Rate Of Performance Improvement Research Articles

Improvement of Thermal Performance using Spineloxides/Water Nanofluids in the Heat Recovery Unit with Air-to-Air Thermosiphone Mechanism

Heat recovery units are used for pre-heating the fresh air in waste heat plants. The nanofluid used as working fluid in the heat pipe will allow the heat recovery unit to benefit from the waste heat at lower temperatures, since it can evaporate at a temperature lower than the temperature of the base fluid. Therefore, the range of the operating temperature of the heat recovery unit will be increased. Different temperatures and flow rates of waste heat were used to find the optimum conditions for the evaporation of spinel nanofluid in the evaporator region of the heat pipe. Similar conditions were investigated for both of the cold fluid and optimal conditions on the condensation of nanofluid in the condenser region. In this study, an experimental setup was designed to improve the thermal performance of the air-to-air containing heat pipe heat recovery unit. The experiments were carried out using pure water-based nanofluids containing nano-sized ZnOAl2O3, MgOAl2O3, FeOAl2O3 particles, and the efficiencies of the heat transfer were investigated in this experiment. A thermosiphon mechanism that consists of five heat pipes, the length of each one is equal to 1 m, and the inner and the outer diameter of them is equal to 23.4 mm and 25.4 mm, respectively, those pipes are vacuumed, without wick and copper material. Heat pipes in the experimental setup; the evaporation zone is positioned to form 450 mm, the condensation zone 400 mm and the middle 150 mm parts to form the adiabatic region. At the beginning of the experiments, the working fluid charged to fill the 1/3 of the evaporation zone volumes of the heat pipes. Two different cooling air flow rates, 30 g·s−1 and 60 g·s−1, and three different heating air flow rates, 50 g·s−1, 70 g·s−1, 90 g·s−1 have been used, to calculate the heat obtained from the condensation zone. All the experiments were carried out by applying two different heaters with a power of, 1000 W and 2000 W, in the evaporation zone. Thus, the optimum temperatures and flow rates were found for all Reynolds numbers in the hot and cold fluid region, and its helps to determining the ranges of the operating temperature in the heat recovery unit. The experimental results that obtained from the two different working fluids (pure water and nanofluid) are compared. The best thermal performance of the system was found in the hot and cold air duct, when the Reynolds number was equal to 12 000. The thermal performance improvement rate was found as 92.8 % when the FeOAl2O3/water nanofluid used as working fluid, and 1000 W heater power and 60 g·s−1 cooling air flow applied.

Functional performance improvement data and patent sets for 30 technology domains with measurements of patent centrality and estimations of the improvement rate.

This article accompanies the study presented in Triulzi et al. (2020) [1]. It briefly describes and makes available the data on functional performance for 30 technology domains, their patent sets, the measurement of patent centrality and the method to estimate the yearly technology performance improvement rate (TIR) that underly that study. Some of this data (performance time series and the lists of patents for 28 domains) has been collected by other authors for previous studies but were previously unavailable to the public. Measurements of patent centrality and other patent-based indicators for the 30 domains, and for 5.259.906 utility patents granted by the United States Patent and Trademark Office between 1976 and 2015 are novel data contributed by Triulzi et al. (2020) [1]. Here we organize, describe and make available the collection of data in its entirety. This allows anyone interested to replicate the study or use the method to estimate the improvement rate of a given technology for which patents can be identified. For a detailed description of the data and methods see Triulzi et al. (2020) [1].

Breaking the athletics world record in the 100 and 400 meters: an alternative method for assessment.

The top 10 athletes in the International Association of Athletic Federations in 100-m and 400-m ranking for each sex were assessed for their history of race times before achieving their personal record (PR). The main goal of this study was to create a new method for optimal performance improvement rate assessment for coaches and athletes aiming the World Record. The difference between PR ('current' season) and the best race time in the last season was defined as the first season improvement rate (1-SIR), whereas the average improvement rate in the last and preceding seasons was the multi-season improvement rate (M-SIR). 1-SIR and M-SIR were calculated for each athlete. The sex comparison for the 100 m event showed a significant difference in the M-SIR in favor of women. No statistical differences were identified for the 400 m event, with a trivial effect in both 1-SIR and M-SIR. As a practical applicability, graph plots were designed to help verifying the improvement rate of athletes and to evaluate whether a long-term training strategy induced an acceptable performance improvement or whether some adjustments needed and check within the plots if the improvement rate is within the average of the top-10 athletes of their event.

