Reduction In Runtime Research Articles

Dynamic performance analysis and optimization research on solar dual-stage evaporation multigeneration system

A novel solar dual-stage evaporation multigeneration system is proposed, with analysis on the impact of the parabolic trough collector area (APTC) and the volume of the heat storage tank (VHST) on performance under dynamic conditions. The particle swarm optimization- genetic algorithm and the minimum path method optimized the system’s overall performance. Results show that reducing APTC and increasing VHST can enhance energy performance within a certain range, but beyond specific thresholds, changes in APTC and VHST no longer affect it. As APTC and VHST increase, economic performance initially improves but then declines. The particle swarm optimization- genetic algorithm increased optimization efficiency and avoided local optima. When using the minimum path method, the dimensionless cumulative error was 0.63, balancing thermal and economic performance. Using toluene as the working fluid, the optimized system achieved thermal and exergy efficiencies of 56.06 % and 9.15 %, respectively, with a net present value of 58.12 M$, a payback period of 5.23 years, and an equivalent CO2 reduction of 24.8 kt, outperforming the solar organic Rankine cycle system in energy efficiency, economic performance, and carbon reduction potential. The solar dual-stage evaporation multigeneration system achieved the best performance with toluene as the working fluid, while heptane provided longer running time and greater carbon reduction potential. At low rated power generation, performance differences between toluene and heptane were minimal, while as the rated power generation increased, toluene showed significant advantages in thermal and economic performance.

QuAC: Quick Attribute-Centric Type Inference for Python

Python’s dynamic typing facilitates rapid prototyping and underlies its popularity in many domains. However, dynamic typing reduces the power of many static checking and bug-finding tools. Python type annotations can make these tools more useful. Type inference tools aim to reduce developers’ burden of adding them. However, existing type inference tools struggle to support dynamic features, infer correct types (especially container type parameters and non-builtin types), and run in reasonable time. Inspired by Python’s duck typing, where the attributes accessed on Python expressions characterize their implicit interfaces, we propose QuAC (Quick Attribute-Centric Type Inference for Python). At its core, QuAC collects attribute sets for Python expressions and leverages information retrieval techniques to predict classes from these attribute sets. It also recursively predicts container type parameters. We evaluate QuAC’s performance on popular Python projects. Compared to state-of-the-art non-LLM baselines, QuAC predicts types with high accuracy complementary to those predicted by the baselines while not sacrificing coverage. It also demonstrates clear advantages in predicting container type parameters and non-builtin types and reduces run times. Furthermore, QuAC is nearly two orders of magnitude faster than an LLM-based method while covering nearly half of its errorless non-trivial type predictions. It is also significantly more consistent at predicting container type parameters and non-builtin types than the LLM-based method, regardless of whether the project has ground-truth type annotations.

Systematic Screening of Autosomal Dominant Tubulointerstitial Kidney Disease- MUC1 27dupC Pathogenic Variant through Exome Sequencing.

The MUC1 gene is associated with autosomal dominant tubulointerstitial kidney disease (ADTKD), leading to CKD. Current methods of sequencing, like exome sequencing, rarely detect MUC1 pathogenic variants because of the variable number of tandem repeats (VNTR) in MUC1 exon2. We demonstrated that combining fast read filtering with a sensitive VNTR genotyping strategy enables systematic screening of 27dupC pathogenic MUC1 variant from exome data. We initially validated our bioinformatics pipeline in a proof-of-concept cohort incorporating exome data from 33 participants with a known MUC1 pathogenic variant identified by Snapshot PCR and confirmed by 54 MUC1-negative individuals for negative control. We then retrospectively analyzed exome sequencing data from January 2019 to October 2023 from 3512 adult participants with nephropathy of unknown origin. Finally, we prospectively validated our pipeline in 825 additional participants enrolled from November 2023. SharkVNTyper accurately identified MUC1 variants in 32 out of 33 participants and excluded its presence in all the 54 negative controls in the proof-of-concept cohort (sensitivity of 97%, specificity of 100%). Integration of Shark tool with VNTyper significantly reduced running time from 6-12 hours to 5-10 minutes per sample, allowing both retrospective and prospective analysis. In the retrospective cohort, SharkVNTyper identified 23 additional positive cases that were not suspected clinically and had been missed in the initial exome analysis; 18 of these cases were confirmed as carrying the MUC1 27dupC mutation by low-throughput Snapshot PCR. In the prospective cohort of 825 CKD cases, systematic screening discovered 13 positive cases, with 12 confirmed by PCR. Overall, of 63 participants (1.4% of 4653) with molecularly confirmed ADTKD-MUC1, comprehensive diagnoses and descriptions of the disease were available for 24 participants. Median age of kidney failure was 50 years, 38% exhibited bilateral multiple kidney cysts, 8% had early-onset gout, and 58% had arterial hypertension. SharkVNTyper enables the analysis of highly repeated regions, such as the MUC1 VNTR, and facilitates the systematic screening of ADTKD-MUC1 from exome data, fostering 27dupC variation identification.

