Fully Systolic FFT Architecture for Giga-sample Applications

High performance VLSI-based FFT architectures are key to signal processing and telecommunication systems since they meet the hard real-time constraints at low silicon area and low power compared to CPU-based solutions. In order to meet these goals, this paper presents a novel VLSI FFT architecture based on combining three consecutive radix-4 stages to result in a 64-point FFT engine. Cascading these 64-point FFT engines consequences an improved architecture design featuring certain characteristics. First, it can efficiently accommodate large input data sets in real time. It also simplifies processing requirements due to the radix-4 calculations. Finally, it reduces memory requirements and latency to one third compared to the fully unfolded radix-4 architecture. Two different implementations are utilized in order to validate the architecture efficiency: a FPGA implementation of a 4096-point FFT achieving a throughput of 4096 point/20.48 usec, and a VLSI implementation sustaining a throughput of 4096 point/3.89 usec.

In recent years, numerous satellite-based systems have been developed to monitor and study volcanic activity from space. This progress reflects the growing demand for accurate and timely monitoring to reduce volcanic risk. Observing volcanoes from a satellite perspective provides key advantages, enabling continuous data acquisition and near-real-time assessment of volcanic activity. Multispectral sensors operating across various regions of the electromagnetic spectrum can detect thermal anomalies associated with lava flows, pyroclastic flows, ash plumes, and volcanic gases. Traditional hotspot detection techniques based on fixed thresholds often miss subtle anomalies on a global scale. In contrast, advanced machine learning algorithms offer a data-driven alternative. We designed and implemented the V-STAR application (Volcanic Satellite Thermal Anomalies Recognition) on Google Earth Engine (GEE) to leverage cloud computing for processing large geospatial datasets in real time. It employs supervised machine learning, specifically Random Forests, to adapt to evolving volcanic conditions. This enhances the accuracy and responsiveness of volcanic monitoring, offering valuable insights into potential eruptive behavior. Here, we present V-STAR as a robust and accessible tool that integrates satellite data and advanced analytics. Through its intuitive interface, V-STAR provides a comprehensive visualization of key volcanic features. The resulting analyses reveal hidden patterns in thermal data, contributing to improved disaster risk reduction strategies associated with volcanic hazards.

This paper addresses the problem of big association rule mining using an evolutionary approach. The mimetic method has been successfully applied to small and medium size databases. However, when applied on larger databases, the performance of this method becomes an important issue and current algorithms have very long execution times. Modern CPU/GPU architectures are composed of many cores, which are massively threaded and provide a large amount of computing power, suitable for improving the performance of optimization techniques. The parallelization of such method on GPU architecture is thus promising to deal with very large datasets in real time. In this paper, an approach is proposed where the rule evaluation process is parallelized on GPU, while the generation of rules is performed on a multi-core CPU. Furthermore, an intelligent strategy is proposed to partition the search space of rules in several independent sub-spaces to allow multiple CPU cores to explore the search space efficiently and without performing redundant work. Experimental results reveal that the suggested approach outperforms the sequential version by up to at 600 times for large datasets. Moreover, it outperforms the-state-of-the-art high performance computing based approaches when dealing with the big WebDocs dataset.

Data mining techniques are used to extract useful patterns from a large data set. k-mean algorithm is one of the most famous partitioning clustering algorithm. But, Euclidean distance is sensitive to outliers and is suitable to only numeric values. Real time datasets have mixed attribute values, missing values and measurements are not in the standard format.The proposed algorithm extends the ability of the kmean algorithm to use a mixed simil arity measure to find the similarity between data objects for clustering mixed datasets.For imp uting missing values, correlation based data imputation is used.In addition, k-mean output depends on the initial cluster centre and local optima suffers from the number of clusters(k). In order to improve the efficiency of the k-mean algorithm, Artificial Bee Colony Optimization (ABC) based clustering algorithm is suggested. ABC is successful at exploring the search space, but endures in leveraging the search space. Collaborative search is used to amplify the search quality of bees to amplify the search quality of bees employees. To determine the number of clusters for the given data set, the Elbow method is used. In order to evaluate the outcome of the proposed algorithm, real time datasets are used. The results showed that the proposed method performs well compared to comparative algorithms.