Estimating technology performance improvement rates by mining patent data

The future direction of technology development depends on the relative yearly rate of functional performance improvement of different technologies. We use patent data to identify accurate and reliable predictors of this rate for 30 technologies. We illustrate how patent-based predictors should be normalized to correct for possible confounding factors introduced by changing patenting dynamics. We test the accuracy and reliability of various predictors by means of a Monte Carlo cross-validation exercise. We find that a measure of the centrality of domains’ patented inventions in the overall US patent citation network is an accurate and highly reliable predictor of improvement rates.

Technological Improvement Rate Estimates for All Technologies: Use of Patent Data and an Extended Domain Description

In this work, we attempt to provide a comprehensive granular account of the pace of technological change. More specifically, we survey estimated yearly performance improvement rates for nearly all definable technologies for the first time. We do this by creating a correspondence of all patents within the US patent system to a set of technology domains. A technology domain is a body of patented inventions achieving the same technological function using the same knowledge and scientific principles. We obtain a set of 1757 domains using an extension of the previously defined classification overlap method (COM). These domains contain 97.14% of all patents within the entire US patent system. From the identified patent sets, we calculated the average centrality of the patents in each domain to estimate their improvement rates, following a methodology tested in prior work. The estimated improvement rates vary from a low of 1.9% per year for the Mechanical Skin treatment- Hair Removal and wrinkles domain to a high of 228.8% per year for the Network management- client-server applications domain. We developed a one-line descriptor identifying the technological function achieved and the underlying knowledge base for the largest 50, fastest 20 as well as slowest 20 of these domains, which cover more than forty percent of the patent system. In general, the rates of improvement were not a strong function of the patent set size and the fastest improving domains are predominantly software-based. We make available an online system that allows for automated searching for domains and improvement rates corresponding to any technology of interest to researchers, strategists and policy formulators.

Dosimetry, efficacy, and safety of three-dimensional printing noncoplanar template-assisted and CT-guided 125I seed implantation for recurrent retroperitoneal lymphatic metastasis after external beam radiotherapy

ObjectivesTo evaluate the dosimetry, efficacy, and safety of radioactive 125I seed implantation (RISI) assisted by three-dimensional printing noncoplanar template (3D-PNCT) and CT for recurrent retroperitoneal lymphatic metastasis (RRLM) after previous external beam radiotherapy. Methods and MaterialsFrom June 2016 to August 2018, 32 patients with RRLM successfully underwent 3D-PNCT–assisted and CT-guided RISI. The dosimetry, pain relief rate, performance improvement rate, overall response rate, disease control rate, local control time (LCT), overall survival (OS), and safety profiles were evaluated. ResultsDosimetric results showed that the D90, D100, V100, V150, V200, and homogeneity index were consistent in preoperation and postoperation (p > 0.05), except for the external index and conformal index (p = 0.048, p = 0.034). After RISI, 81.3% of the patients achieved pain relief, and 71.9% achieved an improvement of performance. The overall response rate and disease control rate were 85.3% and 94.1%, respectively. The LCT rates reached 66.2% and 43.2% in 1 year and two years, respectively, with a median LCT of 15.8 months. The OS rates were 74.1% and 28.1% in 1 year and two year, respectively, with a median OS reaching 17.6 months. Univariate analysis showed that when D90 > 130 Gy or D100 > 63 Gy or tumor size ≤49.8 cm3, LCT was extended significantly, but not for OS. Except for two patients developing Grade 1 retroperitoneal hematomas, no other severe adverse events were observed. Conclusions3D-PNCT and CT guidance provide excellent accuracy for RISI, which can be an effective and safe alternative for RRLM after external beam radiotherapy. Radiation dose and tumor size seem to significantly influence the local control.

Motor Imagery Training With Neurofeedback From the Frontal Pole Facilitated Sensorimotor Cortical Activity and Improved Hand Dexterity.

