Search

GPUs are commonly used to accelerate the execution of applications in domains such as deep learning. Deep learning applications are applied to an increasing variety of scenarios, with edge computing being one of them. However, edge devices present severe computing power and energy limitations. In this context, the use of remote GPU virtualization solutions is an efficient way to address these concerns. Nevertheless, the limited network bandwidth might be an issue. This limitation can be alleviated by leveraging on-the-fly compression within the communication layer of remote GPU virtualization solutions. In this way, data exchanged with the remote GPU is transparently compressed before being transmitted, thus increasing network bandwidth in practice. In this paper, we present the implementation of a parallel compression pipeline designed to be used within remote GPU virtualization solutions. A thorough performance analysis shows that network bandwidth can be increased by a factor of up to 2×. [ABSTRACT FROM AUTHOR]

Summary: Internet of Things (IoT) devices are usually low performance nodes connected by low bandwidth networks. To improve performance in such scenarios, some computations could be done at the edge of the network. However, edge devices may not have enough computing power to accelerate applications such as the popular machine learning ones. Using remote virtual graphics processing units (GPUs) can address this concern by accelerating applications leveraging a GPU installed in a remote device. However, this requires exchanging data with the remote GPU across the slow network. To address the problem with the slow network, the data to be exchanged with the remote GPU could be compressed. In this article, we explore the suitability of using data compression in the context of remote GPU virtualization frameworks in edge scenarios executing machine learning applications. We use popular machine learning applications to carry out such exploration. After characterizing the GPU data transfers of these applications, we analyze the usage of existing compression libraries for compressing those data transfers to/from the remote GPU. Our exploration shows that transferring compressed data becomes more beneficial as networks get slower, reducing transfer time by up to 10 times. Our analysis also reveals that efficient integration of compression into remote GPU virtualization frameworks is strongly required. [ABSTRACT FROM AUTHOR]

This article discusses the topic of accelerator virtualization in the journal Concurrency and Computation: Practice and Experience. Virtualization is a technique used to share computer resources, such as processors and memory, to improve resource utilization and reduce costs. The article explores the history of virtualization and its applications, including virtual machines and network file systems. It also focuses on recent research in virtualization for hardware accelerators, such as GPUs and TPUs, which can improve the performance and energy efficiency of certain applications. The article concludes by highlighting three papers that delve into specific aspects of accelerator virtualization, including IoT monitoring systems, cluster performance, and heterogeneous systems. [Extracted from the article]

Summary: The use of Graphics Processing Units (GPUs) has become a very popular way to accelerate the execution of many applications. However, GPUs are not exempt from side effects. For instance, GPUs are expensive devices which additionally consume a non‐negligible amount of energy even when they are not performing any computation. Furthermore, most applications present low GPU utilization. To address these concerns, the use of GPU virtualization has been proposed. In particular, remote GPU virtualization is a promising technology that allows applications to transparently leverage GPUs installed in any node of the cluster. In this paper, the remote GPU virtualization mechanism is comparatively analyzed across three different generations of GPUs. The first contribution of this study is an analysis about how the performance of the remote GPU virtualization technique is impacted by the underlying hardware. To that end, the Tesla K20, Tesla K40, and Tesla P100 GPUs along with FDR and EDR InfiniBand fabrics are used in the study. The analysis is performed in the context of the rCUDA middleware. It is clearly shown that the GPU virtualization middleware requires a comprehensive design of its communication layer, which should be perfectly adapted to every hardware generation in order to avoid a reduction in performance. This is precisely the second contribution of this work, ie, redesigning the rCUDA communication layer in order to improve the management of the underlying hardware. Results show that it is possible to improve bandwidth up to 29.43%, which translates into up to 4.81% average less execution time in the performance of the analyzed applications. [ABSTRACT FROM AUTHOR]

Summary: Graphics processing units (GPUs) are currently used in data centers to reduce the execution time of compute‐intensive applications. However, the use of GPUs presents several side effects, such as increased acquisition costs and larger space requirements. Furthermore, GPUs require a nonnegligible amount of energy even while idle. Additionally, GPU utilization is usually low for most applications. In a similar way to the use of virtual machines, using virtual GPUs may address the concerns associated with the use of these devices. In this regard, the remote GPU virtualization mechanism could be leveraged to share the GPUs present in the computing facility among the nodes of the cluster. This would increase overall GPU utilization, thus reducing the negative impact of the increased costs mentioned before. Reducing the amount of GPUs installed in the cluster could also be possible. However, in the same way as job schedulers map GPU resources to applications, virtual GPUs should also be scheduled before job execution. Nevertheless, current job schedulers are not able to deal with virtual GPUs. In this paper, we analyze the performance attained by a cluster using the remote Compute Unified Device Architecture middleware and a modified version of the Slurm scheduler, which is now able to assign remote GPUs to jobs. Results show that cluster throughput, measured as jobs completed per time unit, is doubled at the same time that the total energy consumption is reduced up to 40%. GPU utilization is also increased. [ABSTRACT FROM AUTHOR]