Future Generation Computer Systems-The International Journal of Escience最新文献

{"title":"Prompting Robotic Modalities (PRM): A structured architecture for centralizing language models in complex systems","authors":"Bilel Benjdira,&nbsp;Anis Koubaa,&nbsp;Anas M. Ali","doi":"10.1016/j.future.2025.107723","DOIUrl":"10.1016/j.future.2025.107723","url":null,"abstract":"<div><div>Despite significant advancements in robotics and AI, existing systems often struggle to integrate diverse modalities (e.g., image, sound, actuator data) into a unified framework, resulting in fragmented architectures that limit adaptability, scalability, and explainability. To address these gaps, this paper introduces Prompting Robotic Modalities (PRM), a novel architecture that centralizes language models for controlling and managing complex systems through natural language. In PRM, each system modality (e.g., image, sound, actuator) is handled independently by a Modality Language Model (MLM), while a central Task Modality, powered by a Large Language Model (LLM), orchestrates complex tasks using information from the MLMs. Each MLM is trained on datasets that pair modality-specific data with rich textual descriptions, enabling intuitive, language-based interaction. We validate PRM with two main contributions: (1) ROSGPT_Vision, a new open-source ROS 2 package (available at <span><span>https://github.com/bilel-bj/ROSGPT_Vision</span><svg><path></path></svg></span>) for visual modality tasks, achieving up to 66% classification accuracy in driver-focus monitoring—surpassing other tested models in its category; and (2) CarMate, a driver-distraction detection application that significantly reduces development time and cost by allowing rapid adaptation to new monitoring tasks via simple prompt adjustments. In addition, we develop a Navigation Language Model (NLM) that converts free-form human language orders into detailed ROS commands, underscoring PRM’s modality-agnostic adaptability. Experimental results demonstrate that PRM simplifies system development, outperforms baseline vision-language approaches in specialized tasks (e.g., driver monitoring), reduces complexity through prompt engineering rather than extensive coding, and enhances explainability via natural-language-based diagnostics. Hence, PRM lays a promising foundation for next-generation complex and robotic systems by integrating advanced language model capabilities at their core, making them more adaptable to new environments, cost-effective, and user-friendly.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107723"},"PeriodicalIF":6.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143166798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EPPDL: An efficient privacy-preserving distributed ledger for digital asset transfer in Web3.0","authors":"Lichuan Ma ,&nbsp;Lu Zhou ,&nbsp;Hang Huang ,&nbsp;Youyang Qu ,&nbsp;Xuefeng Liu","doi":"10.1016/j.future.2025.107735","DOIUrl":"10.1016/j.future.2025.107735","url":null,"abstract":"<div><div>Nowadays, a new generation of decentralized internet framework, coined as Web3.0, is emerging. However, due to the insufficient computing power on the user side and the know-your-customer regulatory requirements, it is unrealistic to fully achieve decentralization in Web3.0 currently. The service provider-intermediated architecture seems more practical by including federated service providers. At the same time, in order to fully stimulate users to create and share contents in the era of Web3.0, the importance of digital assets, e.g., digit tokens and cryptocurrencies, is increasing. As a result, whether digital asset transfer can be securely and efficiently accommodated determines the further development of Web3.0. Blockchain is believed to be one effective solution to guarantee the security of digital asset transfer. However, existing works either target fully decentralized scenarios or fail to settle digital asset transfers with high efficiency. Thus in this paper, a formal yet novel service provider-intermediated architecture is firstly proposed to closely align with the practical requirements of Web3.0. Then, an efficient privacy-preserving distributed ledger construction protocol, coined as EPPDL, is proposed to safeguard digital asset transfer among users registered at different service providers. Concrete security analysis proves that the proposed EPPDL is secure against different types of adversaries, while comprehensive experiments verify its efficiency and effectiveness.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"167 ","pages":"Article 107735"},"PeriodicalIF":6.