KEYWORDS: Parallel computing, Distributed computing, Data processing, Control systems, Performance modeling, Data analysis, Computing systems, Analytical research
As one of the primary platforms for parallel computing, Spark plays a crucial role in enhancing the performance of largescale parallel processing of big data. In many Spark task scheduling processes, memory considerations are often overlooked, leaving the determination of the number of concurrent task threads to the users. Default parallelism parameters and those set by users for different algorithms or datasets may struggle to harness the maximum computational efficiency of the cluster. To address this issue, we first conduct a comprehensive analysis of the Spark job execution process, establish a job scheduling model, and propose a computation cost estimation for task execution stages. Subsequently, we analyze the impact of task parallelism on memory utilization and task execution efficiency, demonstrating its significant influence on various algorithms and datasets. Finally, we introduce the Memory-Aware Parallelism Setting Algorithm (MAPS), which is designed based on the model and controls task parallelism in real-time through memory sampling to achieve optimal computational efficiency. The MAPS algorithm iteratively executes across various stages of a job, optimizing scheduling strategies based on the computational environment to enhance performance. Experimental results indicate that the MAPS algorithm improves the job execution efficiency of the Spark framework, demonstrating good applicability across different types of jobs.
Semantic segmentation plays a crucial role in practical applications, such as autonomous driving and robot navigation. However, prevalent semantic segmentation networks suffer from two primary challenges: oversized networks with redundant parameters that hinder network inference speed and excessively lightweight network structures that sacrifice semantic segmentation accuracy. Therefore, it is essential to design a semantic segmentation network that strikes a balance between accuracy and inference speed. We propose the asymmetric residual bottleneck module, which incorporates dilated convolution, depth-wise separable asymmetric convolution, channel attention mechanism, and a channel shuffle unit. By utilizing these components, model parameters are effectively reduced, and inference speed is accelerated. Furthermore, a feature aggregation module is designed to integrate features from feature maps with various resolutions, thereby enhancing segmentation accuracy. Based on these advancements, an efficient and lightweight real-time semantic segmentation network called efficiently lightweight asymmetrical network (ELANet) is proposed. Experimental results of the Cityscapes and CamVid datasets demonstrate that ELANet strikes a favorable balance between speed and accuracy. Notably, without any pretrained model and postprocessing scheme, ELANet achieves an impressive mean intersection over union of 72.5% on the Cityscapes test dataset with only 0.82 million parameters, operating at an inference speed of 173.5 frames per second on a single NVIDIA GTX 3090 GPU, with a 512×1024 input image. These findings underscore ELANet’s tremendous potential for real-time applications.
The cloud operating system plays a crucial role in supporting machine learning tasks by offering powerful computing capabilities and abundant resources. This enables the implementation of various complex applications, including health monitoring and noise pollution assessment. However, this paradigm involves centralized data storage and processing on a central server, which raises significant data security concerns. To tackle these security issues, we propose a state-ofthe- art framework called FL4C^2, which amalgamates federated learning (FL) and cloud computing (CC) eliminating their inherent drawbacks. Specifically, FL4C^2 allows users to collaboratively learn a shared model by solely submitting model parameters to the server, which ensures that user data remains localized and private, thereby preventing any potential data leakage. Moreover, FL4C^2 incorporates an encryption parameter transmission mechanism, which safeguards against indirect data leakage resulting from sharing the model parameters. Simultaneously, we employ an anonymity strategy to generate verifiable anonymous identities, which enhances identity privacy protection. Extensive experimental results based on the KunPeng operating system demonstrate that FL4C^2 achieves superior performance while effectively preserving user privacy.
With the development of computer technology, people have higher and higher requirements for computer performance and reliability and security of system operation. Heterogeneous multi-core processor, as a new type of computer processor, has the ability to switch tasks between different types of cores, which provides more space and possibility for realizing the efficient operation of computer system, improving the computing power and enhancing the security of the system. In this paper, the task scheduling algorithm of heterogeneous multi-core processor will be discussed, and its relationship with virtual execution environment and security domain will be explored. This paper combines heterogeneous multi-core processor task scheduling algorithm with three-domain separation of multi-level security and trusted monitoring method to propose an effective scheme to improve the security of computer systems. The combination of security monitoring mechanisms in different fields can effectively prevent all kinds of security problems, so as to ensure the normal operation of computer system and data security.
Selective encryption algorithms have emerged as a popular technique for protecting the privacy of images during real-time transmission. For selectively encrypted images, it is necessary to evaluate their security and usability with visual security indices, and there have been a series of studies in this area. However, those proposed visual security indices (VSI) are often ineffective. We present a multi-directional structure and content-aware features-based visual security index (MCVSI) to perform an objective assessment on selectively encrypted images. Considering that selectively encrypted images prevent the main contents from being easily identified, stable local features in the images are extracted to indicate the degree of image content leakage. Meanwhile, we extract spatial structure information that closely aligns with human visual perception to indicate the level of variation in the overall image skeleton. Next, these features are subjected to similarity measurements to produce two types of similarity, content perception feature similarity and structure feature similarity. Finally, our visual security index is built by connecting all feature similarities and their corresponding visual security scores using the regression module. The experimental results and analyses indicate that the proposed MCVSI outperforms many existing mainstream VSI in terms of higher performance and stronger robustness, particularly on low and medium quality images.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.