To develop a real-time neurofeedback system from the anterior prefrontal cortex (aPFC) using functional near-infrared spectroscopy (fNIRS) for motor rehabilitation, we investigated the effects of motor imagery training with neurofeedback from the aPFC on hand dexterity and cerebral hemodynamic activity during a motor rehabilitation task. Thirty-one right-handed healthy subjects participated in this study. They received motor imagery training six times for 2 weeks under fNIRS neurofeedback from the aPFC, in which they were instructed to increase aPFC activity. The real group subjects (n = 16) were shown real fNIRS neurofeedback signals from the aPFC, whereas the sham group subjects (n = 15) were shown irrelevant randomized signals during neurofeedback training. Before and after the training, hand dexterity was assessed by a motor rehabilitation task, during which cerebral hemodynamic activity was also measured. The results indicated that aPFC activity was increased during the training, and performance improvement rates in the rehabilitation task after the training was increased in the real group when compared with the sham group. Improvement rates of mean aPFC activity across the training were positively correlated with performance improvement rates in the motor rehabilitation task. During the motor rehabilitation task after the training, the hemodynamic activity in the left somatosensory motor-related areas [premotor area (PM), primary motor area (M1), and primary somatosensory area (S1)] was increased in the real group, whereas the hemodynamic activity was increased in the supplementary motor area in the sham group. This hemodynamic activity increases in the somatosensory motor-related areas after the training correlated with aPFC activity during the last 2 days of motor imagery training. Furthermore, improvement rates of M1 hemodynamic activity after the training was positively correlated with performance improvement rates in the motor rehabilitation task. The results suggest that the aPFC might shape activity in the somatosensory motor-related areas to improve hand dexterity. These findings further suggest that the motor imagery training using neurofeedback signals from the aPFC might be useful to patients with motor disability.

Technological development of key domains in electric vehicles: Improvement rates, technology trajectories and key assignees

Technology innovation in electric vehicles is of significant interest to researchers, companies and policy-makers of many countries. Electric vehicles integrate various kinds of distinct technologies and decomposing the overall electric vehicle field into several key domains allows determination of more detailed, valuable information. To provide both broader and more detailed information about technology development in the EV field, unlike most previous studies on electric vehicle innovation which analyzed this field as a whole, this research decomposed the electric vehicle field into domains, which are power electronics, battery, electric motor as well as charging and discharging subdomains and then further extracted the subdomains. Furthermore, In addition, the improvement rates, technology trajectories and major patent assignees in these domains and key subdomains are determined using patents extracted for each domain from the US patent system. The main findings are: (1) The estimated rates of performance improvement per year are 18.3% for power electronics, 7.7% for electric motors, 23.8% for charging and discharging and 11.7% for batteries. The relatively lower improvement rate for electric motors and batteries suggests their potential to hinder the popularization of electric vehicles. Besides, as for the subdomains, the relatively higher technology improvement rate of lithium-ion battery or permanent magnet motor in its domain supports the current trend of battery or motor type quantitively from a patent analysis view. A possible implication for the policy makers encouraging EV development is to issue more incentive plans for innovations in the battery and electric motor domains, especially for lithium-ion battery and permanent magnet motor. (2) The technology trajectories depict the development of four critical subdomains over time, which quantitively proves the focuses and emerging topics of the subdomains and thereby provide guidance to research topic selection. For example, the silicon negative electrode is a promising topic in the subdomain of lithium-ion battery. (3) The key players in the four critical subdomains appear to be Toyota and Honda in hybrid power electronics, E-One Moli Energy Corp in lithium-ion batteries, Panasonic in Permanent Magnet motors and Toyota in discharging. The key players found by the main path method from the view of innovation are also important players in EV from the market view. Other market participants should pay more attention to the adjustment of business strategy of these companies to monitor the market, and make effort to invent important EV related technologies.

Deadline Constrained based Resource Allocation in Cloud Environment

Cloud computing faces a challenge of handling huge amounts of data. The users keep on pushing the data without knowing the challenge in increased storage. Task Scheduling deals with allocating the task to a respective resource pool on a demand basis. Approaches have been built that handle requests from users with deadlines on the amount of request that can be handled. It is important to understand that the mechanism is available to handle the deadlines. The experimental results show that the proposed algorithm produces remarkable performance improvement rate on the total execution cost and total transfer time under meeting the deadline constraint. In view of the experimental results, the proposed algorithm provides a better-quality scheduling solution that is suitable for scientific application task execution in the cloud computing environment.

A deadline constrained scheduling algorithm for cloud computing system based on the driver of dynamic essential path.

