Analytical Performance Model Research Articles

Context-aware temporal network representation of event logs: Model and methods for process performance analysis

Analysing performance of business processes is an important vehicle to improve their operation. Specifically, an accurate assessment of sojourn times and remaining times enables bottleneck analysis and resource planning. Recently, methods to create respective performance models from event logs have been proposed. These works have several limitations, though: They either consider control-flow and performance information separately, or rely on an ad-hoc selection of temporal relations between events. In this paper, we introduce the Temporal Network Representation (TNR) of a log. It is based on Allen’s interval algebra, comprises the pairwise temporal relations for activity executions, and potentially incorporates the context in which these relations have been observed. We demonstrate the usefulness of the TNR for detecting (unrecorded) delays and for probabilistic mining of variants when modelling the performance of a process. In order to compare different models from the performance perspective, we further develop a framework for measuring performance fitness. Under this framework, TNR-based process discovery is guaranteed to dominate existing techniques in measuring performance characteristics of a process. In addition, we show how contextual information in terms of the congestion levels of the process can be mined in order to further improve capabilities for performance analysis. To illustrate the practical value of the proposed models, we evaluate our approaches with three real-life datasets. Our experiments show that the TNR yields an improvement in performance fitness over state-of-the-art algorithms, while congestion learning is able to accurately reconstruct congestion levels from event data.

Aerodynamic Analysis of Total Temperature Probe Thermal Performance Using Conjugate Heat Transfer

Understanding the behavior and design of total temperature probes used in gas turbine engines and other hot high-speed flows remains an active area of work, given the demands set by extreme operating conditions and ever-decreasing sensor size requirements. Where probe strut/mount cooling is necessary for probe survivability, conduction effects can be significant. The most effective mitigation of conduction errors results from convective heat transfer over the sensor. The goal of this paper is to provide a deeper understanding of the flow physics responsible for establishing the convective heat transfer on the sensor surface of vented-shield total temperature probes. This was accomplished by using high-fidelity computational fluid dynamics with conjugate heat transfer to examine the details of the aerodynamic environment establishing the heat transfer on the sensor surface. Probe geometric parameters varied and include the axial vent location, sensor diameter, and shield outer diameter. These parameters are cast into a set of nondimensional parameters that are used to build a set of correlations for local Reynolds numbers over regions of the sensor and Stanton number correlations for film coefficients. These film coefficients can be directly used in the new and existing low-order analytic thermal conduction probe performance models presented here and in the literature.

Implementation and performance analysis of the massively parallel method of characteristics based on GPU

The method of characteristics (MOC) is one of the most common methods for solving the neutron transport equation in practical application. Researches have been focused on the acceleration techniques and the parallel algorithm for improving the efficiency of MOC. The Graphics Processing Unit (GPU) provides an alternative method of parallelizing the MOC neutron transport sweep. In this work, a GPU-paralleled 2D MOC code is implemented, which employs the diamond difference (DD) scheme and the step characteristics (SC) scheme. Different parallel schemes which are ray-level, energy-group-level, and polar-angle-level, are analyzed to choose the proper parallel scheme. The C5G7 2D benchmark is calculated to verify the accuracy and efficiency of the code in different schemes with single precision and double precision. The bottlenecks of the GPU code are identified and the code is classified into three categories, which are compute-bound, memory-bound, and latency-bound, according to the performance analysis model introduced in this paper. In addition, corresponding optimization strategies are applied to improve the performance according to the analysis. Moreover, the speed, power efficiency, and hardware cost are compared for CPU and GPU based on a fictitious quarter core PWR problem. Numerical results demonstrate that the energy group-level parallelization can obtain the optimal performance on GPU. Optimization strategies are effective to improve the efficiency of the calculation on GPU, which indicates that the performance analysis model is useful and effective to locate the limitation of the code. Moreover, the GPU-version code is about 30 times faster than the CPU-version code with double precision and about 100 times faster with single precision, while the desired accuracy is kept. And the GPU delivers superior performance in both speed, energy efficiency, and hardware cost.

