The explosive growth of data poses significant challenges for GPU-based databases, which must balance limited memory capacity with the need for high-speed query execution. Compression has become an essential technique for optimizing memory utilization and reducing data movement. However, its benefits have been limited to the necessary data decompression. Querying compressed data conventionally requires decompression, which causes the query process to be significantly slower than a direct query on uncompressed data. To address this problem, this paper presents a novel GPU-accelerated tile-based direct query framework that successfully eliminates the limitation, significantly enhancing query performance. By employing direct query strategies, the framework minimizes data movement and maximizes memory bandwidth utilization. It incorporates tilebased hardware-conscious execution strategies for direct query, including memory management and control flow coordination, to improve execution efficiency. Additionally, adaptive data-driven compression formats are paired with tailored SQL operators to enable efficient support for diverse queries. Our experiments, conducted using the Star Schema Benchmark, show an average improvement of 3.5× compared to the state-of-the-art tilebased decompression scheme, while maintaining the space-saving advantages of compression. Notably, our solution consistently outperforms existing direct execution schemes for compressed data across all query types.

>Sampling is one of the most widely employed approximations in big data processing. Among various challenges in sampling design, sampling for join is particularly intriguing yet complex. This perplexing problem starts with a classical case where the join of two Bernoulli samples shrinks its output size quadratically and exhibits a strong dependency on the input data, presenting a unique challenge that necessitates adaptive sampling to guarantee both the quantity and quality of the sampled data. The community has made strides in achieving this goal by constructing offline samples and integrating support from indexes or key frequencies. However, when dealing with stream data, due to the need for real-time processing and high-quality analysis, methods developed for processing static data become unavailable. Consequently, a fundamental question arises: Is it possible to achieve adaptive sampling in stream data without relying on offline techniques? To address this problem, we propose FreeSam, which couples hybrid sampling with intra-window join, a key stream join operator. Our focus lies on two widely used metrics: output size, ensuring quantity, and variance, ensuring quality. FreeSam enables adaptability in both the desired quantity and quality of data sampling by offering control on the two-dimensional space spanned by these metrics. Meanwhile, adjustable trade-offs between quality and performance make FreeSam practical for use. Our experiments show that, for every 1% increase in latency limitation, FreeSam can yield a 3.83% increase in the output size while maintaining the level of the estimator's variance. Additionally, we give FreeSam a multi-core implementation and ensure predictability of its latency through both an analytic model and a neural network model. The accuracy of these models is 88.05% and 96.75% respectively.

There are plenty of parking spaces in big cities, but we often find nowhere to park. For example, New York has 1.4 million cars and 4.4 million on-street parking spaces, but it is still not easy to find a parking place near our destination, especially during peak hours. The reason is the lack of prediction of parking behavior. If we could provide parking behavior in advance, we can ease this parking problem that affects human well-being. We observe that parking lots have periodic parking patterns, which is an important factor for parking behavior prediction. Unfortunately, existing work ignores such periodic parking patterns in parking behavior prediction, and thus incurs low accuracy. To solve this problem, we propose PewLSTM, a novel periodic weather-aware LSTM model that successfully predicts the parking behavior based on historical records, weather, environments, weekdays, and events. PewLSTM includes a periodic weather-aware LSTM prediction module and an event prediction module, for predicting parking behaviors in regular days and events. PewLSTM is extremely useful for drivers and parking lot owners to improve customer experience. For example, the probability of parking space that will be available soon can be provided even if the parking lot is full. Based on 910,477 real parking records in 904 days from 13 parking lots, PewLSTM yields 93.84% parking prediction accuracy, which is about 30% higher than the state-of-the-art parking behavior prediction method. Additionally, we have analyzed parking behaviors in events like holidays and COVID-19. PewLSTM can handle parking behavior prediction in events and reaches 90.68 percent accuracy.

In big cities, there are plenty of parking spaces, but we often find nowhere to park. For example, New York has 1.4 million cars and 4.4 million on-street parking spaces, but it is still not easy to find a parking place near our destination, especially during peak hours. The reason is the lack of prediction of parking behavior. If we could provide parking behavior in advance, we can ease this parking problem that affects human well-being. We observe that parking lots have periodic parking patterns, which is an important factor for parking behavior prediction. Unfortunately, existing work ignores such periodic parking patterns in parking behavior prediction, and thus incurs low accuracy. To solve this problem, we propose PewLSTM, a novel periodic weather-aware LSTM model that successfully predicts the parking behavior based on historical records, weather, environments, and weekdays. PewLSTM has been successfully integrated into a real parking space reservation system, ThsParking, which is one of the top smart parking platforms in China. Based on 452,480real parking records in 683 days from 10 parking lots, PewLSTM yields 85.3% parking prediction accuracy, which is about 20% higher than the state-of-the-art parking behavior prediction method. The code and data can be obtained fromhttps://github.com/NingxuanFeng/PewLSTM.

