicted. This method calculates the power consumption of servers and workloads by collecting the utilization of workload resource(CPU, Memory, Network, Disk I/O). The collected power consumption is an input value of LSTM to predict future workload is defined by analyzing the resource utilization of the workload and the server, the weight of the LSTM result is calculated using the two factors. Finally, the target of migration is selected.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132239547","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}
With successful applications of machine learning to various domains, there have been large demands on developing machine learning-based applications. Automated machine learning is crucial to meet the demand because there are not sufficiently many expert machine learning developers to support such various demands. This paper presents an automated machine learning platform which gets some basic information about a task from nonexpert developer and examines several candidate models to develop an effective machine learning model. To choose some candidate machine learning pipeline for the given task, the platform makes use of the HTN-based plans to describe the machine learning plans along with its application conditions. The prototype system has been developed to mainly support machine learning models for tabular data including time-series data.
{"title":"An Automated Machine Learning Platform for Non-experts","authors":"Jin Han, Ki Sun Park, K. Lee","doi":"10.1145/3400286.3418276","DOIUrl":"https://doi.org/10.1145/3400286.3418276","url":null,"abstract":"With successful applications of machine learning to various domains, there have been large demands on developing machine learning-based applications. Automated machine learning is crucial to meet the demand because there are not sufficiently many expert machine learning developers to support such various demands. This paper presents an automated machine learning platform which gets some basic information about a task from nonexpert developer and examines several candidate models to develop an effective machine learning model. To choose some candidate machine learning pipeline for the given task, the platform makes use of the HTN-based plans to describe the machine learning plans along with its application conditions. The prototype system has been developed to mainly support machine learning models for tabular data including time-series data.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131649578","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}
Vincent Bushong, Russell Sanders, Jacob Curtis, Mark Du, T. Cerný, Karel Frajták, Miroslav Bures, Pavel Tisnovsky, Dongwan Shin
Logging is a vital part of the software development process. Developers use program logging to monitor program execution and identify errors and anomalies. These errors may also cause uncaught exceptions and generate stack traces that help identify the point of error. Both of these sources contain information that can be matched to points in the source code, but manual log analysis is challenging for large systems that create large volumes of logs and have large codebases. In this paper, we contribute a systematic mapping study to determine the state-of-the-art tools and methods used to perform automatic log analysis and stack trace analysis and match the extracted information back to the program's source code. We analyzed 16 publications that address this issue, summarizing their strategies and goals, and we identified open research directions from this body of work.
{"title":"On Matching Log Analysis to Source Code: A Systematic Mapping Study","authors":"Vincent Bushong, Russell Sanders, Jacob Curtis, Mark Du, T. Cerný, Karel Frajták, Miroslav Bures, Pavel Tisnovsky, Dongwan Shin","doi":"10.1145/3400286.3418262","DOIUrl":"https://doi.org/10.1145/3400286.3418262","url":null,"abstract":"Logging is a vital part of the software development process. Developers use program logging to monitor program execution and identify errors and anomalies. These errors may also cause uncaught exceptions and generate stack traces that help identify the point of error. Both of these sources contain information that can be matched to points in the source code, but manual log analysis is challenging for large systems that create large volumes of logs and have large codebases. In this paper, we contribute a systematic mapping study to determine the state-of-the-art tools and methods used to perform automatic log analysis and stack trace analysis and match the extracted information back to the program's source code. We analyzed 16 publications that address this issue, summarizing their strategies and goals, and we identified open research directions from this body of work.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131835517","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}
Solid-state drives (SSDs) that consist of flash memory have the advantages of non-volatility, fast speed, shock resistance, low-power consumption, and small size. Two critical characteristics of flash memory are that it does not support in-place updates, and it must write data in units of a page and erase data in units of a block. Due to the two characteristics, when a block is selected as a victim block to erase, we need to copy the remaining valid pages from the victim block to another free block and the additional copy overhead is called write amplification (WA). Therefore, how to reduce the write amplification (WA) is a crucial issue for SSDs. By performing data classification, it is effective to concentrate the invalid pages in specific blocks and decrease the distribution of invalid pages in the flash memory. The advantage is that we can reduce the write amplification due to the valid pages copied. In the paper, we will propose a machine-learning-based data classifier to classify the written data. Data with similar characteristics will be eventually written in the same group of data blocks in flash memory. Through such a design, it can improve the performance of SSDs by concentrating the invalid pages in the same block and reduce the write amplification.
