Great progress has been made in diagnosing medical diseases based on deep learning. Large-scale medical data are expected to improve deep learning performance further. It is almost impossible for a single institution to collect so much data due to the time-consuming and costly collection and labeling of medical data. Many studies have turned attention to data sharing among multiple medical institutions. However, due to different data acquiring and processing procedures, multiple institutions' medical data is characterized by distribution heterogeneity. Besides, the protection of patient privacy in medical data sharing has also been a common concern. To simultaneously address the problems of heterogeneous data distribution and privacy protection, we propose a novel multi-source source free domain adaptation. When aligning distributed heterogeneous data, our method only requires the pre-trained source models rather than direct access to the source domain data, thus protecting patients' privacy. In addition, it has the advantages of being efficient and less costly in network resources. The proposed method is evaluated on the multi-site fMRI database Autism Brain Imaging Data Exchange (ABIDE) and yields an average accuracy of 69.54%. We also analyzed its effectiveness on network resource-saving and conducted additional experiments on Camelyon17 to validate the generalization.

Convolutional Neural Networks (CNNs) have been widely deployed, while traditional cloud data-centers based applications suffer from the bandwidth and latency network demand when applying to Industrial-Internet-of-Things (IIoT) fields. It is critical to migrate the CNN inference to edge devices for efficiency and security concerns. However, it is challenging to deploy complex CNNs on resource-constraint IIoT edge devices due to a large number of parameters and intensive floating-point computations. In this paper, we propose ABM-SpConv-SIMD, an on-device inference optimization framework, aiming at accelerating the network inference by fully utilizing the low-cost and common CPU resource. ABM-SpConv SIMD first adopts a model optimizer with pruning and quantization, which produces Sparse Convolutional models. And then, the Accumulation-Before-Multiplication mechanism is proposed to reduce multiplication operations. Additionally, the SIMD instructions, which are commonly available on cost-effective edge devices, are employed to improve the performance of convolutions. We have implemented ABM-SpConv-SIMD base on the Arm Compute Library software framework and evaluated on Hikey970 and Raspberry Pi devices with two representative models AlexNet and ResNet50. The results show that the ABM SpConv-SIMD can significantly improve the performance, and achieve on average of 1.96x and 1.73x speedup respectively over the baseline implementation with negligible loss of accuracy.

Cloud gaming has been very popular recently, but providing satisfactory gaming experiences to players at a modest cost is still challenging. Colocating several games onto one server could improve server utilization. However, prior work regarding colocating games either ignores the performance interference between games or uses simple performance model to charaterize it, which may make inefficient game colocation decisions and cause QoS violations. In this paper, we address the resource allocation issues for colocating games in cloud gaming. We first propose a novel machine learning based performance model, which is able to capture the complex relationship among the performance interference, the contention features of colocated games and resource partition. Guided by the performance model, we then propose efficient and effective algorithms for two resource allocation scenarios in cloud gaming. We evaluate the proposed solutions through extensive experiments using a large number of real popular games. The results show that our performance model is able to identify whether a colocated game satisfies QoS requirement within an average error of 5%, which significantly outperforms the alternatives. Our resource allocation algorithms are able to increase the resource utilization by up to 60% compared to the state-of-the-art solutions.

Graph analytics are widely used including recommender systems, scientific computing, and data mining. Meanwhile, GPU has become the major accelerator for such applications. However, the graph size increases rapidly and often exceeds the GPU memory, incurring severe performance degradation due to frequent data transfers between the main memory and GPUs. To relieve this problem, we focus on the utilization of data in GPUs by taking advantage of the data reuse across iterations. In our studies, we deeply analyze the memory access patterns of graph applications at different granularities. We have found that the memory footprint is accessed with a roughly sequential scan without a hotspot, which infers an extremely long reuse distance. Based on our observation, we propose a novel framework, called Liberator , to exploit the data reuse within GPU memory. In Liberator , GPU memory is reserved for the data potentially accessed across iterations to avoid excessive data transfer between the main memory and GPUs. For the data not existing in GPU memory, a Merged and Aligned memory access manner is employed to improve the transmission efficiency. We also further optimize the framework by parallel processing of data in GPU memory and data in the main memory. We have implemented a prototype of the Liberator framework and conducted a series of experiments on performance evaluation. The experimental results show that Liberator can significantly reduce the data transfer overhead, which achieves an average of 2.7x speedup over a state-of-the-art approach.