Advances in High-Performance Computing Research and Applications

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: closed (20 October 2023) | Viewed by 35002

Special Issue Editors

SAUDI ARAMCO Cybersecurity Chair, Department of Computer Science, Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia
Interests: high performance computing; real-time systems; internet-of-things; smart homes; big data
Department of Computer Science, Grande Prairie Regional College, Grand Priaire, AB T8V 4C4, Canada
Interests: datacenter optimization; high-performance computing; machine learning; smart grids; internet of things

Special Issue Information

Dear Colleagues,

The multicore era began in 2004 when instead of designing and building faster microprocessors, chip manufacturers put multiple processors on a single integrated circuit. With multicore systems, the personal computers of today can offer the power of super computers of 1990s and thus provide the opportunity to perform extensive computations. To exploit multicore architectures, many serial applications have been converted into parallel versions where it is of paramount interest to discover and exploit parallelism. Advancement in applications high performance computing (HPC) has a significant impact on data analysis, energy research, oil and gas industry, health sectors, weather forecasting, decoding the human genome, deep learning, self-driving vehicles, drug discovery and more. Today, the computing landscape is dominated by multicore architectures, and HPC has evolved tremendously in its ability to offer solutions to complex problems. However, despite the ubiquity of parallel computers, writing programs that take advantage of multiple cores remains challenging.

In this Special Issue, we invite original and unpublished submissions discussing the applications of HPC and related disciplines. This includes developing new models, image processing, oil exploration, control systems, applications of HPC in the health sector, cyber security, and adapting available tools to efficiently run on computer core systems. Computing aspects should be specifically addressed in the full papers to be aligned with the scope of this Special Issue. Additionally, selected full-length papers from “Saudi HPC/AI Conference in Medical 2022 Research” will be invited for submission. Authors from the conference submit directly to the special issue, and once accepted the conference will cover the publication charges for good papers.

Topics of interest for this Special Issue include, but are not limited to:

  • HPC infrastructure;
  • HPC algorithms;
  • HPC programming models;
  • Design and architecture of new CPUs, GPUs and other computational units for high-performance systems;
  • Algorithms for energy-efficient computation;
  • Application development for scalable architectures;
  • HPC tools and architectures;
  • High performance and machine learning;
  • Deep learning and AI-based frameworks;
  • Networking architectures;
  • Path diversity and scalable routing protocols in HPC;
  • Data locality;
  • Fault tolerance;
  • Big data analytics and social media mining for large-scale systems;
  • IoT smart buildings for post-pandemic digitized workplaces;
  • HPC scheduling;
  • Large-scale HPC application integration with fog, edge, cloud, virtualization and containerization;
  • HPC and distributed computing;
  • HPC and data visualization;
  • HPC for transportation and logistics;
  • HPC for healthcare;
  • HPC and Industry 4.0;
  • Large-scale modelling, simulation and analysis in Smart Grid;
  • Applications of HPC in health;
  • Security issues in HPC and cloud environments;
  • HPC for building cyberattack mitigation systems;
  • Cryptographic authentication systems for high-performance computing;
  • Blockchain and high-performance computing for data integrity.

Prof. Dr. Nasro Min-Allah
Dr. Ubaid Abbasi
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers (8 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