{"title":"A Machine-Learning-based Data Classifier to Reduce the Write Amplification in SSDs","authors":"Yi-Ying Lu, Chin-Hsien Wu, Ya-Shu Chen","doi":"10.1145/3400286.3418239","DOIUrl":"https://doi.org/10.1145/3400286.3418239","url":null,"abstract":"Solid-state drives (SSDs) that consist of flash memory have the advantages of non-volatility, fast speed, shock resistance, low-power consumption, and small size. Two critical characteristics of flash memory are that it does not support in-place updates, and it must write data in units of a page and erase data in units of a block. Due to the two characteristics, when a block is selected as a victim block to erase, we need to copy the remaining valid pages from the victim block to another free block and the additional copy overhead is called write amplification (WA). Therefore, how to reduce the write amplification (WA) is a crucial issue for SSDs. By performing data classification, it is effective to concentrate the invalid pages in specific blocks and decrease the distribution of invalid pages in the flash memory. The advantage is that we can reduce the write amplification due to the valid pages copied. In the paper, we will propose a machine-learning-based data classifier to classify the written data. Data with similar characteristics will be eventually written in the same group of data blocks in flash memory. Through such a design, it can improve the performance of SSDs by concentrating the invalid pages in the same block and reduce the write amplification.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128869117","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}
Providing an intelligent campus guiding application is a good example of smart campus. Our smartphone APP focused on monitor the drivers' illegal behaviors, such as illegal parking for safety and management issues. For illegal parking detection, the motion mode recognition algorithm is focused on to detect the walking mode after drivers park their cars. When distinguishing low-speed driving mode and walking mode, the accuracy rate of Decision Tree algorithm is 67.03%, and that of our algorithm is 89.91%. When the walking mode is detected, the car location will be computed and if the location is not within the parking lot, the illegal parking occurs and the illegal parking alert will be sent to the APP. The application will play an important role in smart campus to improve the traffic safety and traffic management in campus.
{"title":"Motion Mode Recognition for Traffic Safety in Campus Guiding Application","authors":"Rukang Yan, Zhan Gao, Lianbing Xu, Lei Cai, Zhengtao Xiang, Yufeng Chen","doi":"10.1145/3400286.3418246","DOIUrl":"https://doi.org/10.1145/3400286.3418246","url":null,"abstract":"Providing an intelligent campus guiding application is a good example of smart campus. Our smartphone APP focused on monitor the drivers' illegal behaviors, such as illegal parking for safety and management issues. For illegal parking detection, the motion mode recognition algorithm is focused on to detect the walking mode after drivers park their cars. When distinguishing low-speed driving mode and walking mode, the accuracy rate of Decision Tree algorithm is 67.03%, and that of our algorithm is 89.91%. When the walking mode is detected, the car location will be computed and if the location is not within the parking lot, the illegal parking occurs and the illegal parking alert will be sent to the APP. The application will play an important role in smart campus to improve the traffic safety and traffic management in campus.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114586435","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}
Seoyeon Kim, Jaehyeok Jeong, Jaehee Kim, Young-Sun Yun, Bongjae Kim, Jinmang Jung
With the recent progress in SNN(Spiking Neural Networks), IoT applications require more intelligent data processing that can operate on neuromorphic hardware. However, as there is no development environment where SNN frameworks can be applied to existing IoT applications, it is difficult for IoT developers to utilize neuromorphic hardware. In this paper, we present a new development framework of IoT applications using SNN as a machine learning solution, called NA-Designer. It has GUI based neural network editor to quickly generate SNN components that can be run on various neuromorphic hardware. In particular, this component is designed to interwork with other IoT framework such as Node-RED. We implement NA-Designer and core neural networks libraries and measure its performance. Our framework can be used to develop Edge computing applications that utilize neuromorphic hardware to reduce energy consumption and latency.
{"title":"Developing IoT Applications Using Spiking Neural Networks Framework","authors":"Seoyeon Kim, Jaehyeok Jeong, Jaehee Kim, Young-Sun Yun, Bongjae Kim, Jinmang Jung","doi":"10.1145/3400286.3418271","DOIUrl":"https://doi.org/10.1145/3400286.3418271","url":null,"abstract":"With the recent progress in SNN(Spiking Neural Networks), IoT applications require more intelligent data processing that can operate on neuromorphic hardware. However, as there is no development environment where SNN frameworks can be applied to existing IoT applications, it is difficult for IoT developers to utilize neuromorphic hardware. In this paper, we present a new development framework of IoT applications using SNN as a machine learning solution, called NA-Designer. It has GUI based neural network editor to quickly generate SNN components that can be run on various neuromorphic hardware. In particular, this component is designed to interwork with other IoT framework such as Node-RED. We implement NA-Designer and core neural networks libraries and measure its performance. Our framework can be used to develop Edge computing applications that utilize neuromorphic hardware to reduce energy consumption and latency.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128256983","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}
Heejin Kim, Kyuho Kim, Jiman Hong, Jun-Young Heo, Joongjin Kook
A dynamic Android application(app) analysis tool is as known as useful for detecting errors or vulnerabilities in the Android apps at runtime by observing the internal behavior changes like specific method calls which can make malicious behaviors. However, existing Android app analysis tools not only cannot extract all methods used in the Android app but also cannot extract code blocks that can check the branch condition of the program. In this paper, we propose EDAroid, an efficient dynamic analysis tool for Android apps. The proposed EDAroid can dynamically extract not only the core system's methods but also the user-defined methods in Android apps. The proposed EDAroid can also represent the extracted methods and code blocks in Android apps in a graph. We also evaluate the functionality and performance of the proposed EDAroid and the evaluation results show that the proposed EDAroid is more efficient than the existing Android app analysis tool.
