Long-term, high-fidelity simulation of slow-changing physical systems, such as the ocean and climate, presents a fundamental challenge in scientific computing. Traditional autoregressive machine learning models often fail in these tasks as minor errors accumulate and lead to rapid forecast degradation. To address this problem, we propose NeuralOM, a general neural operator framework designed for simulating complex, slow-changing dynamics. NeuralOM's core consists of two key innovations: (1) a Progressive Residual Correction Framework that decomposes the forecasting task into a series of fine-grained refinement steps, effectively suppressing long-term error accumulation; and (2) a Physics-Guided Graph Network whose built-in adaptive messaging mechanism explicitly models multi-scale physical interactions, such as gradient-driven flows and multiplicative couplings, thereby enhancing physical consistency while maintaining computational efficiency. We validate NeuralOM on the challenging task of global Subseasonal-to-Seasonal (S2S) ocean simulation. Extensive experiments demonstrate that NeuralOM not only surpasses state-of-the-art models in forecast accuracy and long-term stability, but also excels in simulating extreme events. For instance, at a 60-day lead time, NeuralOM achieves a 13.3% lower RMSE compared to the best-performing baseline, offering a stable, efficient, and physically-aware paradigm for data-driven scientific computing.

With the growing role of artificial intelligence in climate and weather research, efficient model training and inference are in high demand. Current models like FourCastNet and AI-GOMS depend heavily on GPUs, limiting hardware independence, especially for Chinese domestic hardware and frameworks. To address this issue, we present a framework for migrating large-scale atmospheric and oceanic models from PyTorch to MindSpore and optimizing for Chinese chips, and evaluating their performance against GPUs. The framework focuses on software-hardware adaptation, memory optimization, and parallelism. Furthermore, the model's performance is evaluated across multiple metrics, including training speed, inference speed, model accuracy, and energy efficiency, with comparisons against GPU-based implementations. Experimental results demonstrate that the migration and optimization process preserves the models' original accuracy while significantly reducing system dependencies and improving operational efficiency by leveraging Chinese chips as a viable alternative for scientific computing. This work provides valuable insights and practical guidance for leveraging Chinese domestic chips and frameworks in atmospheric and oceanic AI model development, offering a pathway toward greater technological independence.

Object detection, particularly open-vocabulary object detection, plays a crucial role in Earth sciences, such as environmental monitoring, natural disaster assessment, and land-use planning. However, existing open-vocabulary detectors, primarily trained on natural-world images, struggle to generalize to remote sensing images due to a significant data domain gap. Thus, this paper aims to advance the development of open-vocabulary object detection in remote sensing community. To achieve this, we first reformulate the task as Locate Anything on Earth (LAE) with the goal of detecting any novel concepts on Earth. We then developed the LAE-Label Engine which collects, auto-annotates, and unifies up to 10 remote sensing datasets creating the LAE-1M — the first large-scale remote sensing object detection dataset with broad category coverage. Using the LAE-1M, we further propose and train the novel LAE-DINO Model, the first open-vocabulary foundation object detector for the LAE task, featuring Dynamic Vocabulary Construction (DVC) and Visual-Guided Text Prompt Learning (VisGT) modules. DVC dynamically constructs vocabulary for each training batch, while VisGT maps visual features to semantic space, enhancing text features. We comprehensively conduct experiments on established remote sensing benchmark DIOR, DOTAv2.0, as well as our newly introduced 80-class LAE-80C benchmark. Results demonstrate the advantages of the LAE-1M dataset and the effectiveness of the LAE-DINO method.