Over the last few decades, there has been a focus on better determining evapotranspiration and its spatial variability, but for many regions routine prediction is not generally available at a spatial resolution appropriate to the underlying surface heterogeneity. Over agricultural regions especially in China, this is particularly critical, since the spatial extent of typical field scales is not regularly resolved within the pixel resolution of satellite sensors. Clearly, for landscapes with significant variability in vegetation cover, type/architecture, and moisture, the spatial resolution of the remote sensing data is crucial for discriminating fluxes for the different land cover types and hence avoiding significant errors due to application of a land surface model to a mixed pixel containing large contrasts in surface temperature and vegetation cover. High resolution remotely sensed data is seemly to discriminate the differences over heterogeneous landscape, but we are inclined to use MODIS data (high temporal resolution, free of charge) to estimate regional ET. At such coarse spatial resolutions, the capability to monitor the impact of land cover change and disturbances on ET or to evaluate ET from different land covers is severely hampered. So, understanding the role of landscape heterogeneity and its influence on the scaling behavior of surface fluxes as observed by satellite sensors with different spatial resolutions is a critical research needed. In this study, we are inclined to use the classified data provided by high-spatial-resolution remotely sensed image to improve regional ET from coarse-spatial-resolution data.

Routine (i.e., daily to weekly) monitoring of surface energy fluxes, particularly evapotranspiration (ET), using satellite observations of radiometric surface temperature has not been feasible at high pixel resolution because of the low frequency in satellite coverage over the region of interest (i.e., approximately every 2 weeks). Cloud cover further reduces the number of useable observations of surface conditions resulting in high-resolution satellite imagery of a region typically being available once a month, which is not very useful for routine ET monitoring. Radiometric surface temperature observations at more than 1 km pixel resolution are available multiple times per day from several satellites. However, this spatial resolution is too coarse for estimating ET from individual agricultural fields or for defining variations in ET due to land cover changes. In this paper, Landsat ETM+ data in the visible and near-infrared wavelengths, are used for computing vegetation indices provide higher resolution information(60 m resolution) on vegetation cover conditions. Then the vegetation index-radiometric surface temperature relationship is exploited and utilized in a disaggregation procedure for estimating subpixel variation in surface temperature. In addition, a remote sensing-based energy balance model is used to compare output using Landsat ETM+ data versus estimated surface temperatures. From these comparisons, the utility of the surface temperature disaggregation technique appears to be most useful for estimating subpixel surface temperatures at resolutions corresponding to length scales defining agricultural field boundaries across the landscape. Finally, the simplified model to estimate daily ET is revised considering the scaling effect on heterogeneous region.

For semi-analytical or analytical models, absorptions by water constituents are key model parameters. The paper systematically analyzed the spectral characteristics of Taihu Lake water constituents absorption and inherent relationships between Taihu Lake water constituents absorption and main water quality parameters. The results showed that variations of phytoplankton absorption coefficient could be described as functions of chlorophyll a concentration [chla], and absorption spectrum of yellow substance and non-pigment suspended matters were well described by exponential functions, and ANN (Artificial Neural Networks) technology was introduced to model the relationships between ay(440), ad(440) and main water quality parameters concentration, namely DOC (Dissolvable Organic Carbon) and non-pigment suspended matter concentration. Finally based on this analysis, water components absorption models were established respectively, which provided important parameters for semi-analytical model.

Evapotranspiration (ET) plays an important role in energy exchange and hydrological cycle between the land surface and atmosphere. Over the last decades, scientists have developed many models and algorithms to have a better determination of regional ET by remote sensing data. However, at regional scale, the field measured data and remote sensing data are inconsistent with each other due to the underlying surface heterogeneity. In this study, remotely sensed ET was carried out using Landsat ETM+ and MODIS data to understand the role of landscape heterogeneity and its influence on the scaling behavior of surface fluxes. This knowledge is very useful for correction of scale error of ET estimation, and can also be used to study the spatio-temporal effect of remotely sensed ET, which is going to be carried out in the future.

Desertification, which is affecting more and more of the world, has been a major problem throughout the past decades in the north of China. Desertification and its spatio-temporal evolution information are very important to confirm stratagem and measure of desertification and implement macro-management effectively, which is helpful to promote environmental resources sustainable development. Remote sensing has been shown to be a powerful tool in monitoring and assessment desertification. This paper focuses on reviewing the theory and methodology of monitoring and assessment desertification by using high resolution remotely sensed data. On the basis of these theories and methods, the desertification in some counties at Hunshandake sandy desertification area was evaluated. The result indicates that quantitative monitoring and assessment of desertification in large area can be carried out by using high resolution remotely sensed image. It is vital to take reasonable estimation principle, establish scientific evaluation index and remotely sensed image interpreter symbol system, constitute feasible sand desertification classification project, and make use of appropriate interpreter means and technique routes. Compared with and the result from the low-resolution data, the evaluation to this method is more accurate, specific and valuable in application. The result of this method to evaluate regional desertification at Hunshandake desertification area indicates that, from 1970s, the acreage and proportion of the desertification at this area has augmented significantly, and the situation of sandy desertification is very serious.