The Covid-19 pandemic led to an unprecedented reliance on Virtual Private Networks (VPNs) for remote work, exposing critical vulnerabilities in global cybersecurity infrastructures. As organizations rapidly transitioned to remote operations, many lacked the necessary security measures to protect their VPN systems, making them prime targets for cybercriminals. This study synthesizes findings from 106 studies (2020–2023) to analyze the evolution of VPN-targeted cyberattacks, the tactics employed by threat actors, and effective mitigation strategies. Our analysis reveals that the widespread adoption of remote work triggered a 238% surge in VPN-targeted attacks between 2020 and 2022, as adversaries exploited vulnerabilities, misconfigurations, and inadequate security policies. Both independent cybercriminals and state-sponsored actors leveraged phishing, ransomware, and advanced persistent threats (APTs) to gain unauthorized access to corporate networks. In many cases, organizations struggled with outdated VPN protocols, weak authentication mechanisms, and insufficient network segmentation, allowing attackers to infiltrate systems with minimal resistance. To address these challenges, we propose a VPN Hardening Framework incorporating strong authentication, robust encryption, secure configurations, and continuous monitoring, expected to significantly reduce breach risks and enhance VPN resilience in the post-pandemic era. Additionally, we highlight emerging cybersecurity trends, including the role of zero-trust architectures, quantum-resistant encryption, and AI-driven intrusion detection in fortifying VPN security against evolving threats.

Heterogeneous systems, consisting of CPUs and GPUs, offer the capability to address the demands of compute- and data-intensive applications. However, programming such systems is challenging, requiring knowledge of various parallel programming frameworks. This paper introduces COMPAR, a component-based parallel programming framework that enables the exposure and selection of multiple implementation variants of components at runtime. The framework leverages compiler directive-based language extensions to annotate the source code and generate the necessary glue code for the StarPU runtime system. COMPAR provides a unified view of implementation variants and allows for intelligent selection based on runtime context. Our evaluation demonstrates the effectiveness of COMPAR through benchmark applications. The proposed approach simplifies heterogeneous parallel programming and promotes code reuse while achieving optimal performance. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

The utilization of performance monitoring probes is a valuable tool for programmers to gather performance data. However, the manual insertion of these probes can result in an increase in code size, code obfuscation, and an added burden of learning different APIs associated with performance monitoring tools. To mitigate these issues, EDPM, an embedded domain-specific language, was developed to provide a higher level of abstraction for annotating regions of code that require instrumentation in C and C++ programs. This paper presents the design and implementation of EDPM and compares it to the well-known tool PAPI, in terms of required lines of code, flexibility in configuring regions, and performance overhead. The results of this study demonstrate that EDPM is a low-resolution profiling tool that offers a reduction in required lines of code and enables programmers to express various configurations of regions. Furthermore, the design of EDPM is such that its pragmas are ignored by the standard compiler, allowing for seamless integration into existing software processes without disrupting build systems or increasing the size of the executable. Additionally, the design of the EDPM pre-compiler allows for the extension of available performance counters while maintaining a high level of abstraction for programmers. Therefore, EDPM offers a promising solution to simplify and optimize performance monitoring in C and C++ programs. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

The frame rate of a game is important for both the end-user and the developer, and for a game to be considered playable, the minimum requirement of 60 FPS should be maintained. With respect to the programming design pattern, the cur- rent industry standard is to use Object-Oriented Design (OOD). While the demand for efficient game applications is increasing, applications developed using OOD are struggling to efficiently utilize the available computing resources and consequently meet the minimum frame-rate requirements. A design pattern that may be able to cope with the current and future requirements of resource-intensive game applications is the Data-Oriented Design (DOD), which focuses on the efficient CPU memory utilization. The main difference between OOD and DOD is related to the way data is organized and accessed. While DOD is able to efficiently utilize the cache memory, programming using the data- oriented design is perceived as much more complex compared to programming applications using object-oriented design.

In this paper, we will first develop a simple game applica- tion using both programming design patterns. Thereafter, we evaluate the performance of both implementations with respect to the overall execution time, CPU and memory utilization. Furthermore, we will develop the corresponding multi-threading versions to explore how the memory is utilized when multiple cores access data from shared cache memory. The results of the empirical evaluation show that the code sections that use DOD perform significantly faster than the corresponding parts implemented using OOD for both single-threaded and multi- threaded applications. The maximum observed speedup of 13.25 times demonstrates that games and applications developed using DOD can utilize the available resources more efficiently.

Heterogeneous computing systems provide high performance and energy efficiency. However, to optimally utilize such systems, solutions that distribute the work across host CPUs and accelerating devices are needed. In this paper, we present a performance and energy-aware approach that combines AI planning heuristics for parameter space exploration with a machine learning model for performance and energy evaluation to determine a near-optimal system configuration. For data-parallel applications, our approach determines a near-optimal host-device distribution of work, the number of processing units required, and the corresponding scheduling strategy. We evaluate our approach for various heterogeneous systems accelerated with GPU or the Intel Xeon Phi. The experimental results demonstrate that our approach finds a near-optimal system configuration by evaluating only about 7% of reasonable configurations. Furthermore, the performance per Joule estimation of system configurations using our machine learning model is more than 1000 × faster compared to the system evaluation by program execution.