Search for: All records

Time series similarity matrices (informally, recurrence plots or dot-plots), are useful tools for time series data mining. They can be used to guide data exploration, and various useful features can be derived from them and then fed into downstream analytics. However, time series similarity matrices suffer from very poor scalability, taxing both time and memory requirements. In this work, we introduce novel ideas that allow us to scale the largest time series similarity matrices that can be examined by several orders of magnitude. The first idea is a novel algorithm to compute the matrices in a way that removes dependency on the subsequence length. This algorithm is so fast that it allows us to now address datasets where the memory limitations begin to dominate. Our second novel contribution is a multiscale algorithm that computes an approximation of the matrix appropriate for the limitations of the user’s memory/screen-resolution, then performs a local, just-in-time recomputation of any region that the user wishes to zoom-in on. Given that this largely removes time and space barriers, human visual attention then becomes the bottleneck. We further introduce algorithms that search massive matrices with quadrillions of cells and then prioritize regions for later examination by either humans or algorithms. We will demonstrate the utility of our ideas for data exploration, segmentation, and classification in domains as diverse as astronomy, bioinformatics, entomology, and wildlife monitoring.

 

With the proliferation of low-cost sensors and the Internet of Things, the rate of producing data far exceeds the compute and storage capabilities of today’s infrastructure. Much of this data takes the form of time series, and in response, there has been increasing interest in the creation of time series archives in the last decade, along with the development and deployment of novel analysis methods to process the data. The general strategy has been to apply a plurality of similarity search mechanisms to various subsets and subsequences of time series data in order to identify repeated patterns and anomalies; however, the computational demands of these approaches renders them incompatible with today’s power-constrained embedded CPUs. To address this challenge, we present FA-LAMP, an FPGA-accelerated implementation of the Learned Approximate Matrix Profile (LAMP) algorithm, which predicts the correlation between streaming data sampled in real-time and a representative time series dataset used for training. FA-LAMP lends itself as a real-time solution for time series analysis problems such as classification. We present the implementation of FA-LAMP on both edge- and cloud-based prototypes. On the edge devices, FA-LAMP integrates accelerated computation as close as possible to IoT sensors, thereby eliminating the need to transmit and store data in the cloud for posterior analysis. On the cloud-based accelerators, FA-LAMP can execute multiple LAMP models on the same board, allowing simultaneous processing of incoming data from multiple data sources across a network. LAMP employs a Convolutional Neural Network (CNN) for prediction. This work investigates the challenges and limitations of deploying CNNs on FPGAs using the Xilinx Deep Learning Processor Unit (DPU) and the Vitis AI development environment. We expose several technical limitations of the DPU, while providing a mechanism to overcome them by attaching custom IP block accelerators to the architecture. We evaluate FA-LAMP using a low-cost Xilinx Ultra96-V2 FPGA as well as a cloud-based Xilinx Alveo U280 accelerator card and measure their performance against a prototypical LAMP deployment running on a Raspberry Pi 3, an Edge TPU, a GPU, a desktop CPU, and a server-class CPU. In the edge scenario, the Ultra96-V2 FPGA improved performance and energy consumption compared to the Raspberry Pi; in the cloud scenario, the server CPU and GPU outperformed the Alveo U280 accelerator card, while the desktop CPU achieved comparable performance; however, the Alveo card offered an order of magnitude lower energy consumption compared to the other four platforms. Our implementation is publicly available at https://github.com/aminiok1/lamp-alveo. 

Template matching has proven to be an effective method for seismic event detection, but is biased toward identifying events similar to previously known events, and thus is ineffective at discovering events with non‐matching waveforms (e.g., those dissimilar to existing catalog events). In principle, this limitation can be overcome by cross‐correlating every segment (possible template) of a seismogram with every other segment to identify all similar event pairs, but doing so has been previously considered computationally infeasible for long time series. Here we describe a method, called the ‘Matrix Profile’ (MP), a “correlate everything with everything” calculation that can be efficiently and scalably computed. The MP returns the maximum value of the correlation coefficient of every sub‐window of continuous data with every other sub‐window, as well as the best‐correlated sub‐window location. Here we show how MP methods can obtain valuable results when applied to months and years of continuous seismic data in both local and global case studies. We find that the MP can identify many new events in Parkfield, California seismicity that are not contained in existing event catalogs and that it can efficiently find clusters of similar earthquakes in global seismic data. Either used by itself, or as a starting point for subsequent template matching calculations, the MP is likely to provide a useful new tool for seismology research.