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143077807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sufficiency power consideration to run a workload on renewable energy operated datacenter","authors":"Damien Landré ,&nbsp;Laurent Philippe ,&nbsp;Jean-Marc Pierson","doi":"10.1016/j.future.2025.107710","DOIUrl":"10.1016/j.future.2025.107710","url":null,"abstract":"<div><div>Datacenters are an essential part of the internet, but their continuous development requires finding sustainable solutions to limit their impact on climate change. The <span>Datazero2</span> project aims to design datacenters running solely on local renewable energy. In this paper, we tackle the problem of computing the minimum power demand to process a workload under quality of service constraint in a datacenter operated with renewable energy. To solve this problem, we propose a binary search algorithm that requires the computation of machine configurations with maximum computing power. When machines are heterogeneous, we face the problem of choosing the machines and their DVFS (Dynamic Voltage and Frequency Scaling) state. A MILP (Mixed-Integer Linear Programming), to find the optimal solution, and four heuristics that give satisfactory results in a reasonable time are proposed. simulations show that the best heuristics reach an average deviation from the optimal solution of 0.03% to 0.65%. The binary search algorithm is challenged against a real workload to assess the impact of flexibility on the quality of service.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"167 ","pages":"Article 107710"},"PeriodicalIF":6.2,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143437975","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EDP-CVSM model-based multi-keyword ranked search scheme over encrypted cloud data","authors":"Yinfu Deng ,&nbsp;Hua Dai ,&nbsp;Zhangchen Li ,&nbsp;Haiping Huang ,&nbsp;Qian Zhou ,&nbsp;Jian Xu ,&nbsp;Geng Yang","doi":"10.1016/j.future.2025.107726","DOIUrl":"10.1016/j.future.2025.107726","url":null,"abstract":"<div><div>Traditional searchable encryption schemes for clouds are generally based on the term frequency-inverse document frequency (TF-IDF) vector space model, but they ignore the high-dimensional sparse characteristic of encrypted vectors. It could lead to substantial computational cost of the inner product. If the dimensionality and sparsity of encrypted vectors can be reduced or compressed, the search processing will be accelerated. To improve the search efficiency, we propose an encrypted two-layer balance binary tree index-based multi-keyword ranked search scheme (ETMRS) to address this problem in this paper. An equal-length dictionary partition-based compressed vector space model (EDP-CVSM) is presented, which introduces the dictionary partition strategy. It effectively compresses the document and search vectors, which benefits the efficiency of relevance score computation in search processing. In addition, to further improves the search efficiency, a two-layer balance binary tree index (TBBT-index) is proposed, which adopts secure inner product and symmetric encryption to preserve the privacy. The index is able to filter out the sub-dictionaries having no search keywords in the upper layer and identify the result documents in the lower layer, which speeds up the search processing. Experimental results show a good performance of the proposed scheme in file coverage rate, search precision, rank privacy, search efficiency and space consumption.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"167 ","pages":"Article 107726"},"PeriodicalIF":6.2,"publicationDate":"2025-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143165137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An enhanced list scheduling algorithm for heterogeneous computing using an optimized Predictive Cost Matrix","authors":"Min Wang ,&nbsp;Jiawang Chen ,&nbsp;Haoyuan Wang ,&nbsp;Ziyi Gao ,&nbsp;Weihao Bian ,&nbsp;Sibo Qiao","doi":"10.1016/j.future.2025.107733","DOIUrl":"10.1016/j.future.2025.107733","url":null,"abstract":"<div><div>Effective task scheduling is essential for optimizing resource utilization and improving system performance in heterogeneous computing environments. Current algorithms face challenges, particularly their need for more focus on the computational demands of intensive tasks and their inadequate attention to load balancing during processor allocation. To solve these problems, this study introduces the Balanced Prediction Priority Task Scheduling (BPPTS) algorithm, a novel list scheduling approach to improve the scheduling efficiency of compute-heavy tasks in heterogeneous systems. The BPPTS algorithm proposes the Balanced Prediction Cost Matrix (BPCM), which comprehensively evaluates the importance of tasks by considering their average computation cost. At the same time, a computation enhancement factor is introduced in the priority sorting to optimize the scheduling of computation-intensive tasks. The goal is to improve the scheduling efficiency of computation-intensive tasks and achieve load balancing. The BPPTS algorithm has a complexity of <span><math><mrow><mi>O</mi><mrow><mo>(</mo><msup><mrow><mi>v</mi></mrow><mrow><mn>2</mn></mrow></msup><mi>p</mi><mo>)</mo></mrow></mrow></math></span>, where <span><math><mi>v</mi></math></span> represents the number of tasks, and <span><math><mi>p</mi></math></span> denotes the number of processors. Experiments demonstrate that BPPTS outperforms other algorithms in terms of maximum completion time and speedup.