Accurate weather forecasts are important for disaster prevention, agricultural planning, etc. Traditional numerical weather prediction (NWP) methods offer physically interpretable high-accuracy predictions but are computationally expensive and fail to fully leverage rapidly growing historical data. In recent years, deep learning models have made significant progress in weather forecasting, but challenges remain, such as balancing global and regional high-resolution forecasts, excessive smoothing in extreme event predictions, and insufficient dynamic system modeling. To address these issues, this paper proposes a global-regional nested weather forecasting framework (OneForecast) based on graph neural networks. By combining a dynamic system perspective with multi-grid theory, we construct a multi-scale graph structure and densify the target region to capture local high-frequency features. We introduce an adaptive messaging mechanism, using dynamic gating units to deeply integrate node and edge features for more accurate extreme event forecasting. For high-resolution regional forecasts, we propose a neural nested grid method to mitigate boundary information loss. Experimental results show that OneForecast performs excellently across global to regional scales and short-term to long-term forecasts, especially in extreme event predictions. Codes link: \url{https://github.com/YuanGao-YG/OneForecast}.

Large-scale pre-trained neural language model has facilitated to achieve the state-of-the-art performance on Dialogue State Tracking (DST) tasks. One of the existing works models the semantic correlation between the dialogue context and (domain, slot) pair encoded by BERT and make the prediction. Despite the effectiveness, they ignore the fact that there is no perfect semantic correspondence between (domain, slot) pair and the dialogue context. In this paper, we propose a domain-slot aware contrastive learning framework to solve this problem, which proposes three methods to bridge the semantic gap between the dialogue context and the (domain, slot) by constructing training sample pairs to fine-tune the BERT model and use it for base DST model. The experiments demonstrate that our proposed method has improved the performance of the baseline model on the MultiWOZ2.1 and MultiWOZ2.4 datasets, yielding competitive results.

Modeling the evolution of physical dynamics is a foundational problem in science and engineering, and it is regarded as the modeling of an operator mapping between infinite-dimensional functional spaces. Operator learning methods, learning the underlying infinite-dimensional operator in a high-dimensional latent space, have shown significant potential in modeling physical dynamics. However, there remains insufficient research on how to approximate an infinite-dimensional operator using a finite-dimensional parameter space. Inappropriate dimensionality representation of the underlying operator leads to convergence difficulties, decreasing generalization capability, and violating the physical consistency. To address the problem, we present Neural Manifold Operator (NMO) to learn the invariant subspace with the intrinsic dimension to parameterize infinite-dimensional underlying operators. NMO achieves state-of-the-art performance in statistical and physical metrics and gains 23.35% average improvement on three real-world scenarios and four equation-governed scenarios across a wide range of multi-disciplinary fields. Our paradigm has demonstrated universal effectiveness across various model structure implementations, including Multi-Layer Perceptron, Convolutional Neural Networks, and Transformers. Experimentally, we prove that the intrinsic dimension calculated by our paradigm is the optimal dimensional representation of the underlying operators. We release our code at https://github.com/AI4EarthLab/Neural-Manifold-Operators.

Dialogue state error correction has recently been proposed to correct wrong slot values in predicted dialogue states, thereby mitigating the error propagation problem for dialogue state tracking (DST).These approaches, though effective, are heavily intertwined with specific DST models, limiting their applicability to other DST models.To solve this problem, we propose Scalable Dialogue State Correction (Scalable-DSC), which can correct wrong slot values in the dialogue state predicted by any DST model.Specifically, we propose a Structural Template Prompt (STP) that converts predicted dialogue state from any DST models into a standardized natural language sequence as a part of the historical context, associates them with dialogue history information, and generates a corrected dialogue state sequence based on predefined template options.We further enhance Scalable-DSC by introducing two training strategies.The first employs a predictive state simulator to simulate the predicted dialogue states as the training data to enhance the generalization ability of the model.The second involves using the dialogue state predicted by DST as the training data, aiming at mitigating the inconsistent error type distribution between the training and inference.Experiments confirm that our model achieves state-of-the-art results on MultiWOZ 2.0-2.4 △ .

