A Survey on System-Level Design of Neural Network Accelerators

In this paper, we present a brief survey on the system-level optimizations used for convolutional neural network (CNN) inference accelerators. For the nested loop of convolutional (CONV) layers, we discuss the effects of loop optimizations such as loop interchange, tiling, unrolling and fusion on CNN accelerators. We also explain memory optimizations that are effective with the loop optimizations. In addition, we discuss streaming architectures and single computation engine architectures that are commonly used in CNN accelerators. Optimizations for CNN models are briefly explained, followed by the recent trends and future directions of the CNN accelerator design.

Power-setpoint extremum seeking control for maximizing wind power capture of turbine and farm operation

In this paper, we propose a power-setpoint based Extremum Seeking Control (ESC) framework for model-free Region-2 controls for maximizing the power capture for turbine and farm operation, without dependency on wind measurement. As a major obstacle for retrofitting wind turbine/farm controls is that only the power setpoint is accessible, the power-setpoint based ESC framework is proposed with a back-calculation anti-windup structure. If increasing the power demand cannot further increase actual power output, the anti-windup structure automatically holds the power demand setpoint. For farm operation, the proposed method is integrated into the Delay-compensated Nested-loop ESC. The proposed method is evaluated by simulations on the SimWindFarm platform for both single-turbine and farm operation scenarios. The results demonstrate the capability of tracking the achievable optimum power for turbine and farm operation, with only reasonable increase of some loads. The proposed method promises an easy-to-implement model-free retrofitting control strategy for enhancing wind energy capture.

Efficient join algorithms for large database tables in a multi-GPU environment

Relational join processing is one of the core functionalities in database management systems. It has been demonstrated that GPUs as a general-purpose parallel computing platform is very promising in processing relational joins. However, join algorithms often need to handle very large input data, which is an issue that was not sufficiently addressed in existing work. Besides, as more and more desktop and workstation platforms support multi-GPU environment, the combined computing capability of multiple GPUs can easily achieve that of a computing cluster. It is worth exploring how join processing would benefit from the adaptation of multiple GPUs. We identify the low rate and complex patterns of data transfer among the CPU and GPUs as the main challenges in designing efficient algorithms for large table joins. To overcome such challenges, we propose three distinctive designs of multi-GPU join algorithms, namely, the nested loop, global sort-merge and hybrid joins for large table joins with different join conditions. Extensive experiments running on multiple databases and two different hardware configurations demonstrate high scalability of our algorithms over data size and significant performance boost brought by the use of multiple GPUs. Furthermore, our algorithms achieve much better performance as compared to existing join algorithms, with a speedup up to 25X and 2.8X over best known code developed for multi-core CPUs and GPUs respectively.