{"title":"EDAroid","authors":"Heejin Kim, Kyuho Kim, Jiman Hong, Jun-Young Heo, Joongjin Kook","doi":"10.1145/3400286.3418266","DOIUrl":"https://doi.org/10.1145/3400286.3418266","url":null,"abstract":"A dynamic Android application(app) analysis tool is as known as useful for detecting errors or vulnerabilities in the Android apps at runtime by observing the internal behavior changes like specific method calls which can make malicious behaviors. However, existing Android app analysis tools not only cannot extract all methods used in the Android app but also cannot extract code blocks that can check the branch condition of the program. In this paper, we propose EDAroid, an efficient dynamic analysis tool for Android apps. The proposed EDAroid can dynamically extract not only the core system's methods but also the user-defined methods in Android apps. The proposed EDAroid can also represent the extracted methods and code blocks in Android apps in a graph. We also evaluate the functionality and performance of the proposed EDAroid and the evaluation results show that the proposed EDAroid is more efficient than the existing Android app analysis tool.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131690202","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}
T. Fladby, H. Haugerud, S. Nichele, Kyrre M. Begnum, A. Yazidi
Machine-learning based Intrusion Detection and Prevention Systems provide significant value to organizations because they can efficiently detect previously unseen variations of known threats, new threats related to known malware or even zero-day malware, unrelated to any other known threats. However, while such systems prove invaluable to security personnel, researchers have observed that data subject to inspection by behavioral analysis can be perturbed in order to evade detection. We investigated the use of adversarial techniques for adapting the communication patterns between botnet malware and control unit in order to evaluate the robustness of an existing Network Behavioral Analysis solution. We implemented a packet parser that let us extract and edit certain properties of network flows and automated an approach for conducting a grey-box testing scheme of Stratosphere Linux IPS. As part of our implementation, we provided several techniques for providing perturbation to network flow parameters, including a Simultaneous Perturbation Stochastic Approximation method, which was able to produce sufficiently perturbed network flow patterns while adhering to an underlying objective function. Our results showed that network flow parameters could indeed be perturbed to ultimately enable evasion of intrusion detection based on the detection models that were used with the Intrusion Detection System. Additionally, we demonstrated that it was possible to combine evading detection with techniques for optimization problems that aimed to minimize the magnitude of perturbation to network flows, effectively enabling adaptive network flow behavior.
基于机器学习的入侵检测和防御系统为组织提供了重要的价值，因为它们可以有效地检测到以前未见过的已知威胁的变化，与已知恶意软件相关的新威胁，甚至与任何其他已知威胁无关的零日恶意软件。然而，虽然这些系统对安全人员来说是无价的，但研究人员观察到，受行为分析检查的数据可能会受到干扰，以逃避检测。我们研究了使用对抗技术来适应僵尸网络恶意软件和控制单元之间的通信模式，以评估现有网络行为分析解决方案的鲁棒性。我们实现了一个数据包解析器，它允许我们提取和编辑网络流的某些属性，并自动执行一种方法来执行Stratosphere Linux IPS的灰盒测试方案。作为我们实现的一部分，我们提供了几种技术来提供对网络流量参数的扰动，包括同步扰动随机逼近方法，该方法能够在坚持潜在目标函数的同时产生充分扰动的网络流量模式。我们的研究结果表明，基于入侵检测系统所使用的检测模型，网络流参数确实可以被扰动，从而最终能够逃避入侵检测。此外，我们证明了可以将规避检测与优化问题的技术相结合，旨在最大限度地减少对网络流的扰动，有效地实现自适应网络流行为。
{"title":"Evading a Machine Learning-based Intrusion Detection System through Adversarial Perturbations","authors":"T. Fladby, H. Haugerud, S. Nichele, Kyrre M. Begnum, A. Yazidi","doi":"10.1145/3400286.3418252","DOIUrl":"https://doi.org/10.1145/3400286.3418252","url":null,"abstract":"Machine-learning based Intrusion Detection and Prevention Systems provide significant value to organizations because they can efficiently detect previously unseen variations of known threats, new threats related to known malware or even zero-day malware, unrelated to any other known threats. However, while such systems prove invaluable to security personnel, researchers have observed that data subject to inspection by behavioral analysis can be perturbed in order to evade detection. We investigated the use of adversarial techniques for adapting the communication patterns between botnet malware and control unit in order to evaluate the robustness of an existing Network Behavioral Analysis solution. We implemented a packet parser that let us extract and edit certain properties of network flows and automated an approach for conducting a grey-box testing scheme of Stratosphere Linux IPS. As part of our implementation, we provided several techniques for providing perturbation to network flow parameters, including a Simultaneous Perturbation Stochastic Approximation method, which was able to produce sufficiently perturbed network flow patterns while adhering to an underlying objective function. Our results showed that network flow parameters could indeed be perturbed to ultimately enable evasion of intrusion detection based on the detection models that were used with the Intrusion Detection System. Additionally, we demonstrated that it was possible to combine evading detection with techniques for optimization problems that aimed to minimize the magnitude of perturbation to network flows, effectively enabling adaptive network flow behavior.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124682926","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}