To solve the problem of the deadline-constrained task scheduling in the cloud computing system, this paper proposes a deadline-constrained scheduling algorithm for cloud computing based on the driver of dynamic essential path (Deadline-DDEP). According to the changes of the dynamic essential path of each task node in the scheduling process, the dynamic sub-deadline strategy is proposed. The strategy assigns different sub-deadline values to every task node to meet the constraint relations among task nodes and the user’s defined deadline. The strategy fully considers the dynamic sub-deadline affected by the dynamic essential path of task node in the scheduling process. The paper proposed the quality assessment of optimization cost strategy to solve the problem of selecting server for each task node. Based on the sub-deadline urgency and the relative execution cost in the scheduling process, the strategy selects the server that not only meets the sub-deadline but also obtains much lower execution cost. In this way, the proposed algorithm will make the task graph complete within its deadline, and minimize its total execution cost. Finally, we demonstrate the proposed algorithm via the simulation experiments using Matlab tools. The experimental results show that, the proposed algorithm produces remarkable performance improvement rate on the total execution cost that ranges between 10.3% and 30.8% under meeting the deadline constraint. In view of the experimental results, the proposed algorithm provides better-quality scheduling solution that is suitable for scientific application task execution in the cloud computing environment than IC-PCP, DCCP and CD-PCP.

Quantification of technological progress in greenhouse gas (GHG) capture and mitigation using patent data

Greenhouse gas emissions from anthropogenic sources are believed to be the main cause of global warming. We estimate performance improvement rates of various GHG capture and mitigation technologies using a method based on patent centrality.

A scheduling algorithm for applications in a cloud computing system with communication changes

AbstractThis paper proposes a scheduling algorithm to solve the problem of task scheduling in a cloud computing system with time‐varying communication conditions. This algorithm converts the scheduling problem with communication changes into a directed acyclic graph (DAG) scheduling problem for existing fuzzy communication task nodes, that is, the scheduling problem for a communication‐change DAG (CC‐DAG). The CC‐DAG contains both computation task nodes and communication task nodes. First, this paper proposes a weighted time‐series network bandwidth model to solve the indefinite processing time (cost) problem for a fuzzy communication task node. This model can accurately predict the processing time of a fuzzy communication task node. Second, to address the scheduling order problem for the computation task nodes, a dynamic pre‐scheduling search strategy (DPSS) is proposed. This strategy computes the essential paths for the pre‐scheduling of the computation task nodes based on the actual computation costs (times) of the computation task nodes and the predicted processing costs (times) of the fuzzy communication task nodes during the scheduling process. The computation task node with the longest essential path is scheduled first because its completion time directly influences the completion time of the task graph. Finally, we demonstrate the proposed algorithm via simulation experiments. The experimental results show that the proposed DPSS produced remarkable performance improvement rate on the total execution time that ranges between 11.5% and 21.2%. In view of the experimental results, the proposed algorithm provides better quality scheduling solution that is suitable for scientific application task execution in the cloud computing environment than HEFT, PEFT, and CEFT algorithms.

Reducing health inequalities in England: does the demise of NHS Stop Smoking Services matter? Analysis of mandatory monitoring data.

Tobacco smoking is a key cause of mortality, morbidity and health inequalities. The unprecedented English health inequalities strategy (1999-2010) sought to reduce health inequalities, by, in part, instigating NHS Stop Smoking Services (SSS), initially targeted in deprived 'Spearhead' localities. Performance of SSS is assessed here in light of its role supporting the strategy, which evidence suggests achieved a reduction in health inequalities. SSS enrolment and four-week quits in Spearhead and non-Spearhead localities were compared during and after the strategy period, using regression models and routine monitoring data. Changes in SSS expenditure were estimated. After similar increases in enrolment and quits between Spearhead and other localities between 2003/4 and 2008/9, SSS in Spearhead localities experienced a 2-fold better rate of improvement in enrolment and quit performance over the 4 years to 2011/12. Since 2011/12, SSS have dramatically reduced, and expenditure had fallen by half in Spearhead localities by 2016/17. SSS, particularly in Spearhead localities, were expanded up to 2011/12, and this broadly coincides with the reduction in health inequalities. This suggests that although SSS did not achieve the scale anticipated, they have important potential, and the current demise of SSS should not be tolerated.