Reinforcement learning is used to tackle complex tasks with high-dimensional sensory inputs. Over the past decade, a wide range of reinforcement learning algorithms have been developed, with recent progress benefiting from deep learning for raw sensory signal representation. This raises a natural question: how well do these algorithms perform across different robotic manipulation tasks? To objectively compare algorithms, benchmarks use performance metrics. Benchmarks use objective performance metrics to offer a scientific way to compare algorithms. In this paper, we introduce RMBench, the first benchmark for robotic manipulations with high-dimensional continuous action and state spaces. We implement and evaluate reinforcement learning algorithms that take observed pixels as inputs and report their average performance and learning curves to demonstrate their performance and training stability. Our study concludes that none of the evaluated algorithms can handle all tasks well, with soft Actor-Critic outperforming most algorithms in terms of average reward and stability, and an algorithm combined with data augmentation potentially facilitating learning policies. Our code is publicly available at https://github.com/xiangyanfei212/RMBench-2022.git, including all benchmark tasks and studied algorithms.

MOTIVATION: In time-critical clinical settings, such as precision medicine, genomic data needs to be processed as fast as possible to arrive at data-informed treatment decisions in a timely fashion. While sequencing throughput has dramatically increased over the past decade, bioinformatics analysis throughput has not been able to keep up with the pace of computer hardware improvement, and consequently has now turned into the primary bottleneck. Modern computer hardware today is capable of much higher performance than current genomic informatics algorithms can typically utilize, therefore presenting opportunities for significant improvement of performance. Accessing the raw sequencing data from BAM files, e.g. is a necessary and time-consuming step in nearly all sequence analysis tools, however existing programming libraries for BAM access do not take full advantage of the parallel input/output capabilities of storage devices. RESULTS: In an effort to stimulate the development of a new generation of faster sequence analysis tools, we developed quickBAM, a software library to accelerate sequencing data access by exploiting the parallelism in commodity storage hardware currently widely available. We demonstrate that analysis software ported to quickBAM consistently outperforms their current versions, in some cases finishing an analysis in under 3 min while the original version took 1.5 h, using the same storage solution. AVAILABILITY AND IMPLEMENTATION: Open source and freely available at https://gitlab.com/yiq/quickbam/, we envision that quickBAM will enable a new generation of high-performance informatics tools, either directly boosting their performance if they are currently data-access bottlenecked, or allow data-access to keep up with further optimizations in algorithms and compute techniques.

Feature discretization algorithms of remote sensing images are often based on the assumption that a sample only belongs to a single category and cannot describe uncertainty caused by mixed pixels. Fuzzy rough models quantify uncertain information by introducing the memberships of pixels to each category. However, there are large errors in the decomposition model of mixed pixels, making the obtained memberships fuzzy. To overcome this weakness, we propose a feature discretization algorithm based on the generalized interval type-II fuzzy rough set for mixed pixels (GIT2FRSD). We use the fuzzy mean vector and the fuzzy covariance matrix to calculate the primary grades of pixels to each ground object and determine the secondary grades according to the distribution of pixels in the boundary region of the rough set. Then, we construct the fitness function using the magnitude of the reduction of the number of breakpoints and the average approximation precision of the generalized interval type-II fuzzy rough set and search for the best discrete breakpoints in all bands of the remote sensing image using an adaptive genetic algorithm. Our method further fuzzifies the abundance information, more accurately quantifying and evaluating the uncertainty caused by mixed pixels at a time complexity similar to that of the fuzzy rough model. The experimental results on GF-2 and Landsat 8 images show that compared with current mainstream discretization algorithms, our method has better search efficiency. It obtains the minimum number of discrete intervals while ensuring data consistency and achieves the highest classification accuracy.

Outlier detection can identify anomalies in large-scale data. To provide reliability and security for Internet of Things (IoT)-enabled maritime transportation systems (MTSs), in this paper we propose an outlier detection method based on the neighborhood rough residual network (NRRN). We calculate the neighborhood approximation accuracy and neighborhood conditional entropy to obtain the neighborhood combined entropy describing the discrimination ability of the condition attribute subset to the information system. We then delete the redundant attributes according to the attribute combination importance derived from the neighborhood combined entropy. The data after attribute reduction are used to train the convolutional neural network, and the residual network (ResNet50) is used to avoid the degradation of model performance caused by the increase in the number of network layers. The proposed method is compared with mainstream outlier detection algorithms on a fishing vessel operation dataset. Experiments show that the proposed method can greatly improve the accuracy of outlier detection while taking into account interpretability and computational efficiency, thereby ensuring the data integrity of IoT-enabled MTSs.