22 pages, 6719 KiB  
Article
Password Cracking with Brute Force Algorithm and Dictionary Attack Using Parallel Programming
by Ibrahim Alkhwaja, Mohammed Albugami, Ali Alkhwaja, Mohammed Alghamdi, Hussam Abahussain, Faisal Alfawaz, Abdullah Almurayh and Nasro Min-Allah
Appl. Sci. 2023, 13(10), 5979; https://doi.org/10.3390/app13105979 - 12 May 2023
Cited by 5 | Viewed by 20843
Abstract
Studying password-cracking techniques is essential in the information security discipline as it highlights the vulnerability of weak passwords and the need for stronger security measures to protect sensitive information. While both methods aim to uncover passwords, both approach the task in different ways. [...] Read more.
Studying password-cracking techniques is essential in the information security discipline as it highlights the vulnerability of weak passwords and the need for stronger security measures to protect sensitive information. While both methods aim to uncover passwords, both approach the task in different ways. A brute force algorithm generates all possible combinations of characters in a specified range and length, while the dictionary attack checks against a predefined word list. This study compares the efficiency of these methods using parallel versions of Python, C++, and Hashcat. The results show that the NVIDIA GeForce GTX 1050 Ti with CUDA is significantly faster than the Intel(R) HD Graphics 630 GPU for cracking passwords, with a speedup of 11.5× and 10.4× for passwords with and without special characters, respectively. Special characters increase password-cracking time, making the process more challenging. The results of our implementation indicate that parallel processing greatly improves the speed of password-cracking techniques. The brute force algorithm achieved a speedup of 1.9× with six cores, while the dictionary attack showed a speedup of 4.4× with eight-core static scheduling. Studying password-cracking techniques highlights the need for stronger security measures to protect sensitive information and the vulnerability of weak passwords. Full article
(This article belongs to the Special Issue Advances in High-Performance Computing Research and Applications)
Show Figures

Figure 1

12 pages, 5320 KiB  
Article
PC-Allocation: Performance Cliff-Aware Two-Level Cache Resource Allocation Scheme for Storage System
by Song Liu, Chen Zhang, Shiqiang Nie, Keqiang Duan and Weiguo Wu
Appl. Sci. 2023, 13(6), 3556; https://doi.org/10.3390/app13063556 - 10 Mar 2023
Viewed by 827
Abstract
Using the MRC (Miss Rate Curve) to guide cache capacity allocation is a common method in the storage system. However, optimal resource allocation is an NP-complete problem due to the cache performance cliff. Existing studies ignore this phenomenon or they use partitioning technology [...] Read more.
Using the MRC (Miss Rate Curve) to guide cache capacity allocation is a common method in the storage system. However, optimal resource allocation is an NP-complete problem due to the cache performance cliff. Existing studies ignore this phenomenon or they use partitioning technology to eliminate it without considering the performance potential behind the cliff. This paper delves into this potential and proposes a cliff-aware cache resource allocation algorithm based on the inherent relationship between the capacity and the hit rate. Experiments show that these requests where the latency is less than 130 µs is increased by 33.3%. The proposed method obtains a significant cost reduction in DRAM and improves the hitting ratio of the cache layer. Full article
(This article belongs to the Special Issue Advances in High-Performance Computing Research and Applications)
Show Figures

Figure 1

19 pages, 747 KiB  
Article
Improving Deep Echo State Network with Neuronal Similarity-Based Iterative Pruning Merging Algorithm
by Qingyu Shen, Hanwen Zhang and Yao Mao
Appl. Sci. 2023, 13(5), 2918; https://doi.org/10.3390/app13052918 - 24 Feb 2023
Viewed by 1160
Abstract
Recently, a layer-stacked ESN model named deep echo state network (DeepESN) has been established. As an interactional model of a recurrent neural network and deep neural network, investigations of DeepESN are of significant importance in both areas. Optimizing the structure of neural networks [...] Read more.
Recently, a layer-stacked ESN model named deep echo state network (DeepESN) has been established. As an interactional model of a recurrent neural network and deep neural network, investigations of DeepESN are of significant importance in both areas. Optimizing the structure of neural networks remains a common task in artificial neural networks, and the question of how many neurons should be used in each layer of DeepESN must be stressed. In this paper, our aim is to solve the problem of choosing the optimized size of DeepESN. Inspired by the sensitive iterative pruning algorithm, a neuronal similarity-based iterative pruning merging algorithm (NS-IPMA) is proposed to iteratively prune or merge the most similar neurons in DeepESN. Two chaotic time series prediction tasks are applied to demonstrate the effectiveness of NS-IPMA. The results show that the DeepESN pruned by NS-IPMA outperforms the unpruned DeepESN with the same network size, and NS-IPMA is a feasible and superior approach to improving the generalization performance of DeepESN. The newly proposed method has broad application prospects in real-time systems. Full article
(This article belongs to the Special Issue Advances in High-Performance Computing Research and Applications)
Show Figures