Hybrid gradient descent cuckoo search (HGDCS) algorithm for resource scheduling in IaaS cloud computing environment

Resource scheduling is a procedure for the distribution of resources over time to perform a required task and a decision making process in cloud computing. Optimal resource scheduling is a great challenge and considered to be an NP-hard problem due to the fluctuating demand of cloud users and dynamic nature of resources. In this paper, we formulate a new hybrid gradient descent cuckoo search (HGDCS) algorithm based on gradient descent (GD) approach and cuckoo search (CS) algorithm for optimizing and resolving the problems related to resource scheduling in Infrastructure as a Service (IaaS) cloud computing. This work compares the makespan, throughput, load balancing and performance improvement rate of existing meta-heuristic algorithms with proposed HGDCS algorithm applicable for cloud computing. In comparison with existing meta-heuristic algorithms, proposed HGDCS algorithm performs well for almost in both cases (Case-I and Case-II) with all selected datasets and workload archives. HGDCS algorithm is comparatively and statistically more effective than ACO, ABC, GA, LCA, PSO, SA and original CS algorithms in term of problem solving ability in accordance with results obtained from simulation and statistical analysis.

Pre-existing technological core and roots for the CRISPR breakthrough.

This paper applies objective methods to explore the technological origins of the widely acclaimed CRISPR breakthrough in the technological domain of genome engineering. Previously developed patent search techniques are first used to recover a set of patents that well-represent the genome editing domain before CRISPR. Main paths are then determined from the citation network associated with this patent set allowing identification of the three major knowledge trajectories. The most significant of these trajectories for CRISPR involves the core of genome editing with less significant trajectories involving cloning and endonuclease specific developments. The major patents on the core trajectory are consistent with qualitative expert knowledge of the topical area. A second set of patents that we call the CRISPR roots are obtained by finding the patents directly cited by the recent CRISPR patents along with patents cited by that set of patents. We find that the CRISPR roots contain 8 key patents from the genome engineering main path associated with restriction endonucleases and the expected strong connection of CRISPR to prior genome editing technology such as Zn finger nucleases. Nonetheless, analysis of the full CRISPR roots shows that a very wide array of technological knowledge beyond genome engineering has contributed to achieving the CRISPR breakthrough. Such breadth in origins is not surprising since “spillover” is generally perceived as important and previous qualitative studies of CRISPR have shown not only technological breadth in origins but scientific breadth as well. In addition, we find that the estimated rate of functional performance improvement of the CRISPR roots set is about 9% per year compared to the genome engineering set (~4% per year). These estimates indicate below average rates of improvement and may indicate that CRISPR (and perhaps yet undiscovered) genome engineering developments could evolve in effectiveness over an upcoming long rather than short time period.

States of focused attention and sequential action: A comparison of single session meditation and computerised attention task influences on top-down control during sequence learning

Motor sequence learning is considered the result of the outflow of information following cognitive control processes that are shared by other goal-directed behaviours. Emerging evidence suggests that focused-attention meditation (FAM) establishes states of enhanced cognitive control, that then exert top-down control biases in subsequent unrelated tasks. With respect to sequence learning, a single-session of FAM has been shown to entrain stimulus-dependent forms of sequential behaviour in meditation naïve individuals. In the present experiment, we compared single-session effects of FAM and a computerised attention task (CAT) to test if FAM-induced enhanced top-down control is generally comparable to cognitive tasks that require focused attention. We also investigated if effort, arousal or pleasure associated with FAM, or CAT explained the influence of these tasks on sequence learning. Relative to a rest-only control condition, both FAM and CAT resulted in shorter reaction time (RT) in a serial reaction time task (SRTT), and this enhanced RT performance was associated with higher reliance on stimulus-based planning as opposed to sequence representation formation. However, following FAM, a greater rate of improvement in RT performance was observed in comparison to both CAT and control conditions. Neither effort, arousal nor pleasure associated with FAM or CAT explained SRTT performance. These findings were interpreted to suggest that the effect of FAM states on increased top-down control during sequence learning is based on the focused attention control feature of this meditation. FAM states might be associated with enhanced cognitive control to promote the development of more efficient stimulus-response processing in comparison to states induced by other attentional tasks.

Estimating Technology Performance Improvement Rates by Mining Patent Data

The future direction of technology development depends on the relative yearly rate of functional performance improvement of different technologies. We use patent data to identify accurate and reliable predictors of this rate for 30 technologies. We illustrate how patent-based predictors should be normalized to correct for possible confounding factors introduced by changing patenting dynamics. We test the accuracy and reliability of various predictors by means of a Monte Carlo cross-validation exercise. We find that a measure of the centrality of domains' patented inventions in the overall US patent citation network is an accurate and highly reliable predictor of improvement rates.