A design of computational fuzzy set-based semantics for extracting linguistic summaries

Linguistic summarization is to extract a set of summary sentences in the form of natural language, so-called linguistic summaries, from numeric data. The extracted linguistic summaries should be compact and diverse, and have a validity measure greater than a given threshold, so genetic algorithms are applied to extract such linguistic summaries. Besides, the interpretability of linguistic summary content is considered in recent studies in such a way that enlarged hedge algebras are applied to generate multi-semantic structures for linguistic words of linguistic variables ensuring the interpretability of the content of the linguistic summaries. However, the membership function of computational fuzzy-set-based semantics of linguistic words is usually in the shape of trapezoid. In this paper, a membership function of the form S function is applied to improve the quality of extracted linguistic summaries. Besides, the applied algorithm is parallelized to reduce running time. The experimental results with the creep dataset have demonstrated the effectiveness of the proposed method

Quadratic horizontally elastic not-first/not-last filtering algorithm for cumulative constraint

The not-first/not-last rule is a pendant of the edge finding rule, generally embedded in the cumulative constraint during constraint-based scheduling. It is combined with other filtering rules for more pruning of the tree search. In this paper, the Profile data structure in which tasks are scheduled in a horizontally elastic way is used to strengthen the classic not-first/not-last rule. Potential not-first task intervals are selected using criteria (specified later in the paper), and the Profile data structure is applied to selected task intervals. We prove that this new rule subsumes the classic not-first rule. A quadratic filtering algorithm is proposed for the new rule, thus improving the complexity of the horizontally elastic not-first/not-last algorithm from O(n3) to O(n2). The fixed part of external tasks that overlap with the selected task intervals is considered during the computation of the earliest completion time of task intervals. This improvement increases the filtering power of the algorithm while remaining quadratic. Experimental results, on a well-known suite of benchmark instances of Resource-Constrained Project Scheduling Problems (RCPSPs), show that the propounded algorithms are competitive with the state-of-the-art not-first algorithms in terms of tree search and running time reduction.

Applied static analysis and specialization of cross-core syscalls for multi-core AUTOSAR OS

The development of static real-time control systems often follows a closed-world assumption, allowing extensive RTOS-aware whole-program optimization. For single-core systems, previous work could show the high potential of control-flow-aware static system-call tailoring. However, due to an exponential state explosion in the analysis phase, it cannot simply be extended to a multi-core setting, since the core’s relative timing to each other is undetermined. In this work, we present MultiSSE, a multi-core capable and RTOS-aware static whole-system optimization. First, MultiSSE analyzes the system by determining the relative positions of multiple cores only when necessary. For that, it exploits structural control flow and optionally timing information to handle each core separately as much as possible. Based on the analysis result, a synthesis applies lock elision and system-call optimization to generate specialized multi-core real-time systems for AUTOSAR OS. To enable a static prediction of the run-time reduction, we additionally provide cost models for the optimized cross-core system calls and evaluate the approach with synthetic benchmarks and a real-world quadrotor application. MultiSSE was able to optimize or even completely elide costly cross-core system calls and system objects leading to a reduction of up to 14% of a task’s execution time. In this extended version of a conference publication (Entrup et al. 2023), we provide an advanced description, new cost models, and an end-to-end measurement by developing a synthesis complementing the analysis.

An end-to-end framework for private DGA detection as a service.