In this paper we explore the intersection of three domains: immersive virtual reality (VR), collaborative problem solving, and the development of spatial skills. We conducted a narrative review of literature between 2011 and 2021 looking for articles that included keywords associated with spatial skills, collaboration, and immersive virtual reality in 7 well known databases. Searches resulted in a thorough review of 21 articles. These articles included examples of virtual, physical, and applied spatial, collaborative tasks. Most of the tasks in the research were performance tasks that involved spatial skills of object location and spatial navigation by locating objects in a virtual world. Results of these studies also suggest specific design features that can benefit collaboration among users when engaging in spatial tasks. This study contributes to the literature by synthesizing research on collaboration and spatial skills in virtual reality and distilling suggestions for designers of VR experiences.

Terrestrial vegetation dynamics are important for climate variabilities but the understanding of how the vegetation dynamics respond to climate remains limited - not only because the tightly coupled climate-vegetation system makes it tricky to separate water-associated processes (e.g. precipitation and soil moisture) - but the sparse and uneven observations have made it difficult to quantify such links in a larger spatial view. Here, we relate the global vegetation - soil moisture feedbacks to their oscillation characteristics and interpret it in terms of plant-water functional traits from the satellite-based estimates of surface soil moisture (SSM) and normalized difference vegetation index (NDVI). We map the global vegetation - soil moisture oscillation time scales and investigate the spatial distribution across biomes. Our study gives a global quantification on vegetation-soil moisture dynamics, providing references for comparison related to water-and-plant functions with Earth system models.

A soil moisture retrieval algorithm using L-band radar-only observations is applied to soil moisture active and passive (SMAP) radar observations. This algorithm is based on a nonlinear relationship between L-band backscatter and soil moisture, and any ancillary vegetation or roughness information is not needed. This algorithm is based on three limiting cases and end-members: smooth bare soil, rough bare soil and maximum vegetation covered soil. Those parameters is estimated through a iterative process. Three months global soil moisture is retrieved using SMAP radar observations and this algorithm. The accuracy of soil moisture result is validated by the ground network insitu observations and the SMAP standard radar and radiometer product. The soil moisture has similar spatial pattern with that of SMAP standard product at 9 km resolution. In the future, we can estimate the soil moisture at 3 km with SMAP radar data only.

This paper presents an effort for overcoming the complexities of program optimizations on SW26010, the heterogeneous many-core processor that powers Sunway TaihuLight, the world top one supercomputer. The solution centers around a precise, static performance model for modern many-core processor. Through a careful design that leverages the special properties of SW26010 and an effective treatment to massive parallelism, the model achieves a high accuracy, showing less than 5% average errors in estimating program execution performance. The precise performance model opens many opportunities for analyzing and guiding code optimizations. The paper demonstrates the usefulness by revealing a series of insights on the effects of some important code optimizations on SW26010. Moreover, it demonstrates that with such a precise performance model, it is feasible to replace empirical auto-tuning with static auto-tuning for optimizing regular loops on heterogeneous many-core systems. Such a replacement speeds up the tuning process by as much as a factor of 43 while keeping the tuning quality loss below 6%.

Datacenter networks suffer unpredictable performance due to a lack of application level bandwidth guarantees. A lot of attention has been drawn to solve this problem such as how to provide bandwidth guarantees for virtualized machines (VMs), proportional bandwidth share among tenants, and high network utilization under peak traffic. However, existing solutions fail to cope with highly dynamic traffic in datacenter networks. In this paper, we propose eBA, an efficient solution to bandwidth allocation that provides end-to-end bandwidth guarantee for VMs under large numbers of short flows and massive bursty traffic in datacenters. eBA leverages a novel distributed VM-to-VM rate control algorithm based on the logistic model under the control-theoretic framework. eBA's implementation requires no changes to hardware or applications and can be deployed in standard protocol stack. The theoretical analysis and the experimental results show that eBA not only guarantees the bandwidth for VMs, but also provides fast convergence to efficiency and fairness, as well as smooth response to bursty traffic.