HYBRID CAT SWARM OPTIMIZATION AND SIMULATED ANNEALING FOR DYNAMIC TASK SCHEDULING ON CLOUD COMPUTING ENVIRONMENT

The unpredictable number of task arriving at cloud datacentre and the rescaling of virtual processing elements can affect the provisioning of better Quality of Service expectations during task scheduling in cloud computing. Existing researchers have contributed several task scheduling algorithms to provide better QoS expectations but are characterized with entrapment at the local search and high dimensional breakdown due to slow convergence speed and imbalance between global and local search, resulting from lack of scalability. Dynamic task scheduling algorithms that can adjust to long-time changes and continue facilitating the provisioning of better QoS are necessary for cloud computing environment. In this study, a Cloud Scalable Multi-Objective Cat Swarm Optimization-based Simulated Annealing algorithm is proposed. In the proposed method, the orthogonal Taguchi approach is applied to enhance the SA which is incorporated into the local search of the proposed CSMCSOSA algorithm for scalability performance. A multi-objective QoS model based on execution time and execution cost criteria is presented to evaluate the efficiency of the proposed algorithm on CloudSim tool with two different datasets. Quantitative analysis of the algorithm is carried out with metrics of execution time, execution cost, QoS and performance improvement rate percentage. Meanwhile, the scalability analysis of the proposed algorithm using Isospeed-efficiency scalability metric is also reported. The results of the experiment show that the proposed CSM-CSOSA has outperformed Multi-Objective Genetic Algorithm, Multi-Objective Ant Colony and Multi-Objective Particle Swarm Optimization by returning minimum execution time and execution cost as well as better scalability acceptance rate of 0.4811−0.8990 respectively. The proposed solution when implemented in real cloud computing environment could possibly meet customers QoS expectations as well as that of the service providers.

A checkpointed league championship algorithm-based cloud scheduling scheme with secure fault tolerance responsiveness

In its simplest structure, cloud computing technology is a massive collection of connected servers residing in a datacenter and continuously changing to provide services to users on-demand through a front-end interface. The failure of task during execution is no more an accident but a frequent attribute of scheduling systems in a large-scale distributed environment. Recently, some computational intelligence techniques have been mostly utilized to decipher the problems of scheduling in the cloud environment, but only a few emphasis on the issue of fault tolerance. This research paper puts forward a Checkpointed League Championship Algorithm (CPLCA) scheduling scheme to be used in the cloud computing system. It is a fault-tolerance aware task scheduling mechanisms using the checkpointing strategy in addition to tasks migration against unexpected independent task execution failure. The simulation results show that, the proposed CPLCA scheme produces an improvement of 41%, 33% and 23% as compared with the Ant Colony Optimization (ACO), Genetic Algorithm (GA) and the basic league championship algorithm (LCA) respectively as parametrically measured using the total average makespan of the schemes. Considering the total average response time of the schemes, the CPLCA scheme produces an improvement of 54%, 57% and 30% as compared with ACO, GA and LCA respectively. It also turns out significant failure decrease in jobs execution as measured in terms of failure metrics and performance improvement rate. From the results obtained, CPLCA provides an improvement in both tasks scheduling performance and failure awareness that is more appropriate for scheduling in the cloud computing model.

Marathon Specialization in Elites: A Head Start for Africans.

To assess differences in event-specific specialization between elite African and non-African male marathon runners based on age, performance, and career length. The top 90 African marathoners from 2001 to 2015 were compared with the top 90 non-African marathoners from the same time period across various markers related to specialization age, performance, and career length. Independent t tests were used to identify significant differences (P < .05) between the African and non-African groups. Linear regression was used to explore the relationship between first half-marathon and best full-marathon performance. A 1-way ANOVA and Bonferroni correction was used to assess differences in specialization age and rates of performance improvement and decline. African marathoners were found to specialize, reach peak levels of performance, and retire at younger ages than non-African marathoners (P < .001). In addition, African marathoners were found to be faster at these same career time points and in half-marathon performance (P < .001). There was no significant difference in the number of career marathons run between groups, but African marathoners were found to race more frequently than non-African marathoners (P < .001). Half-marathon performance was positively correlated with marathon performance (r2 = .67). Marathon athletes who specialized at early ages experienced significantly higher rates of improvement than those who specialized at older ages. (P < .05). The findings suggest that elite African marathoners achieve a greater level of performance at younger ages than their non-African counterparts. Furthermore, current marathon talent-identification practice may benefit from using half-marathon performance.