In this paper, we present Deepcache a novel Framework for content caching, which can significantly boost cache performance. Our Framework is based on powerful deep recurrent neural network models. It comprises of two main components: i) Object Characteristics Predictor, which builds upon deep LSTM Encoder-Decoder model to predict the future characteristics of an object (such as object popularity) - to the best of our knowledge, we are the first to propose LSTM Encoder-Decoder model for content caching; ii) a caching policy component, which accounts for predicted information of objects to make smart caching decisions. In our thorough experiments, we show that applying Deepcache Framework to existing cache policies, such as LRU and k-LRU, significantly boosts the number of cache hits.

In this paper, hydrostatic cyclic expansion extrusion (HCEE) is developed as a new severe plastic deformation technique for processing of the relatively longer ultrafine grained samples. Increasing the length of the processed sample, decreasing the processing load astonishingly, and increasing the hydrostatic stresses are the main advantages of HCEE. In This process lubricant surrounded around workplaces plays the main role in reducing the friction load and increase pressure hydrostatic. HCEE process was executed to commercial pure aluminum 1050 at room temperature, and microstructural evolution and the mechanical properties were examined. Microstructure evolution of this process was investigated by transmission electron microscopy (TEM) and back-scattered diffraction (EBSD). TEM and EBSD revealed an ultrafine grained microstructure after the two passes of the HCEE process. The average size of grains and subgrain decreased from 50µm in the annealed sample to 0.76µm after two passes of the process. Mechanical properties such strength and hardness improved because of the large effect strain. Yield and ultimate strength were increased from 40 MPa and 52 MPa to 109 MPa and 115 MPa just after one pass of HCEE process, also Microhardness was increased from 36 HV to 45 HV at first passes.

No abstract

No abstract

In this paper, we present DC a novel Framework for content caching, which can signicantly boost cache performance. Our Framework is based on powerful deep recurrent neural network models. It comprises of two main components: i) Object Characteristics Predictor, which builds upon deep LSTM Encoder-Decoder model to predict the future characteristics of an object (such as object popularity)-to the best of our knowledge, we are the rst to propose LSTM Encoder-Decoder model for content caching; ii) a caching policy component, which accounts for predicted information of objects to make smart caching decisions. In our thorough experiments, we show that applying DC Framework to existing cache policies, such as LRU and k-LRU, signicantly boosts the number of cache hits.