Domain Generation Algorithms (DGAs) are used by malware to generate pseudorandom domain names to establish communication between infected bots and command and control servers. While DGAs can be detected by machine learning (ML) models with great accuracy, offering DGA detection as a service raises privacy concerns when requiring network administrators to disclose their DNS traffic to the service provider. The main scientific contribution of this paper is to propose the first end-to-end framework for privacy-preserving classification as a service of domain names into DGA (malicious) or non-DGA (benign) domains. Our framework achieves these goals by carefully designed protocols that combine two privacy-enhancing technologies (PETs), namely secure multi-party computation (MPC) and differential privacy (DP). Through MPC, our framework enables an enterprise network administrator to outsource the problem of classifying a DNS (Domain Name System) domain as DGA or non-DGA to an external organization without revealing any information about the domain name. Moreover, the service provider's ML model used for DGA detection is never revealed to the network administrator. Furthermore, by using DP, we also ensure that the classification result cannot be used to learn information about individual entries of the training data. Finally, we leverage post-training float16 quantization of deep learning models in MPC to achieve efficient, secure DGA detection. We demonstrate that by using quantization achieves a significant speed-up, resulting in a 23% to 42% reduction in inference runtime without reducing accuracy using a three party secure computation protocol tolerating one corruption. Previous solutions are not end-to-end private, do not provide differential privacy guarantees for the model's outputs, and assume that model embeddings are publicly known. Our best protocol in terms of accuracy runs in about 0.22s.

Clock mesh synthesis through dynamic programming with physical parameters consideration

In response to the evolving technological landscape, the traditional clock network architecture faces challenges in meeting the complexities of modern System-on-Chip (SoC) designs. While the clock mesh topology offers resilience against On-Chip Variation (OCV) fluctuations, its manual implementation leaves room for advancements in methodology and swift analytical techniques. This paper introduces an innovative clock mesh synthesis approach, leveraging dynamic programming algorithms and emphasizing compliance with critical physical implementation parameters. Our experimental results demonstrate a significant 26.6% reduction in power consumption compared to baseline methodologies. Moreover, it achieves an impressive average runtime reduction of 78.0% when contrasted with traditional simulation methods. These findings underscore the potential of our methodology to enhance the efficiency and power management of clock mesh designs.

Proactive blockchain deployment mechanism in resource-constrained rate-splitting multiple access IoT networks

In this paper, we propose an innovative blockchain deployment mechanism tailored for resource-constrained rate-splitting multiple access (RSMA) Internet of Things (IoT) networks. To address the storage limitations and security concerns inherent in IoT environments, our approach includes several advanced techniques. First, we utilize a Block Allocation Strategy Contract (BASC) to manage storage efficiently. Second, we employ a deep reinforcement learning (DRL) model to dynamically perform block assignment, ensuring optimal storage utilization. To accommodate the resource constraints of IoT devices, we adopt the Smart Byzantine Fault Tolerance (SBFT) consensus mechanism, which offers low latency and energy efficiency. Our framework demonstrates superior performance in storage optimization and reduced running time compared to existing methods, making it well-suited for large-scale IoT networks. Through extensive simulations, we validate the effectiveness of our proposed solution in enhancing security and operational efficiency in RSMA IoT networks.

An insight into the effect of different parameters for optimizing parallel reservoir simulation performance

Parallel simulation significantly reduces runtime in oil reservoir simulations, enabling higher-resolution analysis and a better understanding of reservoir behavior. This study comprehensively evaluates how simulation time, hardware configuration, and various physical parameters impact the efficiency, scalability, and performance metrics such as run time reduction (RTR) and number of optimal cores (NOC) in parallel simulations. This study introduced the reduced simulation time (RST) concept, an innovative approach that enhances parallel simulation efficiency, and dynamic load balance. In this approach, simulations were performed using only 5% of the total simulation time. The findings reveal that in 80% of the models, switching to RST mode has no effect on NOC but significantly reduces overall execution time in most cases. In complex models, this time reduction can be as high as 60% compared to serial mode. In the remaining 20%, the number of cores in the total simulation time (TST) mode may be higher than in the RST mode, but this difference can be disregarded due to limited hardware resources and no change in RTR.

Dual Kalman Filter Based on a Single Direction under Colored Measurement Noise for INS-Based Integrated Human Localization

For inertial-based integrated pedestrian navigation, the navigation environment might affect the positioning accuracy in different directions. Meanwhile, complex filtering algorithms can reduce computational efficiency. Therefore, one dual Kalman filter (KF) based on a single direction under a colored measurement noise (CMN) scheme is developed herein to improve the robustness and operational efficiency. The proposed method involves designing a data fusion model for the KF that integrates data from an inertial navigation system (INS) and ultrawideband (UWB). Subsequently, the INS/UWB integrated model-based KF under CMN (cKF) will be derived. Then, two sub-cKFs are proposed to fuse the data in the east and north directions, respectively. The empirical findings highlight the superior performance of the proposed approach over the KF for position estimation accuracy and runtime reduction, demonstrating its effectiveness.

