There are several motivations, such as mobility, cost, and secu- rity, that are behind the trend of traditional desktop users transi- tioning to thin-client-based virtual desktop clouds （VDCs）. Such a trend has led ...There are several motivations, such as mobility, cost, and secu- rity, that are behind the trend of traditional desktop users transi- tioning to thin-client-based virtual desktop clouds （VDCs）. Such a trend has led to the rising importance of human-centric performance modeling and assessment within user communities that are increasingly making use of desktop virtualization. In this paper, we present a novel reference architecture and its eas- ily deployable implementation for modeling and assessing objec- tive user quality of experience （QoE） in VDCs. This architec- ture eliminates the need for expensive, time-consuming subjec- tive testing and incorporates finite-state machine representa- tions for user workload generation. It also incorporates slow-mo- tion benchmarking with deep-packet inspection of application task performance affected by QoS variations. In this way, a ＂composite-quality＂ metric model of user QoE can be derived. We show how this metric can be customized to a particular user group profile with different application sets and can be used to a） identify dominant performance indicators and troubleshoot bottlenecks and b） obtain both absolute and relative objective user QoE measurements needed for pertinent selection of thin-client encoding configurations in VDCs. We validate our composite-quality modeling and assessment methodology by us- ing subjective and objective user QoE measurements in a re- al-world VDC called VDPilot, which uses RDP and PCoIP thin-client protocols. In our case study, actual users are pres- ent in virtual classrooms within a regional federated university system.展开更多
Understanding and controlling the self-assembly of vertically oriented carbon nanotube(CNT)forests is essential for realizing their potential in myriad applications.The governing process–structure–property mechanism...Understanding and controlling the self-assembly of vertically oriented carbon nanotube(CNT)forests is essential for realizing their potential in myriad applications.The governing process–structure–property mechanisms are poorly understood,and the processing parameter space is far too vast to exhaustively explore experimentally.We overcome these limitations by using a physics-based simulation as a high-throughput virtual laboratory and image-based machine learning to relate CNT forest synthesis attributes to their mechanical performance.Using CNTNet,our image-based deep learning classifier module trained with synthetic imagery,combinations of CNT diameter,density,and population growth rate classes were labeled with an accuracy of>91%.The CNTNet regression module predicted CNT forest stiffness and buckling load properties with a lower root-mean-square error than that of a regression predictor based on CNT physical parameters.These results demonstrate that image-based machine learning trained using only simulated imagery can distinguish subtle CNT forest morphological features to predict physical material properties with high accuracy.CNTNet paves the way to incorporate scanning electron microscope imagery for high-throughput material discovery.展开更多
Domain name system(DNS)amplification distributed denial of service(DDoS)attacks are one of the popular types of intrusions that involve accessing DNS servers on behalf of the victim.In this case,the size of the respon...Domain name system(DNS)amplification distributed denial of service(DDoS)attacks are one of the popular types of intrusions that involve accessing DNS servers on behalf of the victim.In this case,the size of the response is many times greater than the size of the request,in which the source of the request is substituted for the address of the victim.This paper presents an original method for countering DNS amplification DDoS attacks.The novelty of our approach lies in the analysis of outgoing traffic from the victim’s server.DNS servers used for amplification attacks are easily detected in Internet control message protocol(ICMP)packet headers(type 3,code 3)in outgoing traffic.ICMP packets of this type are generated when accessing closed user datagram protocol(UDP)ports of the victim,which are randomly assigned by the Saddam attack tool.To prevent such attacks,we used a Linux utility and a software-defined network(SDN)module that we previously developed to protect against port scanning.The Linux utility showed the highest efficiency of 99.8%,i.e.,only two attack packets out of a thousand reached the victim server.展开更多
基金supported by VMware and the National Science Foundation under award numbers CNS-1050225 and CNS-1205658
文摘There are several motivations, such as mobility, cost, and secu- rity, that are behind the trend of traditional desktop users transi- tioning to thin-client-based virtual desktop clouds （VDCs）. Such a trend has led to the rising importance of human-centric performance modeling and assessment within user communities that are increasingly making use of desktop virtualization. In this paper, we present a novel reference architecture and its eas- ily deployable implementation for modeling and assessing objec- tive user quality of experience （QoE） in VDCs. This architec- ture eliminates the need for expensive, time-consuming subjec- tive testing and incorporates finite-state machine representa- tions for user workload generation. It also incorporates slow-mo- tion benchmarking with deep-packet inspection of application task performance affected by QoS variations. In this way, a ＂composite-quality＂ metric model of user QoE can be derived. We show how this metric can be customized to a particular user group profile with different application sets and can be used to a） identify dominant performance indicators and troubleshoot bottlenecks and b） obtain both absolute and relative objective user QoE measurements needed for pertinent selection of thin-client encoding configurations in VDCs. We validate our composite-quality modeling and assessment methodology by us- ing subjective and objective user QoE measurements in a re- al-world VDC called VDPilot, which uses RDP and PCoIP thin-client protocols. In our case study, actual users are pres- ent in virtual classrooms within a regional federated university system.
基金The authors would like to acknowledge funding from National Science Foundation(NSF)under award CCMI 2026847 and CMMI 1651538(for T.H.and M.R.M.)partial support from NSF MRI CNS-1429294 and Army Research Laboratory award W911NF-1820285(for K.P.,R.B.,and F.B.)+1 种基金The computation for this work was performed on a GPU cluster from the Army Research Office DURIP award W911NF1910181Any opinions,findings,and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the U.S.Government or agency thereof.
文摘Understanding and controlling the self-assembly of vertically oriented carbon nanotube(CNT)forests is essential for realizing their potential in myriad applications.The governing process–structure–property mechanisms are poorly understood,and the processing parameter space is far too vast to exhaustively explore experimentally.We overcome these limitations by using a physics-based simulation as a high-throughput virtual laboratory and image-based machine learning to relate CNT forest synthesis attributes to their mechanical performance.Using CNTNet,our image-based deep learning classifier module trained with synthetic imagery,combinations of CNT diameter,density,and population growth rate classes were labeled with an accuracy of>91%.The CNTNet regression module predicted CNT forest stiffness and buckling load properties with a lower root-mean-square error than that of a regression predictor based on CNT physical parameters.These results demonstrate that image-based machine learning trained using only simulated imagery can distinguish subtle CNT forest morphological features to predict physical material properties with high accuracy.CNTNet paves the way to incorporate scanning electron microscope imagery for high-throughput material discovery.
基金Russian Foundation for Basic Research(RFBR)(20-37-90002)Andrei Sukhov acknowledge SevSU for a Research(42-01-09/253/2022-1)。
文摘Domain name system(DNS)amplification distributed denial of service(DDoS)attacks are one of the popular types of intrusions that involve accessing DNS servers on behalf of the victim.In this case,the size of the response is many times greater than the size of the request,in which the source of the request is substituted for the address of the victim.This paper presents an original method for countering DNS amplification DDoS attacks.The novelty of our approach lies in the analysis of outgoing traffic from the victim’s server.DNS servers used for amplification attacks are easily detected in Internet control message protocol(ICMP)packet headers(type 3,code 3)in outgoing traffic.ICMP packets of this type are generated when accessing closed user datagram protocol(UDP)ports of the victim,which are randomly assigned by the Saddam attack tool.To prevent such attacks,we used a Linux utility and a software-defined network(SDN)module that we previously developed to protect against port scanning.The Linux utility showed the highest efficiency of 99.8%,i.e.,only two attack packets out of a thousand reached the victim server.