This paper reports our efforts on refactoring and optimizing the Community Atmosphere Model (CAM) on the Sunway TaihuLight supercomputer, which uses a many-core processor that consists of management processing elements (MPEs) and clusters of computing processing elements (CPEs). To map the large code base of CAM to the millions of cores on the Sunway system, we take OpenACC-based refactoring as the major approach, and apply source-to-source translator tools to exploit the most suitable parallelism for the CPE cluster, and to fit the intermediate variable into the limited on-chip fast buffer. For individual kernels, when comparing the original ported version using only MPEs and the refactored version using both the MPE and CPE clusters, we achieve up to 22× speedup for the compute-intensive kernels. For the 25km resolution CAM global model, we manage to scale to 24,000 MPEs, and 1,536,000 CPEs, and achieve a simulation speed of 2.81 model years per day.

Surface wave is the most energetic form of motions in the ocean and is crucially important to navigation safety and climate change. High-resolution global wave model plays a key role in accurate surface wave forecasting. However, operational forecasting systems are still not in high-resolution due to entailed high demand for large computation, as well as low parallel efficiency barrier. Here breakthroughs encompassing the design and application of irregular quasi-rectangular domain decomposition, master-slave cooperative computing workflow and pipelining scheme were applied to a global wave model, which has been used in several operational forecasting systems and earth system models. Our realistic surface wave simulations on Sunway TaihuLight Supercomputer demonstrated that our model had outstanding scalability and achieved 45.43 PFlops in ultra-high resolution of (1/100)°, using full-scale supercomputer with 10,649,600 cores. That provides a highly effective solution for accurate surface wave forecasting and climate change prediction.

High-resolution climate simulations are increasingly in demand and require tremendous computing resources. In the Community Earth SystemModel (CESM), the Parallel Ocean Model (POP) is computationally expensive for high-resolution grids (e.g., 0.1°) and is frequently the least scalable component of CESM for certain production simulations. In particular, the modified Preconditioned Conjugate Gradient (PCG), used to solve the elliptic system of equations in the barotropic mode, scales poorly at the high core counts, which is problematic for high-resolution simulations. In this work, we demonstrate that the communication costs in the barotropic solver occupy an increasing portion of the total POP execution time as core counts are increased. To mitigate this problem, we implement a preconditioned Chebyshev-type iterative method in POP (called P-CSI), which requires far fewer global reductions than PCG. We also develop an effective block preconditioner based on the Error Vector Propagation Method to attain a competitive convergence rate for P-CSI. We demonstrate that the improved scalability of P-CSI results in a 5.2x speedup of the barotropic mode in high-resolution POP on 16,875 cores, which yields a 1.7x speedup of the overall POP simulation. Further, we ensure that the new solver produces an ocean climate consistent with the original one via an ensemble-based statistical method.

One of the most essential and challenging components in climate modeling is the atmospheric model. To solve multiphysical atmospheric equations, developers have to face extremely complex stencil kernels that are costly in terms of both computing and memory resources. This article aims to accelerate the solution of global shallow water equations (SWEs), which is one of the most essential equation sets describing atmospheric dynamics. We first design a hybrid methodology that employs both the host CPU cores and the field-programmable gate array (FPGA) accelerators to work in parallel. Through a careful adjustment of the computational domains, we achieve a balanced resource utilization and a further improvement of the overall performance. By decomposing the resource-demanding SWE kernel, we manage to map the double-precision algorithm into three FPGAs. Moreover, by using fixed-point and reduced-precision floating point arithmetic, we manage to build a fully pipelined mixed-precision design on a single FPGA, which can perform 428 floating-point and 235 fixed-point operations per cycle. The mixed-precision design with four FPGAs running together can achieve a speedup of 20 over a fully optimized design on a CPU rack with two eight-core processorsand is 8 times faster than the fully optimized Kepler GPU design. As for power efficiency, the mixed-precision design with four FPGAs is 10 times more power efficient than a Tianhe-1A supercomputer node.

