VTechWorks staff will be away for the Thanksgiving holiday beginning at noon on Wednesday, November 27, through Friday, November 29. We will resume normal operations on Monday, December 2. Thank you for your patience.

Browsing by Author "Li, Bo"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Class-E Current Source Power Conversion
    Li, Bo (Virginia Tech, 2024-09-16)
    Current source is used in auxiliary power supplies, battery chargers, and LED drivers. The battery chargers are required to provide constant current within a wide output voltage range, similar to LED drivers. The load-independent (LI) Class-E inverter is a promising topology for such applications since it can realize zero-voltage switching (ZVS) within a wide load range. Class-E current source can be achieved by converting constant voltage (CV) Class-E inverter to current source with a trans-susceptance network or using parallel resonant topology. The design and analysis of LI Class-E inverters usually assume a high-Q resonant load tank so that the load current/voltage is sinusoidal. While this is the case in RF applications, it's not required in DC-DC power conversion. Besides, high-Q design leads to high inductance and increased voltage/current stress on the resonant components, increasing converter volume, loss, and cost. This work aims to provide a design guideline for the CC Class-E inverter when significant harmonics are present by reflecting the trade-off between load range and voltage stress, with the help of a modified frequency domain analysis method to eliminate the iteration existing in the time domain analysis. Output current variation and voltage stress can be automatically quantified when circuit parameters vary. Generalized load range contours are obtained to guide the circuit design. With the help of the analysis, a 10-W dual-output Class-E gate power supply is designed with optimized magnetics and reduced isolation capacitance. Compared with CC Class-E based on trans-susceptance network, the parallel resonant CC Class-E inverter has smaller part counts due to its low-order resonant network. However, the current topology suffers from limited maximum output power. In this work, a coupled-inductor based parallel resonant CC Class-E inverter is proposed with more than 2 times maximum power without increasing part counts.
  • Thumbnail Image
    MOANA: Modeling and Analyzing I/O Variability in Parallel System Experimental Design
    Cameron, Kirk W.; Anwar, Ali; Cheng, Yue; Xu, Li; Li, Bo; Ananth, Uday; Lux, Thomas; Hong, Yili; Watson, Layne T.; Butt, Ali R. (Department of Computer Science, Virginia Polytechnic Institute & State University, 2018-04-19)
    Exponential increases in complexity and scale make variability a growing threat to sustaining HPC performance at exascale. Performance variability in HPC I/O is common, acute, and formidable. We take the first step towards comprehensively studying linear and nonlinear approaches to modeling HPC I/O system variability. We create a modeling and analysis approach (MOANA) that predicts HPC I/O variability for thousands of software and hardware configurations on highly parallel shared-memory systems. Our findings indicate nonlinear approaches to I/O variability prediction are an order of magnitude more accurate than linear regression techniques. We demonstrate the use of MOANA to accurately predict the confidence intervals of unmeasured I/O system configurations for a given number of repeat runs – enabling users to quantitatively balance experiment duration with statistical confidence.
  • Thumbnail Image
    Modeling and Runtime Systems for Coordinated Power-Performance Management
    Li, Bo (Virginia Tech, 2019-01-28)
    Emergent systems in high-performance computing (HPC) expect maximal efficiency to achieve the goal of power budget under 20-40 megawatts for 1 exaflop set by the Department of Energy. To optimize efficiency, emergent systems provide multiple power-performance control techniques to throttle different system components and scale of concurrency. In this dissertation, we focus on three throttling techniques: CPU dynamic voltage and frequency scaling (DVFS), dynamic memory throttling (DMT), and dynamic concurrency throttling (DCT). We first conduct an empirical analysis of the performance and energy trade-offs of different architectures under the throttling techniques. We show the impact on performance and energy consumption on Intel x86 systems with accelerators of Intel Xeon Phi and a Nvidia general-purpose graphics processing unit (GPGPU). We show the trade-offs and potentials for improving efficiency. Furthermore, we propose a parallel performance model for coordinating DVFS, DMT, and DCT simultaneously. We present a multivariate linear regression-based approach to approximate the impact of DVFS, DMT, and DCT on performance for performance prediction. Validation using 19 HPC applications/kernels on two architectures (i.e., Intel x86 and IBM BG/Q) shows up to 7% and 17% prediction error correspondingly. Thereafter, we develop the metrics for capturing the performance impact of DVFS, DMT, and DCT. We apply the artificial neural network model to approximate the nonlinear effects on performance impact and present a runtime control strategy accordingly for power capping. Our validation using 37 HPC applications/kernels shows up to a 20% performance improvement under a given power budget compared with the Intel RAPL-based method.
  • Thumbnail Image
    The r/K selection theory and its application in biological wastewater treatment processes
    Yin, Qidong; Sun, Yuepeng; Li, Bo; Feng, Zhaolu; Wu, Guangxue (Elsevier, 2022-06-10)
    Understanding the characteristics of functional organisms is the key to managing and updating biological processes for wastewater treatment. This review, for the first time, systematically characterized two typical types of strategists in wastewater treatment ecosystems via the r/K selection theory and provided novel strategies for selectively enriching microbial community. Functional organisms involved in nitrification (e.g., Nitrosomonas and Nitrosococcus), anammox (Candidatus Brocadia), and methanogenesis (Methanosarcinaceae) are identified as r-strategists with fast growth capacities and low substrate affinities. These r-strategists can achieve high pollutant removal loading rates. On the other hand, other organisms such as Nitrosospira spp., Candidatus Kuenenia, and Methanosaetaceae, are characterized as K-strategists with slow growth rates but high substrate affinities, which can decrease the pollutant concentration to low levels. More importantly, K-strategists may play crucial roles in the biodegradation of recalcitrant organic pollutants. The food-to-microorganism ratio, mass transfer, cell size, and biomass morphology are the key factors determining the selection of r-/K-strategists. These factors can be related with operating parameters (e.g., solids and hydraulic retention time), biomass morphology (biofilm or granules), and operating modes (continuous-flow or sequencing batch), etc., to achieve the efficient acclimation of targeted r-/K-strategists. For practical applications, the concept of substrate flux was put forward to further benefit the selective enrichment of r-/K-strategists, fulfilling effective management and improvement of engineered pollution control bioprocesses. Finally, the future perspectives regarding the development of the r/K selection theory in wastewater treatment processes were discussed.