Container-based Micro Service technology has problems with the limitations of resource expansion and the inability to move services between container platforms to respond to growing user traffic. For this reason, distributed collaborative container plat-form(DCCP) technology has emerged that provides the expansion and availability of services and enables collaboration among locally distributed container platforms. DCCP is an essential technology to overcome the limitations of existing stand-alone container platforms by maximizing flexible expansion of resources and service mobility through collaboration between distributed platforms. In this paper, the inter-platform traffic load balancing technology is proposed to construct DDCP. The proposed technology analyzes the user's traffic, identifies the user's geographic location information, and calculates scores for each distributed platform. The scoring method calculates the geographical score based on the platform's geographical location information and the user's geographical location, and calculates the resource score by analyzing the platform's resource status. Transport traffic through the calculated geographical scores and resource scores to the platform that holds the highest scores. The proposed technology provides analytical load balancing technology that addresses the non-efficient problem of randomly dispersing loads on existing independent platforms by load balancing according to the location and resource status of the platform.
{"title":"Design and Implementation of Analytical Load Balancing between Distributed Collaborative Container Platforms","authors":"Jae-Seung Han, J. An, Younghwan Kim","doi":"10.1145/3400286.3418277","DOIUrl":"https://doi.org/10.1145/3400286.3418277","url":null,"abstract":"Container-based Micro Service technology has problems with the limitations of resource expansion and the inability to move services between container platforms to respond to growing user traffic. For this reason, distributed collaborative container plat-form(DCCP) technology has emerged that provides the expansion and availability of services and enables collaboration among locally distributed container platforms. DCCP is an essential technology to overcome the limitations of existing stand-alone container platforms by maximizing flexible expansion of resources and service mobility through collaboration between distributed platforms. In this paper, the inter-platform traffic load balancing technology is proposed to construct DDCP. The proposed technology analyzes the user's traffic, identifies the user's geographic location information, and calculates scores for each distributed platform. The scoring method calculates the geographical score based on the platform's geographical location information and the user's geographical location, and calculates the resource score by analyzing the platform's resource status. Transport traffic through the calculated geographical scores and resource scores to the platform that holds the highest scores. The proposed technology provides analytical load balancing technology that addresses the non-efficient problem of randomly dispersing loads on existing independent platforms by load balancing according to the location and resource status of the platform.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129129179","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}
This work designs and implements a calibration method between a narrow field of view camera and a depth camera in an endoscope-like scenario. An endoscopy-like scenario has a limited specular reflective surface, a camera with a narrow field of view. Instead of pushing the accuracy of the target marker with low-resolution data, we design a new loss function, which utilizes all of the three dimensions points of the checkerboard measured with the depth camera, and calculates the distance between projected 3D positions onto 2D image surface and the color image. The final re-projected error is improved to be less than 1 millimeters on average.
{"title":"An Extrinsic Depth Camera Calibration Method for Narrow Field of View Color Camera","authors":"C. Shih, Hao-Yu Chen","doi":"10.1145/3400286.3418251","DOIUrl":"https://doi.org/10.1145/3400286.3418251","url":null,"abstract":"This work designs and implements a calibration method between a narrow field of view camera and a depth camera in an endoscope-like scenario. An endoscopy-like scenario has a limited specular reflective surface, a camera with a narrow field of view. Instead of pushing the accuracy of the target marker with low-resolution data, we design a new loss function, which utilizes all of the three dimensions points of the checkerboard measured with the depth camera, and calculates the distance between projected 3D positions onto 2D image surface and the color image. The final re-projected error is improved to be less than 1 millimeters on average.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114178292","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：