While GPU is becoming a compelling acceleration solution for a series of scientific applications, most existing work on climate models only achieved limited speedup. This is due to partial porting of the huge code and the memory bound inherence of these models. In this work, we design and implement a customized GPU-based acceleration of the Princeton Ocean Model (gpuPOM) based on mpiPOM, which is one of the parallel versions of the Princeton Ocean Model. Based on Nvidia's state-of-the-art GPU architectures (K20X and K40m), we rewrite the full mpiPOM model from the original Fortran version into the CUDA-C version. We present the GPU acceleration methods used in the gpuPOM, especially the techniques to ease its memory bound problem through better use of GPU's memory hierarchy. The experimental results indicate that the gpuPOM with one K40m GPU achieves from 6.3-fold to 16.7-fold speedup over different Intel multi-core CPUs and one K20X GPU achieves from 5.8-fold to 15.5-fold speedup.

Atmospheric modeling is an essential issue in the study of climate change. However, due to the complicated algorithmic and communication models, scientists and researchers are facing tough challenges in finding efficient solutions to solve the atmospheric equations. In this paper, we accelerate a solver for the three-dimensional Euler atmospheric equations through reconfigurable data flow engines. We first propose a hybrid design that achieves efficient resource allocation and data reuse. Furthermore, through algorithmic offsetting, fast memory table, and customizable-precision arithmetic, we map a complex Euler kernel into a single FPGA chip, which can perform 956 floating point operations per cycle. In a 1U-chassis, our CPU-DFE unit with 8 FPGA chips is 18.5 times faster and 8.3 times more power efficient than a multicore system based on two 12-core Intel E5-2697 (Ivy Bridge) CPUs, and is 6.2 times faster and 5.2 times more power efficient than a hybrid unit equipped with two 12-core Intel E5-2697 (Ivy Bridge) CPUs and three Intel Xeon Phi 5120d (MIC) cards.

Datacenter networks suffer unpredictable performance due to a lack of application level bandwidth guarantees. A lot of attentions have been drawn to solve this problem such as how to provide bandwidth guarantees for Virtualized Machines (VMs), proportional bandwidth share among tenants, and high network utilization under peak traffic. However, existing solutions fail to cope with highly dynamic traffic in datacenter networks. In this paper, we consider the effects of large numbers of short flows and massive bursty traffic in the datacenter, and design a novel distributed rate allocation algorithm based on the Logistic model under the control-theoretic framework. The theoretical analysis and experimental results using OpenFlow show that our algorithm not only guarantees the bandwidth for VMs, but also provides fast convergence to efficiency and fairness, and smooth response to bursty traffic.

Climate modeling data are usually multidimensional arrays of floating-point numbers. These arrays typically have two or three spatial dimensions and one temporal dimension, describing the evolvement of climate variables in a time span. With the advances of high performance computing, the volume of climate data is expanding exponentially, bringing tough challenges for climate data archiving and sharing. In this paper, we propose a lossless compression algorithm for the time-spatial climate floating-point arrays. Our compression algorithm can eliminate more data redundancy efficiently through adaptive prediction, XOR-differencing, and multi-way compression. In addition, static regions, which are very common in climate data, can be identified and compressed more efficiently. Moreover, to utilize the multi-cores on modern computers, we proposed a method to parallelize our compression algorithm. Evaluations demonstrate that single thread version of our compression method can achieve the best balance in compression ratios, deflating throughputs and inflating throughputs. And the parallel version can achieve 800 MB/s deflating throughputs and over 2600 MB/s inflating throughputs on a 16-core server.

