Search

The development of remote frequency transfer techniques, especially the appearance of optical clocks with unprecedented stability, has prompted geoscientists to study their applications in geodesy. Using remote frequency transfer technique, by frequency comparison of two optical clocks at two points P and Q connected by optical fibers, one can measure the signal's frequency shift between them, and the geopotential difference between them can be determined based on the gravity frequency shift equation. Given the orthometric height of P, the orthometric height of Q can be determined. Since the present stability of the optical clock has achieved 1 × 10−18 or better and comparing the frequency transfer via optical fiber provides stability at 10−19 level, the optical clock network enables determining the orthometric height at centimeter-level. This study provides a formulation to determine the height difference at one-centimeter level between two points on the ground based on the optical fiber frequency transfer technique.

We propose a novel approach for testing the gravitational redshift based on frequency signals transmitted between a spacecraft and a ground station. The main idea is integrating one uplink signal from ground to spacecraft and two downlink signals from spacecraft to ground. Based on the integration and certain correction models, the gravitational shift of the signals between spacecraft and ground station can be detected at high precision level. The gravitational redshift effect can be used to validate the Einstein equivalence principle (EEP), which can be tested at about 10 - 6 ∼ 10 - 8 levels in different cases for a short period (less than a month) if the stability of on-board atomic clock reach 10 - 17 /day. Compared to the scheme of Gravity Probe-A (GP-A) experiment conducted in 1976, in the new approach, any on-board signal transponders is not required, and the frequency values of the three links can be arbitrary. Since the hardware requirement is decreased, a number of spacecrafts can be candidates for a gravitational redshift experiment provided that they are able to emit two different frequency signals and receive a frequency signal from the ground.

We propose a new approach for testing the gravitational redshift based on frequency signals transmission between a spacecraft and a ground station. By a combination of one uplink signal and two downlink signals, the gravitational redshift can be tested at about 6.5×10-6 level for a GNSS satellite (the signals’ frequencies are about 1.2~1.6 GHz), and about 2.2×10-6 level for the International Space Station (the signals’ frequencies are up to 14.7 GHz), under the assumption that the clock accuracy is about 10-17 level. For better desinged cases the accuracy of gravitational redshift test can be improved to several parts in 10-8 level (the signals’ frequencies are about 8~12 GHz). Compared to the scheme of Gravity Probe-A (GP-A) experiment conducted in1976, the new approach does not require any onboard signal transponders, and the frequency values of the three links can be quite arbitrarily given. As the hardware requirement is reduced, a number of spacecrafts could be chosen as candidates for a gravitational redshift experiment. This approach could also be used in gravitational potential determination, which has prospective applications in geodesy. This study is supported by National Natural Science Foundation of China (NSFC) (grant Nos. 42030105, 41721003, 41631072, 41874023, 41804012), Space Station Project (2020)228, and Natural Science Foundation of Hubei Province(grant No. 2019CFB611).

Virtualization technologies provide solutions of cloud computing. Virtual resource scheduling is a crucial task in data centers, and the power consumption of virtual resources is a critical foundation of virtualization scheduling. Containers are the smallest unit of virtual resource scheduling and migration. Although many effective models for estimating power consumption of virtual machines (VM) have been proposed, few power estimation models of containers have been put forth. In this paper, we offer a fast-training piecewise regression model based on decision tree to build a VM power estimation model and estimate the containers' power by treating the container as a group of processes on the VM. In our model, we characterize the nonlinear relationship between power and features and realize the effective estimation of the containers on the VM. We evaluate the proposed model on 13 workloads in PARSEC and compare it with several models. The experimental results prove the effectiveness of our proposed model on most workloads. Moreover, the estimated power of the containers is in line with expectations.

Edge computing, as a new promising paradigm of cloud computing, has gained widespread attention from both academia and industry. Tremendous progress in many aspects of edge computing has been achieved. However, the further development of edge computing also needs to be promoted by economic benefits, while few studies focus on the pricing in the edge-cloud scenario. In this paper, we propose and study the Edge-Cloud Pricing Problem, where a service provider provides both the edge and the cloud platform for executing computational tasks. In order to achieve the optimal social welfare, we design a truthful auction mechanism for the service provider, which is an FPTAS. Theoretical analysis shows that our auction mechanism yields the optimal allocation that maximizes the social welfare while violating the supply constraint by at most a factor of 1+ϵ. Extensive simulations validate our analysis and show that the auction mechanism performs better than the heuristic algorithms, i.e., with ϵ=1, the auction mechanism outperforms the greedy algorithm by 30.2% in average.

