GPUs have become the dominant computing units to meet the need of high performance in various computational fields. But the long operation latency causes the underutilization of on-chip computing resources, resulting in performance degradation when running parallel tasks on GPUs. A good warp scheduling strategy is an effective solution to hide latency and improve resource utilization. However, most current warp scheduling algorithms on GPUs ignore the ability of long operations to hide latency. In this paper, we propose a long-operation-first warp scheduling algorithm, LFWS, for GPU platforms. The LFWS filters warps in the ready state to a ready queue and updates the queue in time according to changes in the status of the warp. The LFWS divides the warps in the ready queue into long and short operation groups based on the type of operations in their instruction buffers, and it gives higher priority to the long-operating warp in the ready queue. This can effectively use the long operations to hide some of the latency from each other and enhance the system's ability to hide the latency. To verify the effectiveness of the LFWS, we implement the LFWS algorithm on a simulation platform GPGPU-Sim. Experiments are conducted over various CUDA applications to evaluate the performance of LFWS algorithm, compared with other five warp scheduling algorithms. The results show that the LFWS algorithm achieves an average performance improvement of 8.01% and 5.09%, respectively, over three traditional and two novel warp scheduling algorithms, effectively improving computational resource utilization on GPU.

No abstract

Vibration measurement is a frequent measurement requirement in a number of areas. Optical vibration sensors have many advantages over electrical counterparts. A common approach is to optically detect the vibration induced mechanical movement of a cantilever. Nevertheless, their practical applications are hindered by the cross-sensitivity of temperature and dynamic instability of the mechanical structure, which lead to unreliable vibration measurements. Here, we demonstrate a temperature insensitive vibration sensor that involves an enclosed suspended cantilever integrated with a readout fiber, providing in-line measurement of vibration. The cantilever is fabricated from a highly birefringent photonic crystal fiber by chemical etching and fused to a single-polarization fiber. Mechanical vibration induced periodic bending of the cantilever can significantly modify the state of polarization of the light that propagates along the photonic crystal fiber. The single-polarization fiber finally converts the state of polarization fluctuation into the change of output optical power. Therefore, the vibration could be demodulated by monitoring the output power of the proposed structure. Due to the special design of the structure, the polarization fluctuation induced by a variation of the ambient temperature can be significantly suppressed. The sensor has a linear response over the frequency range of 5 Hz to 5 kHz with a maximum signal-to-noise ratio of 60 dB and is nearly temperature independent.