Understanding the inherent system characteristics is crucial to the design and optimization of cloud storage system, and few studies have systematically investigated its data characteristics and access patterns. This paper presents an analysis of file system snapshot and five-month access trace of a campus cloud storage system that has been deployed on Tsinghua campus for three years. The system provides online storage and data sharing services for more than 19,000 students and 500 student groups. We report several data characteristics including file size and file type, as well as some access patterns, including read/write ratio, read-write dependency and daily traffic. We find that there are many differences between cloud storage system and traditional file systems: our cloud storage system has larger file sizes, lower read/write ratio, and smaller set of active files than those of a typical traditional file system. With a trace-driven simulation, we find that the cache efficiency can be improved by 5 times using the guidance from our observations.

The Gaussian Copula Model (GCM) plays an important role in the state-of-the-art financial analysis field for modeling the dependence of financial assets. However, the existing implementations of GCM are all computationallydemanding and time-consuming. In this paper, we propose a Dataflow Engine (DFE) design to accelerate the GCM computation. Specifically, a commonly used CPU-friendly GCM algorithm is converted into a fully-pipelined dataflow graph through four steps of optimization: recomposing the algorithm to be pipeline-friendly, removing unnecessary computation, sharing common computing results, and reducing the computing precision while maintaining the same level of accuracy for the computation results. The performance of the proposed DFE design is compared with three CPU-based implementations that are well-optimized. Experimental results show that our DFE solution not only generates fairly accurate result, but also achieves a maximum of 467x speedup over a single-thread CPU-based solution, 120x speedup over a multi-thread CPUbased solution, and 47x speedup over an MPI-based solution.

One of the most essential and challenging components in a climate system model is the atmospheric model. To solve the multi-physical atmospheric equations, developers have to face extremely complex stencil kernels. In this paper, we propose a hybrid CPU-FPGA algorithm that applies single and multiple FPGAs to compute the upwind stencil for the global shallow water equations. Through mixed-precision arithmetic, we manage to build a fully pipelined upwind stencil design on a single FPGA, which can perform 428 floating-point and 235 fixed-point operations per cycle. The CPU-FPGA algorithm using one Virtex-6 FPGA provides 100 times speedup over a 6-core CPU and 4 times speedup over a hybrid node with 12 CPU cores and a Fermi GPU card. The algorithm using four FPGAs provides 330 times speedup over a 6-core CPU; it is also 14 times faster and 9 times more power efficient than the hybrid CPU-GPU node.

Multidimensional arrays are commonly used in scientific and engineering applications. The disk layout for the multidimensional arrays will obviously affect the performance of data querying. Homogeneous Replica method are widely used to maintain the data reliability in most of the distributed storage systems and used to improve the data locality in some parallel processing systems. In this paper, we propose a novel method, that is heterogeneous replicas, to makes better use of the replica method to optimize the performance of multidimensional arrays querying. The experimental results shows that heterogeneous replicas method can significantly reduce the overhead of disk I/O for most of the queries. With three heterogeneous replicas, the performance of random generated range queries for multidimensional datasets can be improved for 30% on the average.

Replica management method has been widely used in distributed file system to improve parallelism and reliability. Traditional replica management method is synchronous replication management (SRM) which updates replicas synchronously. Because of synchronous updating, it is difficult to provide high throughput and low access latency. If the system achieves SRM in the memory hierarchy, it will take a lot of memory space. In order to improve write performance, we design a Two Layered Replica Management Method (TLRMM) based on memory and disk. It can be used for disk-based or memory-based data storage system. TLRMM maintains 3 replicas for every chunk and store one in memory. Using asynchronous update, it can improve the system's I/O performance significantly. We integrated TLRMM in Carrier which is a distributed file system designed by Tsinghua University, and use version number to ensure replica consistency. Furthermore, in order to ensure the reliability of Carrier, we design a replica recovery strategy to solve the failure of single point. We have implemented and evaluated the integrated prototype system. The experimental results show that TLRMM can deliver high performance on distributed file system. Compared with SRM, write throughput of TLRMM increased by a factor of 1.62~2.05, without read performance degradation.