Figure 1

18 pages, 498 KiB  
Article
Quantum Image Steganography Schemes for Data Hiding: A Survey
by Nasro Min-Allah, Naya Nagy, Malak Aljabri, Mariam Alkharraa, Mashael Alqahtani, Dana Alghamdi, Razan Sabri and Rana Alshaikh
Appl. Sci. 2022, 12(20), 10294; https://doi.org/10.3390/app122010294 - 13 Oct 2022
Cited by 6 | Viewed by 3031
Abstract
Quantum steganography plays a critical role in embedding confidential data into carrier messages using quantum computing schemes. The quantum variant of steganography outperforms its classical counterpart from security, embedding efficiency and capacity, imperceptibility, and time-complexity perspectives. Considerable work has been carried out in [...] Read more.
Quantum steganography plays a critical role in embedding confidential data into carrier messages using quantum computing schemes. The quantum variant of steganography outperforms its classical counterpart from security, embedding efficiency and capacity, imperceptibility, and time-complexity perspectives. Considerable work has been carried out in the literature focusing on quantum steganography. However, a holistic view of available schemes is missing. This paper provides an overview of latest advances in the field of quantum-steganography and image-steganography schemes. Moreover, the paper includes discussion of improvements made in the aforementioned fields, a brief explanation of the methodologies used for each presented algorithm, and a comparative study of existing schemes. Full article
(This article belongs to the Special Issue Advances in High-Performance Computing Research and Applications)
Show Figures

Figure 1

11 pages, 815 KiB  
Article
Hyperparameter Tuning with High Performance Computing Machine Learning for Imbalanced Alzheimer’s Disease Data
by Fan Zhang, Melissa Petersen, Leigh Johnson, James Hall and Sid E. O’Bryant
Appl. Sci. 2022, 12(13), 6670; https://doi.org/10.3390/app12136670 - 01 Jul 2022
Cited by 8 | Viewed by 2291
Abstract
Accurate detection is still a challenge in machine learning (ML) for Alzheimer’s disease (AD). Class imbalance in imbalanced AD data is another big challenge for machine-learning algorithms working under the assumption that the data are evenly distributed within classes. Here, we present a [...] Read more.
Accurate detection is still a challenge in machine learning (ML) for Alzheimer’s disease (AD). Class imbalance in imbalanced AD data is another big challenge for machine-learning algorithms working under the assumption that the data are evenly distributed within classes. Here, we present a hyperparameter tuning workflow with high-performance computing (HPC) for imbalanced data related to prevalent mild cognitive impairment (MCI) and AD in the Health and Aging Brain Study-Health Disparities (HABS-HD) project. We applied a single-node multicore parallel mode to hyperparameter tuning of gamma, cost, and class weight using a support vector machine (SVM) model with 10 times repeated fivefold cross-validation. We executed the hyperparameter tuning workflow with R’s bigmemory, foreach, and doParallel packages on Texas Advanced Computing Center (TACC)’s Lonestar6 system. The computational time was dramatically reduced by up to 98.2% for the high-performance SVM hyperparameter tuning model, and the performance of cross-validation was also improved (the positive predictive value and the negative predictive value at base rate 12% were, respectively, 16.42% and 92.72%). Our results show that a single-node multicore parallel structure and high-performance SVM hyperparameter tuning model can deliver efficient and fast computation and achieve outstanding agility, simplicity, and productivity for imbalanced data in AD applications. Full article
(This article belongs to the Special Issue Advances in High-Performance Computing Research and Applications)
Show Figures