Hybrid PDES Simulation of HPC Networks Using Zombie Packets

Although high-fidelity network simulations have proven to be reliable and cost-effective tools to peer into architectural questions for high-performance computing (HPC) networks, they incur a high resource cost. The time spent in simulating a single millisecond of network traffic in the highest detail can take hours, even for static, well-behaved traffic patterns such as uniform random. Surrogate models offer a significant reduction in runtime, yet they cannot serve as complete replacements and should only be used when appropriate. Thus, there is a need for hybrid modeling, where high-fidelity simulation and surrogates run side-by-side. We present a surrogate model for HPC networks in which: packets bypass the network, while the network state is left untouched, i.e , suspended. To bypass the network, we use historical data to estimate the arrival time at which every packet should be scheduled at; to suspend the network, all in-flight packets are scheduled to arrive at their destinations, and are kept in the system to awaken as zombies when switching back to high-fidelity. Speedup for a hybrid model is relative to the proportion of surrogate to high-fidelity. This light-weight surrogate obtained up to 76 × speedup. Keeping the zombies in the network showed an increase in the accuracy of the high-fidelity simulation on restart when compared to restarting the network from an empty state.

NxtSPR: A deadlock-free shortest path routing dedicated to relaying for Triplet-Based many-core Architecture

Deadlock-free routing is a significant challenge in Network-on-Chip (NoC) design as it affects the network’s latency, power consumption, and load balance, impacting the performance of multi-processor systems-on-chip. However, achieving deadlock-free routing will routinely result in expensive overhead as previous solutions either sacrifice performance or power efficiency to proactively avoid deadlocks or impose high hardware complexity to resolve deadlocks when they occur reactively. Utilizing the various characteristics of NoC to implement deadlock-free routing can be significantly more cost-effective with less impact on performance. This paper proposes a relay routing algorithm (NxtSPR) with a shortest path property and a deadlock prevention mechanism based on a synchronized Hamiltonian ring. The proposal is based on an in-depth study of the characteristics of a Triplet-Based many-core Architecture (TriBA) NoC. We establish various important topology-related theories and perform a formal verification (proof-based) for them. By utilizing the critical subgraph and apex of TriBA, NxtSPR can pre-calculate downstream nodes forwarding ports for packets by using a concise judgment strategy. This significantly reduces the computational overhead required for data transmission while optimizing the pipeline design of routers to decrease packet transmission latency and power consumption compared to other TriBA routing algorithms. We group the data transmissions according to the levels of maximum Hamiltonian edges a packet will traverse during its transmission life cycle. Independent data transmissions between groups can avoid mutual interference and resource competition, eliminating potential deadlocks. Gem5 simulation results show that, under the synthetic traffic patterns, compared to the representative (Table) and up-to-date (SPR4T) routing algorithms, NxtSPR achieves a 20.19%, 14.76%, and 5.54%, 4.66% reduction in average packet latency and per-packet power consumption, respectively. Moreover, it has an average of 18.50% and 4.34% improvement in throughput, as compared to them. PARSEC benchmark results show that NxtSPR reduces application runtime by up to a maximum of 22.30% and 12.82% compared to Table and SPR4T, and running the same applications with TriBA results in a maximum runtime reduction of 10.77% compared to 2D-Mesh.

Skeletal muscle p53-depletion uncovers a mechanism of fuel usage suppression that enables efficient energy conservation.