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107733"},"PeriodicalIF":6.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143077809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hybrid fuzzy grammar dynamic graph diffusing attention network for traffic flow prediction","authors":"Dongxue Zhang ,&nbsp;Zhao Zhang ,&nbsp;Xiaohong Jiao ,&nbsp;Yahui Zhang","doi":"10.1016/j.future.2025.107725","DOIUrl":"10.1016/j.future.2025.107725","url":null,"abstract":"<div><div>Accurate and real-time traffic flow prediction is an indispensable part of the intelligent transportation system and is essential in improving traffic planning capability. However, due to the highly nonlinear and spatiotemporal fluctuation characteristics of the large-scale traffic network data, it is a challenging issue to establish an accurate and effective prediction model. In this regard, a hybrid fuzzy grammar dynamic graph diffusing attention network is proposed for traffic flow prediction. Firstly, the network utilizes the grammar network structure composed of grammar rules to synchronously capture the interactive information of observable traffic parameters and the dynamic spatio-temporal correlation of each node. Secondly, the network utilizes an improved graph attention network for spatio-temporal node aggregation and dynamic edge information extraction, effectively mitigating over-smoothing. Finally, the network combines hidden features captured by the grammar structure with the change rate of the traffic flow through the fuzzy network to deduce the blend of hidden features of observable and unobservable information. Simulation results on three real datasets show that the proposed model outperforms existing prediction methods under traffic networks.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"167 ","pages":"Article 107725"},"PeriodicalIF":6.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143165138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TEMPORISE: Extracting semantic representations of varied input executions for silent data corruption evaluation","authors":"Junchi Ma,&nbsp;Yuzhu Ding,&nbsp;Sulei Huang,&nbsp;Zongtao Duan,&nbsp;Lei Tang","doi":"10.1016/j.future.2025.107734","DOIUrl":"10.1016/j.future.2025.107734","url":null,"abstract":"<div><div>The continuous advancement of technology has led to increasingly complex computing systems, but it has also made them more susceptible to soft errors. Among the challenges posed by soft errors, silent data corruption (SDC) stands out as a particularly insidious threat, often occurring without warning. Estimating SDC probabilities for a program is a formidable task due to the diversity of inputs it can encounter, resulting in significant variations in these probabilities. This paper introduces TEMPORISE, a novel approach designed to tackle this challenge. TEMPORISE leverages the control data flow graph and calling context tree to represent the commonalities and distinctions between different input executions. The embeddings of these graphs are learned through structured graph attention network and AttrE2vec. These embeddings are then combined and input into a regression model to calculate SDC probabilities. The experiments demonstrate that TEMPORISE excels in predicting SDC probabilities, achieving a 78.4 % reduction in mean absolute error compared to vTRIDENT, the state-of-the-art baseline model. Moreover, TEMPORISE improves the rank correlation of SDC probabilities for various inputs by 11.4 % compared to vTRIDENT, indicating its superior ability to capture the relative ordering of SDC probabilities. In terms of computational efficiency, TEMPORISE boasts an impressive 91.3 % reduction in time cost compared to the traditional fault injection approach.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107734"},"PeriodicalIF":6.