For time series forecasting, the weight distribution among multivariables and the long-short-term time dependence are always very important and challenging. Traditional machine forecasting can’t automatically select the effective features of multivariable input and can’t capture the time dependence of sequences. The key to solve this problem is to capture the spatial correlations at the same time, the spatiotemporal relationships at different times and the long-term dependence of the temporal relationships between different series. In this paper, inspired by human attention mechanism including encoder-decoder model, we propose DPAST-based RNN (DPAST-RNN) for long-term time series prediction. Specifically, in the first phase we use attention mechanism to extract relevant features at each time adaptively then we use stacked LSTM units to extract hidden information of time series both from time and space dimensions. In the second phase, we use another attention mechanism to select the related hidden state in encoder to the hidden state of the decoder at the current time to make context vector which is embed into recurrent neural network in decoder. Thorough empirical studies based upon the VM-Power dataset we collected on OpenStack and the NASDAQ 100 Stock dataset demonstrate that the DPAST-RNN can outperform state-of-the-art methods for time series prediction.

The increasing numbers of the applications and requirement of cloud computing have made huge power consumption in data centers, which brings the problems of the high cost and resource waste. This problem attracts significant attention from academia and industry. A critical approach to solve this problem is constructing an intelligent energy management system for data centers. Furthermore, an efficient assessment and prediction module of power consumption in data centers is an essential part of the management system. It facilitates cloud service providers to perform workflow scheduling at the minimal cost and energy efficiency management with the requirement of QoS. Since the assessment and prediction of power consumption correlate, this paper presents a multi-granularity approach for power consumption prediction in data centers, which combines multi-task learning with the LSTM network. We first transfer a multi-granularity power prediction problem into a multi-task regression problem to assess and predict the power consumption of data center system maintenance and scheduling operations. Due to the time requirement for workflow and container scheduling, the prediction interval is 30 seconds. Then we propose an efficient long short-term memory network for the multigranularity prediction. The experimental results show our model outperforms other prediction models on the real datasets.

As the size of data centers and cloud computing continue to expand, power consumption in data centers is rapidly increasing. It has a great significance to predict and analyze power consumption in the data center because power consumption prediction can help data center operators perform workflow scheduling, manage energy efficiency, provide high quality-of-service (QoS), and meet the requirements of green energy use. The current methods are mainly divided into two scopes: the one is establishing a static relationship between power consumption and relevant components/applications, and the other one is treating power consumption as sequential temporal data. However, the first scope does not consider the dynamic fluctuation of power, and the other one ignores the characteristics of the power consumption data. To solve these issues, in this paper, we present a power consumption prediction framework called PCP-2LSTM based on the mean smoothing and long short-term memory (LSTM) network. We first build a power consumption system to collect data and analyze the stationary of the power series. Then we use the mean smoothing to remove the noise from the time series of power consumption. After data preprocessing, because the time for workflow and container scheduling is usually 30 seconds, we use a stacked LSTM model to predict 30s power consumption in the future. The experimental result indicates that our approach outperforms other baselines.

Background In an EQA scheme, peer group mean is conventionally regarded as the target after removing values exceeding the mean ± 3SD. However, this computation fails if there are some outliers. Methods The outliers in each original data set were separately deleted by traditional and robust statistical methods. The data distributions of original and trimmed data sets were separately tested by Shapiro–Wilks or Kolmogorov–Smirnov Z test. The means of peer groups were used to set IgE targets. The difference between targets set by both methods was tested by 1-sample t test. Results The original data sets were not all Gaussian distributed. After deletion of outliers using robust statistical method, the modified data sets of all tested groups were normally distributed. IgE targets established by robust and traditional methods were significantly different in some groups. The incompatible IgE targets derived all from the original and trimmed data sets which were not Gaussian distributed. Conclusions The premise for choosing a traditional method to delete outliers and using peer group means as targets was that the data sets must be Gaussian distributed. The reasons that caused the targets to be incompatible possibly were the 2 rules to be broken. Robust statistical method was used for deletion of outliers due to the distorted distribution.

Aim: This study is aimed to discuss the present situation, existing problems, and relevant measures concerning the management of clinical laboratories in China. Methods: This study was carried out through the analysis of the problems relevant to the quality management of the laboratory medicine, based on the advanced experiences of ISO and developed countries in the management of clinical laboratories. Results: Many problems still remain unsolved concerning the quality management of clinical laboratories. Conclusion: Only legalized and standardized management guarantees the quality of clinical inspection.