MapReduce is a distributed programming paradigm to process large scale data set. Meanwhile, with the development of coprocessors, heterogeneous architecture is widely used for getting high performance. Therefore, it is natural to try to leverage both of them for big data processing. In this paper, we propose an optimized MapReduce framework for CPU-MIC heterogeneous Cluster, which mainly provides the following new features: First, a runtime is developed for MIC management, fault tolerance, and task scheduling. Second, we design SIMD friendly map and pipelined reduce to improve the efficiency of resources utilization. In addition, a memory management scheme is implemented for accessing $$<$$ key, value $$>$$ pairs on MIC efficiently. The experimental results show that our system is up to 2.4x and 8.1x faster than Hadoop for different applications.

We have setup a new course on the large scale data processing using clusters. It introduces the concepts and design of distributed systems. Many newly developed ideas such as Google file system and MapReduce programming framework for processing large scale data sets are introduced. Students will gain practical experience with distributed programming technologies via several small labs and one large multi-week final project. Labs and projects will be completed using Hadoop, an open-source implementation of Google's distributed file system and MapReduce programming model. We have taught this class named Mass Data Processing Technology on Large Scale Clusters for two years. This paper will describe the design, perform of the course as well as the experiences and lessons learned.

High-performance adaptive texture streaming for planetary-scale high-mobility information visualization

Gait stability in exercise is an inevitable and vexing problem in mechanics, artificial intelligence, sports, and rehabilitation medicine research. With the rapid development and popularization of science and technology, it becomes a reality for researchers to obtain large-scale human motion data sets in real time with higher efficiency. However, at present, the analysis of gait stability of moving joints is still based on image recognition technology, which is ten times less accurate and inefficient. In this paper, Vicon 3D motion capture system, dynamometer, and surface electromyography system were used to obtain the parameters of the lower limbs of the subjects. Using Anywhere modeling and simulation system, simulation experiments were carried out, and the reaction force data of lower limb joints under two environments were obtained. The gait characteristics of human gait were analyzed from the angle of internal and external adjustment mechanism. Combining one-way ANOVA and incremental occupancy rate, the adjustment process of gait stability is described comprehensively. The findings of this study can provide a theoretical basis for the research of lower limb con-assistive devices and can guide the design and development of bipedal anthropomorphic robots.

The VIS-5D system provides highly interactive visual access to five-dimensional data sets containing up to 50 million data points. VIS-5D runs on the Stardent ST-1000 and ST-2000 workstations and generates animated three-dimensional graphics from gridded data sets in real time. It provides a widget-based user interface and fast visual response which allows scientists to interactively explore their data sets. VIS-5D generates literal and intuitive depictions of data, has user controls that are data oriented rather than graphics oriented, and provides a WYSIWYG (what-you-see-is-what-you-get) response. The result is a system that enables scientists to produce and direct their own animations. >

Archaeology is a destructive science. Photographs and videos preserve some aspects of the sequence of events inherent within the archaeological excavation process, but cannot replicate the spatiality and detail of the downward progression of the digging entailed through excavation. Time lapse sequences of properly adapted and employed terrestrial laser scanning to create a temporal sequence of point clouds of the archaeological methodologies can, however, serve as an innovative step towards accurate documentation of crucial data for future archaeologists interested in the site. Over the course of the National Science Foundation Integrative Graduate Education and Research Traineeship funded Sediment Intervals and Site Deformation Processes: Exploring Time Lapse Laser Scanning Capabilities and Methodologies for Archaeology, temporal scanning was tested on the beaches of San Diego to establish a baseline capability for data capture. Subsequently, the methodologies for data collection were utilized as a part of the excavation workflow at the University of California, San Diego's Edom Lowlands Regional Archaeological Project's excavation of a Roman era section of tell Khirbat Faynan in southern Jordan. With the data collected from the excavation as the impetus for new system development, original visualization processing designed with the archaeological problems and end-goals in mind is being created at the University of California, San Diego Center of Interdisciplinary Science for Art, Architecture, and Archaeology in order to meaningful display the shifting data sets in real time and exhibit not just the archaeological data, but the archaeological process.