Analysis of Licklider Transmission Protocol for Reliable File Delivery in Space Vehicle Communications With Random Link Interruptions

Delay/disruption tolerant networking (DTN) was proposed as a networking architecture for reliable data delivery in space vehicle communications despite long propagation delays and/or lengthy link disruptions. As one of the main data transport protocols of DTN, Licklider transmission protocol (LTP) was proposed by the NASA for reliable data delivery in challenging space environments in which the existing transport protocols of DTN, TCP, and UDP do not perform effectively. Pair-wise comparisons among LTP, TCP, and UDP running in the DTN stack have previously been done using statistical methods; evaluation results and performance analysis are available in the existing LTP research literature. However, there is as yet no solid analytical model of LTP performance in the presence of link disruptions. In this paper, we present an analytical method to predict transmission performance of LTP for reliable file delivery in space vehicle communications with focus on the expected file delivery delay (or latency) over space channels characterized by random link disruptions. Realistic file-transmission experiments using a PC-based infrastructure are presented to validate the analytical models. The analysis and study results presented in this paper are expected to be useful in understanding the impact of transmission conditions and configurations, on the efficiency of LTP when deployed for reliable data delivery in space missions using communication links that are subject to frequent and random interruptions.

Measurement of productivity changes for general network production systems with stochastic data

Performance evaluation of network production systems has been widely studied in recent Data Envelopment Analysis (DEA) literature where internal relations of sub-units are taken into consideration. Most of prior work assumes network systems to have simple series or parallel structures. Complexities of some practical production processes require development of DEA models for their effective analysis. However; input, intermediate products and/or output data are often stochastic and linked to exogenous random variables in most applications. The current study extends Malmquist Productivity Index (MPI) for investigating productivity changes of general network production units with stochastic data in a DEA framework. The proposed stochastic performance analysis models are then transformed into deterministic equivalent non-linear forms so they could be simplified to deterministic programming with quadratic constraints. Numerical examples including an application to productivity evaluation of branches of a university system are presented to illustrate the applicability of the proposed framework.

Probabilistic Analysis of a Redundant Repairable System with Two Service Operations

A redundant system with two types of maintenance services, with single repair server and single replacement server, is considered. The replacement and repair times are random variables. Mathematical model for dependability and performance analysis is constructed and investigated in transient (time-dependent mode). An explicit solution in terms of matrix exponential calculus for transient probabilities is obtained.

The RELATIONSHIP OF SERVICE QUALITY WITH GOJEK CUSTOMER SATISFACTION IN THE ENVIRONMENT OF INDONESIAN CHRISTIAN UNIVERSITY OF CAWANG

The advantage of ride hailing business in society is huge nowadays. Go-jek plays role in service company, provides opportunity for job applicants from various background to achieve good occupation. To improve customer satisfaction, Go-jek made standard for service quality. As many competitors in ride hailing business, it is important to provide better service quality for customer. From this study result, it could be concluded that Go-jek has performed in content and satisfied service level using Importance Performance Analysis (IPA) and Cartesius Analysis Quadrant model. IPA in all questions for level of customer satisfaction reached 80% which means “strongly agree”; besides for question: According to your experience with Go-jek, is Go-jek’s fare proper to your travelled distance? Reached 79.29; question: Is Go-jek driver is keen to offer mask to you? Reached 71.12; question: Does Go-jek driver offer rain coat to you when rain downs? Reached 78.83 in “agree” scale. In the formation and examination of this study, using IPA model and Cartesius Analysis Quadrant, resulted that customer satisfaction in Go-jek service performance has been located in first and second quadrant, therefore second quadrant should be maintained and first quadrant should be improved to maximize Go-jek’s customer satisfaction. Keyword: service quality, Go-jek

