A machine learning approach for predicting computational intensity and domain decomposition in parallel geoprocessing

ABSTRACT

High performance computing is required for fast geoprocessing of geospatial big data. Using spatial domains to represent computational intensity (CIT) and domain decomposition for parallelism are prominent strategies when designing parallel geoprocessing applications. Traditional domain decomposition is limited in evaluating the computational intensity, which often results in load imbalance and poor parallel performance. From the data science perspective, machine learning from Artificial Intelligence (AI) shows promise for better CIT evaluation. This paper proposes a machine learning approach for predicting computational intensity, followed by an optimized domain decomposition, which divides the spatial domain into balanced subdivisions based on the predicted CIT to achieve better parallel performance. The approach provides a reference framework on how various machine learning methods including feature selection and model training can be used in predicting computational intensity and optimizing parallel geoprocessing against different cases. Some comparative experiments between the approach and traditional methods were performed using the two cases, DEM generation from point clouds and spatial intersection on vector data. The results not only demonstrate the advantage of the approach, but also provide hints on how traditional GIS computation can be improved by the AI machine learning.     

A possible method for the determination of the magnitude and inclination of a sunspot magnetic-field

A method is presented in this paper for the determination of both the H and the γ of a sunspot magnetic field by means of two polarizers with transmission axes inclined at ± 45° to the fast axis of a quarter-wave plate. Previously, this arrangement was used to determine H only.It is shown that the profiles of the Stokes parameters are sensitive to the sunspot atmosphere model used whereas the results given by the present method are not. It is shown, more ever, that the formula for determining H from the wave length shift, Δλ = 4.67 × 10^?5gHλ², cannot be used in many cases.     

行星天文学的重要地位及其热点

对1995年至2001年《Natue》和《Science》上发表的天文学论文的统计表明，行星天文学领域的论文数量明显超过天文学的其它分支学科，占天文学总论文数的1/3左右。从这个角度来看，行星天文学是天文学最活跃和重要的分支学科之一。对这些论文具体内容的分析可以给出当前行星天文学领域的若干热点问题。相比之下，我国对这一重要领域的关注和投入还远远不够。     

STUDY ON DEVELOPMENT STRATEGY OF SHIPPING CENTERS AND TRANSPORTATION NETWORKS IN THE YELLOW SEA RIM IN THE YELLOW SEA RIM

It is very important to establish cooperative mechanism to guarantee all members to develop their e-conomies in the Yellow Sea Rim. In this paper, the development strategies ofshipplng centers and transportation networkare discussed based on economic giobalization tendency. The results argue that a united transportation network should bebuilt in order to promote the economic competition of Northeast Asia in the world. As a key component of the economiccooperation, a hierarchical shipping centers network should be established with Hong Kong, Shanghai, Pusan, Koho,and Tokyo as cores. The authorities of China, Japan, R. O. Korea and D. P. B. Korea should make more efforts tobuild a set of cooperation institutions based on raising the transportation efficiency.     

A Linear Dependence between the Magnitude and the Logarithm of the Length of the Time the Quasars OJ287 and 3C345 Stay below this Magnitude

We have studied the optical historical light curves of two Active Galactic Nuclei (AGNs): OJ287 and 3C345. We find a linear dependence between the magnitude and the logarithm of the length of the time the objects stay below this magnitude. The relation may originate from a fractal structure of the light curve. This revised version was published online in July 2006 with corrections to the Cover Date.     

Energy flow in Branchiura sowerbyi (Oligochaeta: tubificidae) in a shallow macrophyte-dominated lake, Biandantang Lake

The energy flow ofBranchiura sowerbyi was studied for the first time in China in a shallow macrophytic lake, Biandantang Lake, Hubei Province. The energy flow was calculated from the measurement of flesh production (12.5241kJ/m²a), egestion (517.7302kJ/m²a), metabolism (38.3273 kJ/m²a), and excretion (4.3798kJ/m²a). The net growth efficiency of the species is about 22.7%, which accords well with the generally reported value for oligochaetes. In addition, the relationship between starvation respiration (R, mgO₂/ind·d), wet weight (Ww, mg) and temperature (T, °C) were also measured, with the regression function beingR=0.008Ww^0.736 e^0.050T. Project supported by NSFC (30270278, 3960019), the foundation of the government of Hubei Province (No. 2000J109), and the foundation of Ecological Station, CAS in the Institute of Hydrobiology.     

5个新发现的X选BL Lac天体

利用ＲＯＳＡＴＶＬＡ方法筛选，从ＲＯＳＡＴＸ射线源中选出了一批新的ＢＬＬａｃ天体和类星体的候选体．１９９６年１２月８日至１７日，利用北京天文台２．１６ｍ望远镜和新从美国引进的ＯＭＲ摄谱仪，对这批新候选体进行了光谱认证．经ＳＵＮ工作站处理，又发现了５个新的Ｘ选ＢＬＬａｃ天体．此外，还对去年作者在ＯＭＲ引进之前发现的ＢＬＬａｃ天体进行了检验，结果发现，去年发现的７个ＢＬＬａｃ天体中，２３２２＋３４３的ＣａＩＩＨ＆Ｋ“Ｂｒｅａｋｓｔｒｅｎｇｔｈ”为２６．９％，稍大于判据２５％，因此，２３２２＋３４３到底是ＢＬＬａｃ天体还是射电星系还有待于进一步测光及偏振观测的验证．     

ENERGY FLOW OF BELIAMYA AERUGINOSA IN A SHALLOW ALGAL LAKE, HOUHU LAKE （WUHAN, CHINA）

The energy budget ofBellamya earuginosa in a shallow algal lake, Houhu Lake (Wuhan, China) was investigated by the measurement of flesh production (32.8 kJ/(m²·a)), egestion (337.7 kJ/(m²·a)), metabolism (246.7 kJ/(m²·a)), and estimation of excretion (21.4kJ/(m²·a)). The net growth efficiency of the species is about 10.9%, which accords with the generally reported value for gastropods. In addition, the relationships between starvation respiration (R, mgO₂/(Ind·d)), body weight (Wd, mg in dry wt) and temperature (T, °C) were also determined. The regression equationR=0.044Wd ^0.537 e ^0.061T was obtained by the least square method, The measured SDA of the species is 26.51% of its gross metabolism. Project No. 3960019 and 39430101 supported by NSFC and also a granted for the East Lake Ecological Experimental Station, CAS, Wuhan.     

On the microwave millisecond spike emission and its associated phenomena during the impulsive phase of large solar flare

We propose a model of two acceleration regions, which can explain, on the basis of microwave maser caused by a “hollow-beam” distribution of electrons, the presence of millisecond spikes in the event of 1981 May 16 and their absence in the event of 1981 October 12, and the enhanced continuous emission in the latter. We have also uncovered relations among the features, e.g. the Type III_G, Type IV_DCIM and hard x-ray bursts, that accompany the microwave millisecond spikes during the impulsive phase of a large flare.