Modern communication networks have become very complicated and highly dynamic, which makes them hard to model, predict control. In this paper, we develop a novel experience-driven approach that can learn well control network from its own experience rather than an accurate mathematical just as human learns new skill (such driving, swimming, etc). Specifically, we, for the first time, propose leverage emerging Deep Reinforcement Learning (DRL) enabling model-free in networks; present effective...

In this paper, we propose to leverage the emerging deep learning techniques for spatiotemporal modeling and prediction in cellular networks, based on big system data. First, perform a preliminary analysis dataset from China Mobile, use traffic load as an example show non-zero temporal autocorrelation spatial correlation among neighboring Base Stations (BSs), which motivate us discover both dependencies our study. Then present hybrid model prediction, includes novel autoencoder-based Long...

Mobile cloud computing (MCC) offers significant opportunities in performance enhancement and energy saving for mobile, battery-powered devices. Applications running on mobile devices may be represented by task graphs. This work investigates the problem of scheduling tasks (which belong to same or possibly different applications) MCC environment. More precisely, involves following steps: (i) determining offloaded onto cloud, (ii) mapping remaining (potentially heterogeneous) local cores...

Electrical energy is a high quality form of that can be easily converted to other forms with efficiency and, even more importantly, it used control lower grades ease. However, building cost-effective electrical storage (EES) system challenging task despite steady advances in the design and manufacturing EES elements including various battery supercapacitor technologies. As today, no single type element fulfills density, power delivery capacity, low cost per unit storage, long cycle life,...

The rapidly developing cloud computing and virtualization techniques provide mobile devices with battery energy saving opportunities by allowing them to offload computation execute applications remotely. A device should judiciously decide whether which portion of application be offloaded the cloud. In this paper, we consider a (MCC) interaction system consisting multiple facilities. We nested two stage game formulation for MCC system. first stage, each determines its service requests remote...

Cloud computing has attracted significant attention due to the increasing demand for low-cost, high performance, and energy-efficient computing. Profit maximization cloud service provider (CSP) is a key objective in large-scale, heterogeneous, multi-user environment of system. This paper addresses problem minimizing operation cost system by maximizing its energy efficiency while ensuring that user deadlines as defined Service Level Agreements are met. The workload can be modeled independent...

A hybrid electrical energy storage (HEES) system consists of multiple banks heterogeneous (EES) elements placed between a power source and some load devices providing charge retrieval functions. For an HEES to perform its desired functions 1) reducing electricity costs by storing obtained from the grid at off-peak times when price is lower, for use peak instead that must be bought then higher prices, 2) alleviating problems, such as excessive fluctuation undependable supply, which are...

Automatic decision-making approaches, such as reinforcement learning (RL), have been applied to (partially) solve the resource allocation problem adaptively in cloud computing system. However, a complete framework exhibits high dimensions state and action spaces, which prohibit usefulness of traditional RL techniques. In addition, power consumption has become one critical concerns design control systems, degrades system reliability increases cooling cost. An effective dynamic management...

Cloud computing has attracted significant attention due to the increasing demand for low-cost, high performance, and energy-efficient computing. Profit maximization cloud service provider (CSP) is a key objective in large-scale, heterogeneous, multi-user environment of system. This paper addresses problem minimizing operation cost system by maximizing its energy efficiency while ensuring that user deadlines as defined Service Level Agreements are met. The workload can be modeled independent...

Recently, Deep Convolutional Neural Networks (DCNNs) have made unprecedented progress, achieving the accuracy close to, or even better than human-level perception in various tasks. There is a timely need to map latest software DCNNs application-specific hardware, order achieve orders of magnitude improvement performance, energy efficiency and compactness. Stochastic Computing (SC), as low-cost alternative conventional binary computing paradigm, has potential enable massively parallel highly...

Intelligent fault diagnosis of mechanical equipment is crucial to ensure reliable operation. However, cloud-based methods often encounter challenges such as time delays and data loss. Therefore, edge computing-based has emerged a promising alternative. the limited hardware resources devices in Industrial Internet Things (IoT) pose significant striking balance between diagnostic capabilities operational efficiency. This paper introduces novel lightweight intelligent method, which tailored for...

Large language models (LLMs) excel across various tasks, but standard first-order (FO) fine-tuning demands considerable memory, significantly limiting real-world deployment. Recently, zeroth-order (ZO) optimization stood out as a promising memory-efficient training paradigm, avoiding backward passes and relying solely on forward for gradient estimation, making it attractive resource-constrained scenarios. However, ZO method lags far behind FO in both convergence speed accuracy. To bridge the...

The thermal management is a crucial design problem for mobile devices because it greatly affects not only the device reliability, but also leakage energy consumption. Conventional dynamic (DTM) techniques work well computer systems. However, due to limitation of physical space in devices, coupling effect between major heat generation components, such as application processor (AP) and battery, plays an important role determining temperature inside package. Due this effect, behavior one part...

Deep neural network (DNN) has emerged as a powerful machine learning technique for various artificial intelligence applications. Due to the unique advantages on speed, area, and power, specific hardware design become very attractive solution efficient deployment of DNN. However, huge resource cost multipliers makes fully-parallel implementations multiplication-intensive DNN still prohibitive in many real-time resource-constrained embedded This brief proposes area-efficient stochastic design....

With recent advancing of Internet Things (IoTs), it becomes very attractive to implement the deep convolutional neural networks (DCNNs) onto embedded/portable systems. Presently, executing software-based DCNNs requires high-performance server clusters in practice, restricting their widespread deployment on mobile devices. To overcome this issue, considerable research efforts have been conducted context developing highly-parallel and specific DCNN hardware, utilizing GPGPUs, FPGAs, ASICs....

Cloud Computing is a promising approach to handle the growing needs for computation and storage in an efficient cost-effective manner. Towards this end, characterizing workloads cloud infrastructure (e.g., data center) essential performing optimizations such as resource provisioning energy minimization. However, there huge gap between characteristics of actual they tend be bursty exhibit fractal behavior) existing optimization algorithms, which rely on simplistic assumptions about workloads....

Mobile cloud computing (MCC) offers significant opportunities in performance enhancement and energy saving mobile, battery-powered devices. An application running on a mobile device can be represented by task graph. This work investigates the problem of scheduling tasks (which belong to same or possibly different applications) an MCC environment. More precisely, involves following steps: (i) determining offloaded cloud, (ii) mapping remaining onto (potentially heterogeneous) cores device,...