In infrastructure as a service(IaaS)cloud mode equipment simulated training,to keep the resource utilization ratio in a rational high level,improve the training effect and reduce the system running cost,the problem of...In infrastructure as a service(IaaS)cloud mode equipment simulated training,to keep the resource utilization ratio in a rational high level,improve the training effect and reduce the system running cost,the problem of training virtual machine(TVM)placement needs to be resolved first.We make analysis to the problem and give the mathematical formulation to the problem.Then,we figure out the principle and target of the TVM placement.Based on above analysis,we propose a constrained immune memory and immunodominance clone(CIMIC)TVM placement optimization algorithm.By reverse optimization of the initial antibody population,the searching range is reduced.The common antibody population and the immunodominance antibody population evolve simultaneously,which realizes the simultaneous progressing of global searching and local searching of solutions.Further,local optimal is avoided by this means.Memory antibody makes ful use of the unfeasible solutions and the diversity of antibody population is maintained.The constraint information of the problem is utilized to improve the optimization effect.Experiment results show that the CIMIC algorithm improves the overall optimization effect of TVM placement,reduces the server number and improves the resource utilization and system stability.展开更多
Based on normalized six-hourly black body temperature （TBB） data of three geostationary meteorological satellites,the leading modes of the mei-yu cloud system between 1998 and 2008 were extracted by the Empirical Or...Based on normalized six-hourly black body temperature （TBB） data of three geostationary meteorological satellites,the leading modes of the mei-yu cloud system between 1998 and 2008 were extracted by the Empirical Orthogonal Function （EOF） method,and the transition processes from the first typical leading mode to other leading modes were discussed and compared.The analysis shows that,when the southern mode （EOF1） transforms to the northeastern mode （EOF3）,in the mid-troposphere,a low trough develops and moves southeastward over central and eastern China.The circulation pattern is characterized by two highs and one low in the lower troposphere.A belt of low pressure is sandwiched between the weak high over central and western China and the strong western North Pacific subtropical high （WNPSH）.Cold air moves southward along the northerly flow behind the low,and meets the warm and moist air between the WNPSH and the forepart of the low trough,which leads to continuous convection.At the same time,the central extent of the WNPSH increases while its ridge extends westward.In addition,transitions from the southern mode to the dual centers mode and the tropical-low-influenced mode were found to be atypical,and so no common points could be concluded.Furthermore,the choice of threshold value can affect the number of samples discussed.展开更多
For the multi-mode radar working in the modern electronicbattlefield, different working states of one single radar areprone to being classified as multiple emitters when adoptingtraditional classification methods to p...For the multi-mode radar working in the modern electronicbattlefield, different working states of one single radar areprone to being classified as multiple emitters when adoptingtraditional classification methods to process intercepted signals,which has a negative effect on signal classification. A classificationmethod based on spatial data mining is presented to address theabove challenge. Inspired by the idea of spatial data mining, theclassification method applies nuclear field to depicting the distributioninformation of pulse samples in feature space, and digs out thehidden cluster information by analyzing distribution characteristics.In addition, a membership-degree criterion to quantify the correlationamong all classes is established, which ensures classificationaccuracy of signal samples. Numerical experiments show that thepresented method can effectively prevent different working statesof multi-mode emitter from being classified as several emitters,and achieve higher classification accuracy.展开更多
基金Equipment Pre-research Fund of China under Grant No.9140A04030214JB34058.
文摘In infrastructure as a service(IaaS)cloud mode equipment simulated training,to keep the resource utilization ratio in a rational high level,improve the training effect and reduce the system running cost,the problem of training virtual machine(TVM)placement needs to be resolved first.We make analysis to the problem and give the mathematical formulation to the problem.Then,we figure out the principle and target of the TVM placement.Based on above analysis,we propose a constrained immune memory and immunodominance clone(CIMIC)TVM placement optimization algorithm.By reverse optimization of the initial antibody population,the searching range is reduced.The common antibody population and the immunodominance antibody population evolve simultaneously,which realizes the simultaneous progressing of global searching and local searching of solutions.Further,local optimal is avoided by this means.Memory antibody makes ful use of the unfeasible solutions and the diversity of antibody population is maintained.The constraint information of the problem is utilized to improve the optimization effect.Experiment results show that the CIMIC algorithm improves the overall optimization effect of TVM placement,reduces the server number and improves the resource utilization and system stability.
基金supported by the National Natural Science Foundation of China (Grant No. 40975023)the Special Promotion Program for Meteorology (Grant No. GYHY201406011 and No. GYHY201106044)the National High Technology Research and Development Project of China (Grant No. 2012AA120903)
文摘Based on normalized six-hourly black body temperature （TBB） data of three geostationary meteorological satellites,the leading modes of the mei-yu cloud system between 1998 and 2008 were extracted by the Empirical Orthogonal Function （EOF） method,and the transition processes from the first typical leading mode to other leading modes were discussed and compared.The analysis shows that,when the southern mode （EOF1） transforms to the northeastern mode （EOF3）,in the mid-troposphere,a low trough develops and moves southeastward over central and eastern China.The circulation pattern is characterized by two highs and one low in the lower troposphere.A belt of low pressure is sandwiched between the weak high over central and western China and the strong western North Pacific subtropical high （WNPSH）.Cold air moves southward along the northerly flow behind the low,and meets the warm and moist air between the WNPSH and the forepart of the low trough,which leads to continuous convection.At the same time,the central extent of the WNPSH increases while its ridge extends westward.In addition,transitions from the southern mode to the dual centers mode and the tropical-low-influenced mode were found to be atypical,and so no common points could be concluded.Furthermore,the choice of threshold value can affect the number of samples discussed.
基金supported by the National Natural Science Foundation of China(61371172)the International S&T Cooperation Program of China(2015DFR10220)+1 种基金the Ocean Engineering Project of National Key Laboratory Foundation(1213)the Fundamental Research Funds for the Central Universities(HEUCF1608)
文摘For the multi-mode radar working in the modern electronicbattlefield, different working states of one single radar areprone to being classified as multiple emitters when adoptingtraditional classification methods to process intercepted signals,which has a negative effect on signal classification. A classificationmethod based on spatial data mining is presented to address theabove challenge. Inspired by the idea of spatial data mining, theclassification method applies nuclear field to depicting the distributioninformation of pulse samples in feature space, and digs out thehidden cluster information by analyzing distribution characteristics.In addition, a membership-degree criterion to quantify the correlationamong all classes is established, which ensures classificationaccuracy of signal samples. Numerical experiments show that thepresented method can effectively prevent different working statesof multi-mode emitter from being classified as several emitters,and achieve higher classification accuracy.