A simplified dynamic model based on response factor method for thermal performance analysis of capillary radiant floors

Capillary radiant floors are much thinner and lighter than concrete radiant floors by embedding capillary tubes in insulation layers to supply heat, so that they can respond to temperature adjustment more quickly. In this study, a simplified dynamic model for the thermal performance analysis of capillary radiant floors was developed based on the response factor method. Next an experiment was conducted, the results of which were compared to results predicted by the model. The calculation results agree well with experimental data. The maximum deviation between the calculated surface temperatures and the experimental mean surface temperatures is 0.4 °C during the period of the starting and the maximum deviation of heat flux of the water side in the first minute is 8%, respectively. The developed dynamic model can calculate the surface temperature of the capillary tube and the heat fluxes of both the water side and the indoor air side directly if the water temperature and the indoor temperature are known regardless of whether they are constant or varied. The transient heat supplied by water and the heat stored in the floor can also be determined then. The model can be used to calculate the upward heat flow into the room above the floor and the downward heat flow into the room below the floor. At last, the effects of the surface layer and the tube spacing on surface temperature, heat flux, and time constants of capillary radiant floors are also analyzed by the simplified dynamic model.

Sampled Simulation of Task-Based Programs

Sampled simulation is a mature technique for reducing simulation time of single-threaded programs. Nevertheless, current sampling techniques do not take advantage of other execution models, like task-based execution, to provide both more accurate and faster simulation. Recent multi-threaded sampling techniques assume that the workload assigned to each thread does not change across multiple executions of a program. This assumption does not hold for dynamically scheduled task-based programming models. Task-based programming models allow the programmer to specify program segments as tasks which are instantiated many times and scheduled dynamically to available threads. Due to variation in scheduling decisions, two consecutive executions on the same machine typically result in different instruction streams processed by each thread. In this paper, we propose TaskPoint, a sampled simulation technique for dynamically scheduled task-based programs. We leverage task instances as sampling units and simulate only a fraction of all task instances in detail. Between detailed simulation intervals, we employ a novel fast-forwarding mechanism for dynamically scheduled programs. We evaluate different automatic techniques for clustering task instances and show that DBSCAN clustering combined with analytical performance modeling provides the best trade-off of simulation speed and accuracy. TaskPoint is the first technique combining sampled simulation and analytical modeling and provides a new way to trade off simulation speed and accuracy. Compared to detailed simulation, TaskPoint accelerates architectural simulation with 8 simulated threads by an average factor of 220x at an average error of 0.5 percent and a maximum error of 7.9 percent.

An analytical model of thermal performance for an eccentric heat source on a rectangular plate with double-sided convective cooling

Heat sources on rectangular plates with double-sided convective cooling are a common application of heat transfer in electronic packaging. However, it is considerably difficult to obtain analytical solutions for improving thermal performance because of the mixed boundary conditions on a plate’s upper surface. In this study, mixed boundary conditions were adjusted twice. Based on the method of separation of variables, a new analytical model was established and used to calculate the thermal performance of a random application case, including temperature distribution, thermal spreading resistance, and total thermal resistance. The results indicated that accurate analytical solutions can be obtained with few computing resources when more than 30 terms are used in each of single and double summations with a total mesh number of 196 in the heat source region. Subsequently, a series of thermal performances were calculated for random application cases and simulations were also conducted using the commercial software Icepak to offer a reference. The data comparisons revealed that the analytical model is able to calculate the temperature values at arbitrary point coordinates accurately over wide ranges of a plate’s thermal conductivity and heat transfer coefficient.

