Shared resource

About: Shared resource is a research topic. Over the lifetime, 7536 publications have been published within this topic receiving 123491 citations. The topic is also known as: network share.

Systems and methods for controlling resource usage by a driver domain on behalf of a virtual machine

Abstract: A method comprises determining an amount to which resource usage by a driver domain on behalf of a given virtual machine (VM) is to be limited. The method further comprises controlling the resource usage by the driver domain on behalf of the given VM so as not to exceed the determined amount. A system comprises a VM, a driver domain, and a shared resource that is usable by the VM and by the driver domain. The system further comprises a controller for limiting usage of the shared resource by the driver domain on behalf of the VM to prevent the usage of the shared resource by the driver domain on behalf of the VM from exceeding a determined amount.

Node Participation in Ad Hoc and Peer-to-Peer Networks: A Game-Theoretic Formulation

Abstract: Ad hoc and peer-to-peer networks sometimes operate as voluntary resource sharing networks, relying on users’ willingness to spend their own resources for the common good. As the costs of such resource sharing (what we call “node participation”) outweigh the benefits perceived by the nodes, users are less likely to participate, compromising overall network goals. The contribution of this paper is to formalize some of the relevant tradeoffs as a first step toward the design of appropriate incentive structures. We formulate a game theoretic model for node participation and derive conditions that will lead to a socially desirable equilibrium. We also analyze the impact of threats posed by a rogue node in the network.

Interference Management in D2D-Enabled Heterogeneous Cellular Networks Using Matching Theory

Abstract: To support the increasing data traffic demand in wireless communications networks, a multi-tier architecture that consists of multiple small cells is adopted in traditional (macro) cellular networks. Nevertheless, due to the expansion of various wireless devices and services, there is still a need to increase the network capacity in new generation networks such as 5G. Hence, the integration of Device-to-Device (D2D) communication as a promising technology with heterogeneous networks has been proposed. This integration not only increases network capacity but also improves spectral efficiency and alleviates the traffic load of base stations. However, the co/cross-tier interferences between D2D and cellular communications caused by resource sharing is a significant challenge. In this paper, we propose a distributed resource allocation algorithm based on matching theory to minimize these interferences and optimize the network performance. It is shown in this paper that the proposed algorithm converges to a stable matching and terminates after finite iterations. Simulation results show that the proposed algorithm is able to achieve more than 90 percent of the optimum network performance with much lower overhead and complexity.

DeepPicar: A Low-cost Deep Neural Network-based Autonomous Car

Abstract: We present DeepPicar, a low-cost deep neural network based autonomous car platform. DeepPicar is a small scale replication of a real self-driving car called DAVE-2 by NVIDIA. DAVE-2 uses a deep convolutional neural network (CNN), which takes images from a front-facing camera as input and produces car steering angles as output. DeepPicar uses the same network architecture---9 layers, 27 million connections and 250K parameters---and can drive itself in real-time using a web camera and a Raspberry Pi 3 quad-core platform. Using DeepPicar, we analyze the Pi 3's computing capabilities to support end-to-end deep learning based real-time control of autonomous vehicles. We also systematically compare other contemporary embedded computing platforms using the DeepPicar's CNN-based real-time control workload. We find that all tested platforms, including the Pi 3, are capable of supporting the CNN-based real-time control, from 20 Hz up to 100 Hz, depending on hardware platform. However, we find that shared resource contention remains an important issue that must be considered in applying CNN models on shared memory based embedded computing platforms; we observe up to 11.6X execution time increase in the CNN based control loop due to shared resource contention. To protect the CNN workload, we also evaluate state-of-the-art cache partitioning and memory bandwidth throttling techniques on the Pi 3. We find that cache partitioning is ineffective, while memory bandwidth throttling is an effective solution.

Addressing shared resource contention in multicore processors via scheduling

Abstract: Contention for shared resources on multicore processors remains an unsolved problem in existing systems despite significant research efforts dedicated to this problem in the past. Previous solution...