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143077808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evolutionary optimization of spatially-distributed multi-sensors placement for indoor surveillance environments with security levels","authors":"Luis M. Moreno-Saavedra ,&nbsp;Vinícius G. Costa ,&nbsp;Adrián Garrido-Sáez ,&nbsp;Silvia Jiménez-Fernández ,&nbsp;J. Antonio Portilla-Figueras ,&nbsp;Sancho Salcedo-Sanz","doi":"10.1016/j.future.2025.107727","DOIUrl":"10.1016/j.future.2025.107727","url":null,"abstract":"<div><div>The surveillance multi-sensor placement is an important optimization problem that consists of positioning several sensors of different types to maximize the coverage of a determined area while minimizing the cost of the deployment. In this work, we tackle a modified version of the problem, consisting of spatially distributed multi-sensor placement for indoor surveillance. Our approach is focused on security surveillance of sensible indoor spaces, such as military installations, where distinct security levels can be considered. We propose an evolutionary algorithm to solve the problem, in which a novel special encoding (integer encoding with binary conversion) and effective initialization have been defined to improve the performance and convergence of the proposed algorithm. We also consider the probability of detection for each surveillance point, which depends on the distance to the sensor at hand, to better model real-life scenarios. We have tested the proposed evolutionary approach in different instances of the problem, varying both size and difficulty and obtained excellent results regarding the cost of sensors’ placement and convergence time of the algorithm.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107727"},"PeriodicalIF":6.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143077824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient Number Theoretic Transform accelerator on the versal platform powered by the AI Engine","authors":"Zhenshan Bao,&nbsp;Tianhao Zang,&nbsp;Yiqi Liu,&nbsp;Wenbo Zhang","doi":"10.1016/j.future.2025.107728","DOIUrl":"10.1016/j.future.2025.107728","url":null,"abstract":"<div><div>Lattice-based cryptography, essential for fully homomorphic encryption, primarily relies on the computationally intensive Number Theoretic Transform (NTT). This paper proposes an NTT accelerator based on AMD/Xilinx Versal ACAP and AI Engine (AIE), featuring data engines on Programmable Logic (PL) and compute engines on the AIE. For inter-core parallelism on the AIE array, we propose an efficient method that applies the communication avoidance strategy to meet resource constraints; for intra-core data parallelism, we explore the modular multiplication algorithm suitable for AIE’s SIMD processors, proposing optimized software to support extensive NTT parameters while ensuring efficiency. Specialized data units are also proposed to compensate the slow DDR interface, enhancing data flow and overall performance. Our design outperforms CPU-based solutions by an average of 8.30<span><math><mo>×</mo></math></span> and Tesla V100 GPU-based solutions by 1.44<span><math><mo>×</mo></math></span> to 1.89<span><math><mo>×</mo></math></span>. Compared to most FPGA-based solutions, our approach shows shorter latency, improving by an average of 2.62<span><math><mo>×</mo></math></span>, while ensuring scalability and flexibility.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107728"},"PeriodicalIF":6.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143077810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive incremental transfer learning for efficient performance modeling of big data workloads","authors":"Mariano Garralda-Barrio,&nbsp;Carlos Eiras-Franco,&nbsp;Verónica Bolón-Canedo","doi":"10.1016/j.future.2025.107730","DOIUrl":"10.1016/j.future.2025.107730","url":null,"abstract":"<div><div>The rise of data-intensive scalable computing systems, such as Apache Spark, has transformed data processing by enabling the efficient manipulation of large datasets across machine clusters. However, system configuration to optimize performance remains a challenge. This paper introduces an adaptive incremental transfer learning approach to predicting workload execution times. By integrating both unsupervised and supervised learning, we develop models that adapt incrementally to new workloads and configurations. To guide the optimal selection of relevant workloads, the model employs the coefficient of distance variation (CdV) and the coefficient of quality correlation (CqC), combined in the exploration–exploitation balance coefficient (EEBC). Comprehensive evaluations demonstrate the robustness and reliability of our model for performance modeling in Spark applications, with average improvements of up to 31% over state-of-the-art methods. This research contributes to efficient performance tuning systems by enabling transfer learning from historical workloads to new, previously unseen workloads. The full source code is openly available.</div></div>","PeriodicalId":55132,"journal":{"name":"Future Generation Computer Systems-The International Journal of Escience","volume":"166 ","pages":"Article 107730"},"PeriodicalIF":6.2,"publicationDate":"2025-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143077825","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}