The ability of skeletal muscle to respond adequately to changes in nutrient availability, known as metabolic flexibility, is essential for the maintenance of metabolic health and loss of flexibility contributes to the development of diabetes and obesity. The tumour suppressor protein, p53, has been linked to the control of energy metabolism. We assessed its role in the acute control of nutrient allocation in skeletal muscle in the context of limited nutrient availability. A mouse model with inducible deletion of the p53-encoding gene, Trp53, in skeletal muscle was generated using the Cre-loxP-system. A detailed analysis of nutrient metabolism in mice with control and knockout genotypes was performed under ad libitum fed and fasting conditions and in exercised mice. Acute deletion of p53 in myofibres of mice activated catabolic nutrient usage pathways even under ad libitum fed conditions, resulting in significantly increased overall energy expenditure (+10.6%; P=0.0385) and a severe nutrient deficit in muscle characterized by depleted intramuscular glucose and glycogen levels (-62,0%; P<0.0001 and -52.7%; P<0.0001, respectively). This was accompanied by changes in marker gene expression patterns of circadian rhythmicity and hyperactivity (+57.4%; P=0.0068). These metabolic changes occurred acutely, within 2-3days after deletion of Trp53 was initiated, suggesting a rapid adaptive response to loss of p53, which resulted in a transient increase in lactate release to the circulation (+46.6%; P=0.0115) from non-exercised muscle as a result of elevated carbohydrate mobilization. Conversely, an impairment of proteostasis and amino acid metabolism was observed in knockout mice during fasting. During endurance exercise testing, mice with acute, muscle-specific Trp53 inactivation displayed an early exhaustion phenotype with a premature shift in fuel usage and reductions in multiple performance parameters, including a significantly reduced running time and distance (-13.8%; P=0.049 and -22.2%; P=0.0384, respectively). These findings suggest that efficient nutrient conservation is a key element of normal metabolic homeostasis that is sustained by p53. The homeostatic state in metabolic tissues is actively maintained to coordinate efficient energy conservation and metabolic flexibility towards nutrient stress. The acute deletion of Trp53 unlocks mechanisms that suppress the activity of nutrient catabolic pathways, causing substantial loss of intramuscular energy stores, which contributes to a fasting-like state in muscle tissue. Altogether, these findings uncover a novel function of p53 in the short-term regulation of nutrient metabolism in skeletal muscle and show that p53 serves to maintain metabolic homeostasis and efficient energy conservation.

Graph-Based Hotspot Detection of Socio-Economic Data Using Rough-Set

The term hotspot refers to a location or an area where the occurrence of a particular phenomenon, event, or activity is significantly higher than in the surrounding areas. The existing statistical methods need help working well on discrete data. Also, it can identify a false hotspot. This paper proposes a novel graph-based hotspot detection using a rough set (GBHSDRS) for detecting the hotspots. This algorithm works well with discrete spatial vector data. Furthermore, it removes the false hotspot by finding the statistical significance of the identified hotspots. A rough set theory is applied to the graph of the spatial polygon data, and the nodes are divided into lower, boundary, and negative regions. Therefore, the candidate hotspot belongs to the lower region of the set, and the boundary value analysis will ensure the identification of the hotspots if the hotspot is present in the dataset. The p-value is used to find the statistical significance of the hotspots. The algorithm is tested on the socioeconomic data of Uttar Pradesh (UP) from 1991 on medical facilities. The average gain in density and Hotspot Prediction Accuracy Index (HAPI) of the detected hotspots is 26.54% and 23.41%, respectively. An average reduction in runtime is 27.73%, acquired compared to all other methods on the socioeconomic data.

MethParquet: an R package for rapid and efficient DNA methylation association analysis adopting Apache Parquet.

Genome-wide DNA methylation (DNAm) profiling is indispensable for unveiling how DNAm regulates biological pathways and individual phenotypes. However, managing and analyzing extensive DNAm data generated from large cohort studies present computational obstacles. Apache Parquet is a data file format that allows for efficient data storage, retrieval, and manipulation, alleviating computational hurdles associated with conventional row-based formats. We here introduce MethParquet, the first R package leveraging the columnar Parquet format for efficient DNAm data analysis. It can be used for data extraction, methylation risk score calculation, epigenome-wide association analyses, and other standard post-quality control tasks. The package flexibly implements diverse regression models. Via a public methylation dataset, we show the efficiency of this package in reducing running time and RAM usage in large-scale EWAS. The MethParquet R package is publicly available on the GitHub repository https://github.com/ZWangTen/MethParquet. It includes a vignette and a toy dataset derived from a public resource.

The Development of a Rapid, Cost-Effective, and Green Analytical Method for Mercury Speciation.

