Performance of end-to-end neural networks on a given hardware platform is a function of its compute and memory signature, which in-turn, is governed by a wide range of parameters such as topology size, primitives used, framework used, batching strategy, latency requirements, precision etc. Current benchmarking tools suffer from limitations such as a) being either too granular like DeepBench [1] (or) b) mandate a working implementation that is either framework specific or hardware-architecture specific or both (or) c) provide only high level benchmark metrics. In this paper, we present NTP (Neural Net Topology Profiler), a sophisticated benchmarking framework, to effectively identify memory and compute signature of an end-to-end topology on multiple hardware architectures, without the need for an actual implementation. NTP is tightly integrated with hardware specific benchmarking tools to enable exhaustive data collection and analysis. Using NTP, a deep learning researcher can quickly establish baselines needed to understand performance of an end-to-end neural network topology and make high level architectural decisions. Further, integration of NTP with frameworks like Tensorflow, Pytorch, Intel OpenVINO etc. allows for performance comparison along several vectors like a) Comparison of different frameworks on a given hardware b) Comparison of different hardware using a given framework c) Comparison across different heterogeneous hardware configurations for given framework etc. These capabilities empower a researcher to effortlessly make architectural decisions needed for achieving optimized performance on any hardware platform. The paper documents the architectural approach of NTP and demonstrates the capabilities of the tool by benchmarking Mozilla DeepSpeech, a popular Speech Recognition topology.Preprint. Under review.

Ahstract-Z-Source Inverter (ZSI) overcomes all the limitations of traditional Voltage and Current Source Inverters. Besides both buck and boost operation can be accomplished using ZSI. All advantages are made possible in ZSI by using an extra switching state called Shoot-Through state where the legs of inverter are shorted. A modified ZSI called Quasi Z-source inverter for renewable energy applications are investigated in this paper. In the first section, different PWM schemes like Simple Boost, Maximum Boost, Maximum Boost with Third Harmonic Injection, Maximum Constant Boost, Double Carrier PWM methods are reviewed. In next section, a novel maximum boost control method is presented based on the Space Vector Pulse-Width Modulation (SVPWM) technique for the three-phase Quasi Z-source inverter. Compared with the traditional carrier-based maximum boost control strategy, the SV method has a wider linear operation range, lesser switching loss and is easier for digital implementation. THD and voltage gains of schemes are compared. The simulation results in Matlab/Simulink as well as comparative performance of traditional VSI, ZSI and quasi ZSI have been provided and discussed, which have validated all theoretical analysis in the paper.

No abstract