Efficient Use of GPU Memory for Large-Scale Deep Learning Model Training

To achieve high accuracy when performing deep learning, it is necessary to use a large-scale training model. However, due to the limitations of GPU memory, it is difficult to train large-scale training models within a single GPU. NVIDIA introduced a technology called CUDA Unified Memory with CUDA 6 to overcome the limitations of GPU memory by virtually combining GPU memory and CPU memory. In addition, in CUDA 8, memory advise options are introduced to efficiently utilize CUDA Unified Memory. In this work, we propose a newly optimized scheme based on CUDA Unified Memory to efficiently use GPU memory by applying different memory advise to each data type according to access patterns in deep learning training. We apply CUDA Unified Memory technology to PyTorch to see the performance of large-scale learning models through the expanded GPU memory. We conduct comprehensive experiments on how to efficiently utilize Unified Memory by applying memory advises when performing deep learning. As a result, when the data used for deep learning are divided into three types and a memory advise is applied to the data according to the access pattern, the deep learning execution time is reduced by 9.4% compared to the default Unified Memory.

A Reconstruction Attack Scheme on Secure Outsourced Spatial Dataset in Vehicular Ad-Hoc Networks

In the cloud-based vehicular ad-hoc network (VANET), massive vehicle information is stored on the cloud, and a large amount of data query, calculation, monitoring, and management are carried out at all times. The secure spatial query methods in VANET allow authorized users to convert the original spatial query to encrypted spatial query, which is called query token and will be processed in ciphertext mode by the service provider. Thus, the service provider learns which encrypted records are returned as the result of a query, which is defined as the access pattern. Since only the correct query results that match the query tokens are returned, the service provider can observe which encrypted data are accessed and returned to the client when a query is launched clearly, and it leads to the leakage of data access pattern. In this paper, a reconstruction attack scheme is proposed, which utilizes the access patterns in the secure query processes, and then it reconstructs the index of outsourced spatial data that are collected from the vehicles. The proposed scheme proves the security threats in the VANET. Extensive experiments on real-world datasets demonstrate that our attack scheme can achieve quite a high reconstruction rate.

TA-CLOCK: Tendency-Aware Page Replacement Policy for Hybrid Main Memory in High-Performance Embedded Systems

Recently, high-performance embedded systems have adopted phase change memory (PCM) as their main memory because PCMs have attractive advantages, such as non-volatility, byte-addressability, high density, and low power consumption. However, PCMs have disadvantages, such as limited write endurance in each cell and high write latency compared to DRAMs. Therefore, researchers have investigated methods for enhancing the limitations of PCMs. In this paper, we propose a page replacement policy called tendency-aware CLOCK (TA-CLOCK) for the hybrid main memory of embedded systems. To improve the limited write endurance of PCMs, TA-CLOCK classifies the page access tendency of the victim page through access pattern analysis and determines the migration location of the victim page. Through the classification of the page access tendency, TA-CLOCK reduces unnecessary page migrations from DRAMs to PCMs. Unnecessary migrations cause an increase in write operations in PCMs and the energy consumption of the hybrid main memory in embedded systems. Thus, our proposed policy improves the limited write endurance of PCMs and enhances the access latency of the hybrid main memory of embedded systems by classifying the page access tendency. We compared the TA-CLOCK with existing page replacement policies to evaluate its performance. In our experiments, TA-CLOCK reduced the number of write operations in PCMs by 71.5% on average, and it enhanced the energy delay product by 38.3% on average compared with other page replacement policies.