Mercury is a naturally occurring metal found in various inorganic and organic forms within the environment. Due to its high toxicity, there is global concern regarding human exposure to this element. The combination of high-performance liquid chromatography and inductively coupled plasma mass spectrometry (HPLC-ICP-MS) is commonly used to analyze the different forms of mercury in a sample due to its high sensitivity and ability to selectively detect mercury. However, the traditional HPLC-ICP-MS methods are often criticized for their lengthy analysis times. In this study, we have refined the conventional approach by transitioning to ultra-high performance liquid chromatography coupled with inductively coupled plasma mass spectrometry (UHPLC-ICP-MS). This modification has resulted in significant reductions in runtime as well as reagent and argon usage, thereby offering a more rapid, environmentally friendly, and cost-effective method. We successfully adapted an HPLC-ICP-MS method to UHPLC-ICP-MS, achieving the analysis of Hg2+ and MeHg+ within 1 min with a mobile phase consumption of only 0.5 mL and a sample volume of 5.0 µL; this is a major advance compared to HPLC analysis with run times generally between 5 and 10 min. The method's performance was assessed by analyzing muscle and liver tissue samples (serving as reference material) from fish, demonstrating the versatility of the method in relation to different complex matrices.

A Model Based on Fuzzy Neural Networks for Sharing Digital Educational Resources in English

The ineffective digitization of core English course resources, due to limited autonomy, fragmentation, and inadequate management, has prompted the development of digital teacher libraries using multimedia and Internet technologies. Fuzzy neural networks (FNNs), combining fuzzy and neural network controls, have emerged as a promising approach for mathematical modeling. This paper presents an FNN-based model for sharing digital English educational resources and proposes an effective guidance mechanism for sharing digital science education resources. Experimental results show that the FNN outperforms the Apriori algorithm in training sample error by 0.02069 and reduces running time by 0.0034 seconds on average. This indicates the FNN's superior approximation capabilities with appropriate initial fuzzy rules. Consequently, the FNN-based model enhances the value and quality of educational resources, expands the capacity of information resource libraries, and effectively mitigates the creation of low-quality resources.

Shared-Memory Parallel Edmonds Blossom Algorithm for Maximum Cardinality Matching in General Graphs.

The Edmonds Blossom algorithm is implemented here using depth-first search, which is intrinsically serial. By streamlining the code, our serial implementation is consistently three to five times faster than the previously fastest general graph matching code. By extracting parallelism across iterations of the algorithm, with coarse-grain locking, we are able to further reduce the run time on random regular graphs four-fold and obtain a two-fold reduction of run time on real-world graphs with similar topology. Solving very sparse graphs (average degree less than four) exhibiting community structure with eight threads led to a slow down of three-fold, but this slow down is replaced by marginal speed up once the average degree is greater than four. We conclude that our parallel coarse-grain locking implementation performs well when extracting parallelism from this augmenting-path-based algorithm and may work well for similar algorithms.

Accelerating Lagrangian transport simulations on graphics processing units: performance optimizations of Massive-Parallel Trajectory Calculations (MPTRAC) v2.6

Abstract. Lagrangian particle dispersion models are indispensable tools for the study of atmospheric transport processes. However, Lagrangian transport simulations can become numerically expensive when large numbers of air parcels are involved. To accelerate these simulations, we made considerable efforts to port the Massive-Parallel Trajectory Calculations (MPTRAC) model to graphics processing units (GPUs). Here we discuss performance optimizations of the major bottleneck of the GPU code of MPTRAC, the advection kernel. Timeline, roofline, and memory analyses of the baseline GPU code revealed that the application is memory-bound, and performance suffers from near-random memory access patterns. By changing the data structure of the horizontal wind and vertical velocity fields of the global meteorological data driving the simulations from structure of arrays (SoAs) to array of structures (AoSs) and by introducing a sorting method for better memory alignment of the particle data, performance was greatly improved. We evaluated the performance on NVIDIA A100 GPUs of the Jülich Wizard for European Leadership Science (JUWELS) Booster module at the Jülich Supercomputing Center, Germany. For our largest test case, transport simulations with 108 particles driven by the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5 reanalysis, we found that the runtime for the full set of physics computations was reduced by 75 %, including a reduction of 85 % for the advection kernel. In addition to demonstrating the benefits of code optimization for GPUs, we show that the runtime of central processing unit (CPU-)only simulations is also improved. For our largest test case, we found a runtime reduction of 34 % for the physics computations, including a reduction of 65 % for the advection kernel. The code optimizations discussed here bring the MPTRAC model closer to applications on upcoming exascale high-performance computing systems and will also be of interest for optimizing the performance of other models using particle methods.