Since TCP congestion control is ill-suited for high speed networks, designing a replacement for TCP has become a challenge. To address this problem, we extend the population ecology theory to design a novel congestion control algorithm. We treat the network flows as the species in nature, the throughput of the flows as the population number, and the bottleneck bandwidth as the food resources. Then we use the key idea of constructing population ecology models to develop a novel congestion control model, and implement the corresponding end-to-end transport protocol through measurement, which called Population Ecology TCP (PE-TCP). The theoretical analysis and simulation results validate that PE-TCP achieves high utilization, fast convergence, fair bandwidth allocation, and near-zero packet drops. These qualities are desirable for high speed networks.

The various link adaptation techniques employed by High Speed Downlink Packet Access (HSDPA) in the third generation (3G) networks augment the bandwidth oscillation, which is identified as one of the most important factors resulting in the throughput deterioration of Transmission Control Protocol (TCP). In this paper, we firstly explain why the bandwidth oscillation degrades the TCP performance through a special simulation experiment. Subsequently, a split connection Window Adaptation TCP Proxy is proposed to improve the TCP throughput over HSDPA networks. In this solution, the built-in attributes of HSDAP system are sufficiently utilized. In order to effectively use the precious cellular link resources, the length of the queue connected with it is intentionally kept around the reference value through adjusting the sending window size of TCP proxy based on the dynamic values of varying bandwidth. A discrete-time stochastic state space model is formulated to analyze the system stability. The validity of enhanced scheme is verified through simulation experiments. The performance of TCP proxy is compared with the standard TCP protocol. The numerical results show that our TCP proxy is able to keep the cellular link utilization over 90%, and to improve TCP throughput by 100% under most conditions.

The traditional TCP congestion control is inefficient for high speed networks and it is a challenge to design a high speed replacement for TCP. By simulating some existing high speed protocols, we find that these high speed protocols have limitations in convergence and stability. To address these problems, we apply a population ecology model to design a novel congestion control algorithm-Coupling Logistic TCP(CLTCP). It is based on bandwidth pre-assignment that is similar to XCP and MaxNet. The pre-assignment rate factor is computed in the routers based on the information of the router capacity, the aggregate incoming traffic and the queue length. Then the senders adjust the sending rate according to the pre-assignment rate factor which carries by the packet to strengthen the convergence and stability of transport protocol. The theoretical analysis and simulation results show that CLTCP provides not only fast convergence and strong stability, but also high utilization and fair bandwidth allocation regardless of round trip time.

As an application-driven network, wireless sensor network generally requires high data reliability to maintain detection and response capabilities. Although two approaches, which are retransmission and redundancy, have been proposed to enhance data reliability, the theoretical work is required to evaluate their impact on transmission reliability and energy efficiency. In this paper, we offer a comprehensive theoretical study on the packet arrival probability and average energy consumption for both approaches. Our analysis indicates that when loss probability remains low or moderate, erasure coding, a scheme based on redundancy, is more reliable and energy efficient than retransmission. However, the performance of erasure coding would largely deteriorate under high packet loss condition. We also demonstrate that its resistance capability against packet loss weakens as hop number increases. Furthermore, with the increase in redundancy, erasure coding has to sacrifice the advantage of energy efficiency for reliability.

More attention has been paid to congestion control in the emerging area of wireless sensor network (WSN). However, most research works in the past stayed at the level of the current algorithms design or modification, and seldom sought solutions on the viewpoint of architecture. In this paper, Backpressure(BP) under Active Network(AN) architecture is used to make congestion control more responsive to detect/recover congestion in WSN. We design a simple Active Backpressure mechanism to allocate bandwidth Proportional to the Size of tree (ABPS), and we present a fluid-based analytical model of ABPS using stochastic differential equations. ABPS introduces programs in each data packet that tell nodes how to react to congestion, and quickly converge to a fair and efficient rate. We demonstrate a deterministic approach to analyse the stochastic model, in which we obtain a set of ordinary differential equations from our model, and we derive the average behavior of queue length and flow throughput from the ordinary differential equations. Finally, we evaluate ABPS extensively on a 50-node wireless sensor network. Simulation results validate the effectiveness of our ABPS and match well with the theoretic analysis.