Search

Despite increases in qubit count and connectivity in quantum annealers, a quantum speedup has yet to be observed for problems of practical significance. In order to further improve annealer performance, some researchers focus on tuning annealer parameters, such as the annealing schedule. In this work, we focus on pausing, an annealing schedule modification that has been shown to improve the probability of solving an optimization problem by orders of magnitude. However, a challenge associated with pausing is selecting an appropriate pause location, as pausing is only effective in problem-dependent regions and ineffective elsewhere. Moreover, there is little advice on how to determine where pausing is effective. Thus, pausing effectively is difficult and often inaccessible to the majority of users. To address these issues, we propose a data-driven method that leverages machine learning to predict optimal pause locations. First, we construct a dataset consisting of optimization problems and their corresponding optimal pause locations. The optimal pause locations are determined using spin-vector Monte Carlo, a method known to yield results similar to quantum annealing. Next, we train a convolutional neural network on this dataset, demonstrating its ability to distinguish between problem types and accurately predict the optimal pause location. Finally, we evaluate the model on multiple types of optimization problems. Our results show that the pause locations predicted by our method improve solution quality for all selected problem types. Additionally, our model can be pre-trained and easily distributed, making the power of pausing accessible to users unfamiliar with annealing schedule modifications.

As the first experiment at BL10U in NanoTerasu, tender X-ray ptychographic coherent diffraction imaging (PCDI) was conducted using a photon energy of 3.5 keV. The ptychographic diffraction patterns from a 200 nm thick Ta test chart and a micrometer-sized particle of sulfurized polymer were collected. Subsequently, phase images were reconstructed with resolutions of sub-20 nm and sub-50 nm, respectively. In the near future, tender X-ray PCDI with sub-10 nm resolution is anticipated to potentially revolutionize the visualization of nanoscale structures and chemical states in various functional materials composed of light elements.

The number of patients with heat illness transported by ambulance has been gradually increasing due to global warming. In intense heat waves, it is crucial to accurately estimate the number of cases with heat illness for management of medical resources. Ambient temperature is an essential factor with respect to the number of patients with heat illness, although thermophysiological response is a more relevant factor with respect to causing symptoms. In this study, we computed daily maximum core temperature increase and daily total amount of sweating in a test subject using a large-scale, integrated computational method considering the time course of actual ambient conditions as input. The correlation between the number of transported people and their thermophysiological temperature is evaluated in addition to conventional ambient temperature. With the exception of one prefecture, which features a different Köppen climate classification, the number of transported people in the remaining prefectures, with a Köppen climate classification of Cfa, are well estimated using either ambient temperature or computed core temperature increase and daily amount of sweating. For estimation using ambient temperature, an additional two parameters were needed to obtain comparable accuracy. Even using ambient temperature, the number of transported people can be estimated if the parameters are carefully chosen. This finding is practically useful for the management of ambulance allocation on hot days as well as public enlightenment.

Dedicated infrastructures are commonly used for urgent computations. However, using dedicated resources is not always affordable due to budget constraints. As a result, utilizing shared infrastructures becomes an alternative solution for urgent computations. Since the infrastructures are meant to serve many users, the urgent jobs may arrive when regular jobs are using the necessary resources. In such a case, it is necessary to preempt the regular jobs so that urgent jobs can be executed immediately. Most conventional methods for job scheduling have focused on reducing the response times and waiting times of all jobs. However, these methods can delay urgent jobs and hinder them from being completed within a stipulated deadline. Furthermore, in heterogeneous systems with coprocessors, preemption becomes more difficult because coprocessors rely on several system software functionalities provided by the host processor. In this paper, we propose a parallel job scheduling method to effectively use shared heterogeneous systems for urgent computations. Our method employs an in-memory process swapping mechanism to preempt jobs running on the coprocessor devices. The results of our simulations show that our method can achieve a significant reduction in the response time and slowdown of regular jobs without substantial delays of urgent jobs.

The mapping of tasks to processor cores, called task mapping, is crucial to achieving scalable performance on multicore processors. On modern NUMA (non-uniform memory access) systems, the memory congestion problem could degrade the performance more severely than the data locality problem because heavy congestion on shared caches and memory controllers could cause long latencies. Conventional work on task mapping mostly focuses on improving the locality of memory accesses. However, our previous work showed that on modern NUMA systems, maximizing the locality can degrade the performance due to memory congestion. In this work, we propose a task mapping method that addresses the locality and the memory congestion problems to improve the performance of parallel applications. In the proposed method, first, the spatial and temporal communication behaviors of the applications are analyzed from the time-series dataset of communications among the parallel tasks. Then, a data clustering technique is employed to detect groups of tasks that potentially cause the memory congestion. Finally, this information is used to compute the task mapping to improve the locality and reduce the memory congestion. We also provide a set of metrics to describe the communication behaviors and to evaluate if the target application can benefit from our method. The proposed method is evaluated with the NPB and PARSEC applications on a real NUMA system and a multicore simulator. A detailed analysis of the sources of performance gain is also provided. Experimental results show that our method can achieve up to a 61% performance improvement compared with the state-of-the-art locality-based method.

CFD simulations with uniform grids have been paid attention as a next-generation CFD simulation on a large-scale supercomputing system. The Building-Cube Method (BCM) is one of the next-generation CFD methods. The basic idea is to balance loads of calculations among processing elements on a supercomputing system by dividing the whole calculations into many parallel tasks with the same amount of computation. Thus, it is suitable for highly parallel computation on supercomputing systems. This paper firstly implements BCM on five supercomputing systems as an example of a next-generation CFD simulation in the upcoming postpeta-scale era. Then, by theoretical analyses and performance evaluations, this paper clarifies the requirements of future supercomputing systems for a next-generation CFD simulation. The performance evaluations show that as the number of processing elements increases, the imbalance of data exchanges among nodes becomes more serious than that of calculations even in a next-generation CFD simulation. While the calculation time can ideally be reduced according to the number of processing elements, the data transfer time becomes dominant in the total execution time. Different from the massively-parallel system architecture, the number of nodes in a system should be as small as possible to prevent the data transfer. The performance analyses also show that the memory bandwidth limits the performance of BCM and use of an on-chip memory is effective to improve the performance. A memory subsystem that achieves a higher sustained memory bandwidth is required. Therefore, a supercomputing system that consists of a small number of high-performance nodes is essential to achieve high sustained performance of the next-generation CFD in the up coming postpeta-scale era by reducing the data transfers, which becomes eventually a bottleneck in large-scale simulation.