Figure 1

14 pages, 367 KiB  
Article
3D Tiled Code Generation for Nussinov’s Algorithm
by Włodzimierz Bielecki, Piotr Błaszyński and Marek Pałkowski
Appl. Sci. 2022, 12(12), 5898; https://doi.org/10.3390/app12125898 - 09 Jun 2022
Cited by 1 | Viewed by 1237
Abstract
Current state-of-the-art parallel codes used to calculate the maximum number of pairs for a given RNA sequence by means of Nussinov’s algorithm do not allow for achieving speedup close up to the number of the processors used for execution of those codes on [...] Read more.
Current state-of-the-art parallel codes used to calculate the maximum number of pairs for a given RNA sequence by means of Nussinov’s algorithm do not allow for achieving speedup close up to the number of the processors used for execution of those codes on multi-core computers. This is due to the fact that known codes do not make full use of and derive benefit from cache memory of such computers. There is a need to develop new approaches allowing for increasing cache exploitation in multi-core computers. One of such possibilities is increasing the dimension of tiles in generated target tiled code and assuring a similar size of generated tiles. The article presents an approach allowing us to produce 3D parallel code with tiling calculating Nussinov’s RNA folding, i.e., code with the maximal tile dimension possible for the loop nest, executing Nussinov’s algorithm. The approach guarantees that generated tiles are of a similar size. The code generated with the presented approach is characterized by increased code locality and outperforms all closely related ones examined by us. This allows us to considerably reduce execution time required for computing the maximum number of pairs of any nested structure for larger RNA sequences by means of Nussinov’s algorithm. Full article
(This article belongs to the Special Issue Advances in High-Performance Computing Research and Applications)
Show Figures

Figure 1

18 pages, 1462 KiB  
Article
Investigation of Performance and Configuration of a Selected IoT System—Middleware Deployment Benchmarking and Recommendations
by Robert Kałaska and Paweł Czarnul
Appl. Sci. 2022, 12(10), 5212; https://doi.org/10.3390/app12105212 - 21 May 2022
Cited by 2 | Viewed by 1379
Abstract
Nowadays Internet of Things is gaining more and more focus all over the world. As a concept it gives many opportunities for applications for society and it is expected that the number of software services deployed in this area will still grow fast. [...] Read more.
Nowadays Internet of Things is gaining more and more focus all over the world. As a concept it gives many opportunities for applications for society and it is expected that the number of software services deployed in this area will still grow fast. Especially important in this context are properties connected with deployment such as portability, scalability and balance between software requirements and hardware capabilities. In this article, we present results of practical tests with multiple clients representing sensors sending notifications to an IoT middleware—DeviceHive. Firstly, we investigate performance using two deployment configurations—containerized and bare-metal showing small overhead of the former under different loads by various numbers of IoT clients. We present scaling of the middleware on the server side using various numbers of cores as well as HyperThreading for all aforementioned configurations. Furthermore, we also investigated how containarization affects performance when the system is scaled with various numbers of nodes each using a predefined number of cores, considering memory usage of various configurations. The latter could be found useful when assigning cores to Docker nodes in cloud environments. Full article
(This article belongs to the Special Issue Advances in High-Performance Computing Research and Applications)
Show Figures

Figure 1

18 pages, 4296 KiB  
Article
Trusted Electronic Contract for Enabling Peer-to-Peer HPC Resource Sharing
by Kajornsak Piyoungkorn, Siriboon Chaisawat and Chalee Vorakulpipat
Appl. Sci. 2022, 12(10), 5153; https://doi.org/10.3390/app12105153 - 20 May 2022
Viewed by 1172
Abstract
With the growing need for HPC resource usage in Thailand, this study aims to foster the creation of an HPC resource sharing ecosystem based on available in-house computing infrastructure. The model of computing resource sharing based on blockchain technology is presented for bridging [...] Read more.
With the growing need for HPC resource usage in Thailand, this study aims to foster the creation of an HPC resource sharing ecosystem based on available in-house computing infrastructure. The model of computing resource sharing based on blockchain technology is presented for bridging communication between multiple clusters of HPC systems. The use of blockchain technology allows states among HPC systems to be synchronized and extends capabilities in enforcing governing rules. A smart contract was deployed on the blockchain network to enable users to request computing resources. Upon a request being made, a matching scheme performs the automatic selection of a suitable cluster based on current cluster utilization data and distance from users. Since users and clusters are anonymized from each other, a trusted payment scheme and permission access control are presented to assure both parties. As the system leverages off-chined and on-chained data exchange to carry out the operation, the secure gateway is proposed to mitigate technical difficulty from the client’s perspective and ensure information is securely flowing to and from legitimate actors. The result of this work ensures HPC service providers can maximize the utilization of their resources and monetize idle computing time, while users can access demanded resources conveniently and pay at a reasonable price. Full article
(This article belongs to the Special Issue Advances in High-Performance Computing Research and Applications)
Show Figures

Figure 1

Back to TopTop