With the progress of power energy storage technologies in capacity, cycle life and reliability, data center can optimize its utilization of energy storage battery to reduce its Total Cost of Ownership (TCO). Data center can cut the peak and fill the valley of their power consumption graphs with proper management of battery charging and discharging, while maintaining uninterruptible power system (UPS) capacity. This paper studies the control technology of data center battery charging and discharging based on Deep Reinforcement Learning (DRL). According to the electricity price and status and cycle life of batteries, the appropriate time is selected to charge and discharge batteries in order to maximize the electricity bill savings. To achieve higher benefits, the system state, charging and discharging actions, reward function and a neural network structure were designed in detail. According to the simulation results, the proposed algorithm can infer the best savings strategy in both USA and Beijing electricity price systems. Compared with the baseline algorithm, the priority experience playback Deep Q-network (DQN) can increase the energy storage savings up to 47% and 55% with the electricity prices of the USA and Beijing respectively.

Summary Convolutional neural network (CNN) inference usually runs on high‐performance graphic processing units (GPUs). Since GPU is a high power consumption unit, that makes the energy consumption increases sharply due to the deep learning tasks. The energy efficiency of CNN inference is not only related to the software and hardware configurations, but also closely related to the application requirements of inference tasks. However, it is not clear on GPUs at present. In this paper, we conduct a comprehensive study on the model‐level and layer‐level energy efficiency of popular CNN models. The results point out several opportunities for further optimization. We also analyze the parameter settings (i.e., batch size, dynamic voltage and frequency scaling) and propose a revenue model to allow an optimal trade‐off between energy efficiency and latency. Compared with the default settings, the optimal settings can improve revenue by up to 15.31 × . We obtain the following main findings: (i) GPUs do not exploit the parallelism from the model depth and small convolution kernels, resulting in low energy efficiency. (ii) Convolutional layers are the most energy‐consuming CNN layers. However, due to the cache, the power consumption of all layers is relatively balanced. (iii) The energy efficiency of TensorRT is 1.53 × than that of TensorFlow.

The task of Target-oriented Opinion Words Extraction aims to extract the corresponding opinion words for a given opinion target from the sentence. Recently, the methods based on recurrent neural networks have shown promising results for this task. However, these approaches only considered the sequential information of the sentences and ignored the syntactic structure. In this paper, we propose a novel graph convolutional network with adversarial training to extract the opinion words. We present a graph convolutional network based on dependency tree to learn the syntactic representation of the input. Besides, we train our model with the mixture of original examples and adversarial examples, which can improve the robustness of the model. We conduct experiments on four benchmarking datasets and the results illustrate that our proposed model consistently outperforms the state-of-the-art methods.

Current QoS-aware automatic service composition queries over a network of web services are often one-time in nature. After a network of web services is built, such queries are issued once, and answers are found from the scratch. The underlying assumption is that the participating web services are rather static, which means that their functional and non-functional parameters seldom change. However, such an assumption is often baseless. New services come and go, service APIs change gradually, and QoS values fluctuate in practice. Therefore, a support for efficiently adapting service composition is desired. In this paper, we propose an event driven continuous query algorithm to intelligently cope with different types of dynamic services, and thus enable evolution of service composition. Evaluation using both real QoS data and synthetic web service data shows its superior performance, compared with the state-of-the-art solution which won the performance championship of Web Service Challenge in 2009 and 2010. Moreover, we generalize a new graph problem: dynamic single-source optimal-Directed Acyclic Graphs (DAG) problem based on the above work. It is argued that API recommendation for framework evolution, as another typical application, could also be modeled and addressed efficiently using the proposed new graph approach.

With the proliferation of Web services, service engineers demand automatic service composition algorithms that not only synthesize the correct service compositions from thousands of services but also satisfy the quality requirements of users. This is known as QoS-aware automatic service composition problem. Our observation is that current research of only finding the optimal service composition result has several shortcomings. Users have to utilize the optimal one, which will make it rigid, and consequently bring about problems, such as overload of “hot services” and lack of choices for users. To cope with these problems, a top k query mechanism is introduced in this paper, and a progressive and incremental Key-Path-Based Loose (KPL) algorithm with 100 percent accuracy is proposed. Our QSynth, which won the performance championship of Web Service Challenge 2009 and 2010, is extended to support top k query based on KPL algorithm. Evaluations show that, compared to the state of the art, KPL algorithm achieves superior scalability and accuracy with respect to a large variety of composition scenarios. Moreover, we generalize a new graph problem: top k DAGs (Directed Acyclic Graphs) problem based on the above work. Applications of this new graph problem contain API recommender, supply chain, and so on. KPL algorithm illustrated in this paper can address them efficiently, too.

Current QoS-aware automatic service composition queries over a network of Web services are often one-time innature. After a network of Web services is built, such queries are issued once, and answers are found from the scratch. The underlying assumption is that the participating Web services are rather static so that their functional and non-functional parameters seldom change. However, such an assumption is often baseless. New services come and go, service APIs change gradually, and QoS values fluctuate. Therefore, a support for efficiently handling "continuous" service composition queries is desired. In this paper, we propose an event driven continuous query algorithm for QoS-aware automatic service composition problem to cope with different types of dynamic services. Moreover, we integrated this algorithm in our service composition system, QSynth. Finally, we evaluate our proposal using both real QoS data and synthetic Web service data and show the superior performance of ours, compared to the state-of-the art solution which won the performance championship of Web Service Challenge in 2009 and 2010.

While the usage of web services has increased explosively in recent years, very few studies examined the characteristics of web services using large-scale real data for a long period of time. In this paper, we present one such a large scale longitudinal analysis of publicly available web services of SOAP-based and RESTful types. For the period of roughly one year and from five different world-wide locations, we closely monitor the ups and downs of various basic properties of web services and their QoS values using a total of 825,132 real web services.

With the proliferation of Web services, service engineers demand good automatic service composition algorithms that not only synthesize the correct work plans from thousands of services but also satisfy the quality requirements of the users. Our observation is that conventional approaches suffer from serious limitations in scalability and accuracy when addressing both requirements simultaneously. We have designed and implemented a tool QSynth to use QoS objectives of service requests as the search directives. This approach effectively prunes the search space and significantly improves the accuracy of the search results. Evaluations show that, compared to the state of the art, QSynth achieves superior scalability and accuracy with respect to a large variety of composition scenarios. Our design of QSynth won the performance championship of Web Services Challenge 2009.