All applications in current trends need to use Machine Learning and Big Data in this huge data world.Machine learning is a type of artificial intelligence that allows software applications to become more accurate at predicting outcomes without being explicitly programmed to do so.Machine learning use historical data as input to predict new output values.Small Scale and even large developers cannot maintain Machine Learning Algorithms and other big data processing techniques all by themselves.To harvest, process and analyze the data we need a data processing engine.Our project is to provide them with a machine learning engine which runs on.a specified port on their servers which collects the data sets in real time.This which can be queried using a simple Query Language and analyze the database on the query.The purpose of our project is to create a Data Processing Engine for Big Data Analytics. Machine learning and Big Data plays a major role in the current huge data world. Our project is to maintain a dedicated machine learning and data processing engine, so that even small developers without the knowledge of machine learning and big data analysis can produce expected results based on Machine Learning to make their application smooth. All applications in current trends need to use Machine Learning and Big Data in this huge data world. Machine learning is a subset of artificial intelligence that allows software applications to become more accurate at predicting outcomes without being explicitly programmed to do so. Machine learning uses historical data as input to predict new output values. Small Scale and even large developers cannot maintain Machine Learning Algorithms and other big data processing techniques all by themselves. To harvest, process and analyze the data we need a data processing engine. This project is to provide them a data processing engine which runs on a specified port on their servers which collects the data sets in real time. This can be queried using a simple Query Language and analyze the data and produce respected results using ML based on the query

Visunlizing three dimensional discrete datasets has been a topic of many research projects and papers in the past decade. We discuss the issues that come up when designing a whole computer system cnpable of visualizing these datasets in real time. We explain the three way chicken and egg problem and discuss HewlettPackard’s effort at breaking it with the Voxelator API extensions to OpenGL. We enumerate what a good hardware design should accomplish, We discuss what system issues are important and show how to integrate volume visualization hardware in one of Hewlett-Packard’s graphics accelerators, the VISUALIZE-48XP. We show why the Voxelator is an efficient and well designed API by explnining how various existing hardware engines will easily tit into the Voxelator framework. CR

3 Introduction 4 Background 4 Applications 5 Systems-Level Investigations 13 Selected References and Important Web Sites 27

Power efficiency - performance relative to power - is one of the most important concerns when designing RADAR processing systems. This paper analyzes power and performance trade-offs for a typical Space Time Adaptive Processing (STAP) application. We study STAP implementations for CUDA and OpenMP on two architectures, Intel Haswell Core I7-4770TE and NVIDIA Kayla with a GK208 GPU. We analyze the power and performance of STAP's computationally intensive kernels across the two hardware testbeds. We discuss an efficient parallel implementation for the Haswell CPU architecture. We also show the impact and trade-offs of GPU optimization techniques. The GPU architecture is able to process large size data sets without increase in power requirement. The use of shared memory has a significant impact on the power requirement for the GPU. Finally, we show that a balance between the use of shared memory and main memory access leads to an improved performance in a typical STAP application.

3D rendering is a kind of application which is not only data intensive but also computation intensive and we can render different frames in parallel. For a 3D rendering system, the I/O performance and the ability to handle the failures of both hardware and software are very important. Traditional method uses OpenPBS to schedule the rendering tasks and puts the data on a NFS data server. But NFS data server can very easily be the bottleneck of the system and OpenPBS can't reschedule the failing tasks, so new method is required to solve those problems. MapReduce is a programming model serving for processing large scale data sets in a parallel manner and hadoop is an open source implementation of the MapReduce programming model. In this paper we propose a method which uses hadoop to do the 3D rendering work in parallel. The proposed method is based on hadoop-0.20.2, evaluation on our method and the traditional method shows that for a scene with 40 frames, our method can reduce the execution time more significantly than traditional method as the number of rendering nodes increases, and our method can also handle failures of both the rendering node and rendering program.