 

With the proliferation of low-cost sensors and the Internet-of-Things (IoT), the rate of producing data far exceeds the compute and storage capabilities of today’s infrastructure. Much of this data takes the form of time series, and in response, there has been increasing interest in the creation of time series archives in the last decade, along with the development and deployment of novel analysis methods to process the data. The general strategy has been to apply a plurality of similarity search mechanisms to various subsets and subsequences of time series data in order to identify repeated patterns and anomalies; however, the computational demands of these approaches renders them incompatible with today’s power-constrained embedded CPUs. To address this challenge, we present FA-LAMP, an FPGA-accelerated implementation of the Learned Approximate Matrix Profile (LAMP) algorithm, which predicts the correlation between streaming data sampled in real-time and a representative time series dataset used for training. FA-LAMP lends itself as a real-time solution for time series analysis problems such as classification and anomaly detection, among others. FA-LAMP provides a mechanism to integrate accelerated computation as close as possible to IoT sensors, thereby eliminating the need to transmit and store data in the cloud for posterior analysis. At its core, LAMP and FA-LAMP employ Convolution Neural Networks (CNNs) to perform prediction. This work investigates the challenges and limitations of deploying CNNs on FPGAs when using state-of-the-art commercially-supported frameworks built for this purpose, namely, the Xilinx Deep Learning Processor Unit (DPU) overlay and the Vitis AI development environment. This work exposes several technical limitations of the DPU, while providing a mechanism to overcome these limits by attaching our own hand-optimized IP block accelerators to the DPU overlay. We evaluate FA-LAMP using a low-cost Xilinx Ultra96-V2 FPGA, demonstrating performance and energy improvements of more than an order of magnitude compared to a prototypical LAMP deployment running on a Raspberry Pi 3. Our implementation is publicly available at https://github.com/fccm2021sub/fccm-lamp. 

Microbial interactions in harmful algal bloom (HAB) communities have been examined in marine systems, but are poorly studied in fresh waters. To investigate HAB-microbe interactions, we isolated bacteria with close associations to bloom-forming cyanobacteria, Microcystis spp., during a 2017 bloom in the western basin of Lake Erie. The genomes of five isolates ( Exiguobacterium sp. JMULE1, Enterobacter sp. JMULE2, Deinococcus sp. JMULE3, Paenibacillus sp. JMULE4, and Acidovorax sp. JMULE5.) were sequenced on a PacBio Sequel system. These genomes ranged in size from 3.1 Mbp ( Exiguobacterium sp. JMULE1) to 5.7 Mbp ( Enterobacter sp. JMULE2). The genomes were analyzed for genes relating to critical metabolic functions, including nitrogen reduction and carbon utilization. All five of the sequenced genomes contained genes that could be used in potential signaling and nutrient exchange between the bacteria and cyanobacteria such as Microcystis . Gene expression signatures of algal-derived carbon utilization for two isolates were identified in Microcystis blooms in Lake Erie and Lake Tai ( Taihu ) at low levels, suggesting these organisms are active and may have a functional role during Microcystis blooms in aggregates, but were largely missing from whole water samples. These findings build on the growing evidence that the bacterial microbiome associated with bloom-forming algae have the functional potential to contribute to nutrient exchange within bloom communities and interact with important bloom formers like Microcystis . 

Discovery and classification of motifs (repeated patterns) and discords (anomalies) in time series is fundamental to many scientific fields. These and related problems have effectively been solved for offline analysis of time series; however, these approaches are computationally intensive and do not lend themselves to streaming time series, such as those produced by IoT sensors, where the sampling rate imposes real-time constraints on computation and there is strong desire to locate computation as close as possible to the sensor. One promising solution is to use low-cost machine learning models to provide approximate answers to these problems. For example, prior work has trained models to predict the similarity of the most recently sampled window of data points to the time series used for training. This work addresses a more challenging problem, which is to predict not only the “strength” of the match, but also the relative location in the representative time series where the strongest matching subsequences occur. We evaluate our approach on two different real world datasets; we demonstrate speedups as high as about 30x compared to exact computations, with predictive accuracy as high as 87.95%, depending on the granularity of the prediction.