The Northern Caspian Sea: Consequences of climate change for level fluctuations during the Holocene

Quaternary Caspian Sea level variations depended on geophysical processes (affecting the opening and closing of gateways and basin size/shape) and hydro-climatological processes (affecting water balance). Disentangling the drivers of past Caspian Sea level variation, as well as the mechanisms by which they impacted the Caspian Sea level variation, is much debated. In this study we examine the relative impacts of hydroclimatic change, ice-sheet accumulation and melt, and isostatic adjustment on Caspian Sea level change. We performed model analysis of ice-sheet and hydroclimate impacts on Caspian Sea level and compared these with newly collated published palaeo-Caspian sea level data for the last glacial cycle. We used palaeoclimate model simulations from a global coupled ocean-atmosphere-vegetation climate model, HadCM3, and ice-sheet data from the ICE-6G_C glacial isostatic adjustment model. Our results show that ice-sheet meltwater during the last glacial cycle played a vital role in Caspian Sea level variations, which is in agreement with hypotheses based on palaeo-Caspian Sea level information. The effect was directly linked to the reorganization and expansion of the Caspian Sea palaeo-drainage system resulting from topographic change. The combined contributions from meltwater and runoff from the expanded basin area were primary factors in the Caspian Sea transgression during the deglaciation period between 20 and 15 kyr BP. Their impact on the evolution of Caspian Sea level lasted until around 13 kyr BP. Millennial scale events (Heinrich events and the Younger Dryas) negatively impacted the surface water budget of the Caspian Sea but their influence on Caspian Sea level variation was short-lived and was outweighed by the massive combined meltwater and runoff contribution over the expanded basin.

Analytical models of floating content in a vehicular urban environment

Among the many proposed opportunistic content sharing schemes, Floating Content (FC) is of special interest for the vehicular environment, not only for cellular traffic offloading, but also as a natural communication paradigm for location-based context-aware vehicular applications. Previously published results on the performance of vehicular FC have mostly focused on the conditions under which content persists over time in a given region of space, without addressing other important aspects of vehicular FC performance, such as the effectiveness with which content is replicated and made available, and the system conditions that enable good FC performance. This work presents a first analytical model of FC performance in vehicular networks in urban settings. It is based on a new synthetic mobility model (called District Mobility Model - DMM), and it does not require a detailed knowledge of the road grid geometry. We validate our model extensively, by comparison against numerical simulations based on real-world traces, and we prove our model accuracy under a variety of mobility patterns and traffic conditions. Our analytical and simulation results provide evidence of the effectiveness of theFC paradigm in realistic urban settings over a wide range of traffic conditions.

Prediction of adsorption properties for ionic and neutral pharmaceuticals and pharmaceutical intermediates on activated charcoal from aqueous solution via LFER model

In this study, we performed modeling analysis based on isothermal experimental data to develop linear free energy relationship (LFER) model, which is a very useful concept for predicting the molecular interaction of an adsorbate in a certain system. Thirty compounds containing cationic, anionic, and neutral were used, and the effect of diverse intermolecular interactions was captured by isotherm fitting of the Langmuir model. The predicted equation of distribution coefficient (Kd), adsorption capacity (qm), and adsorption affinity (Kad) of activated charcoal (AC) for compounds was developed. For the Kd values, an activity-dependent LFER model was developed, and could predict log Kd within R2 of 0.85. The determined system parameters and their statistical results revealed that the molecular volume (V) of compounds encouraged the adsorption from water to AC, whereas the hydrogen basicity (B) discouraged such adsorption. In the cases of qm and Kad, AC had relatively good capacities for neutrals, which were higher than those for cations and anions. Whereas, AC had stronger adsorption affinities with anions, compared to those for neutrals and cations. The phenomena were explained by the developed LFER models, in which the p−/n− electron interaction (E) made a great positive contribution to increase the adsorption capacity and affinity, while Coulombic interactions (J+ and J−) demonstrated a negative trend. Finally, the developed models can predict the qm and Kad in R2 of 0.86 and 0.76, respectively.

A Survey of Communication Performance Models for High-Performance Computing

This survey aims to present the state of the art in analytic communication performance models, providing sufficiently detailed descriptions of particularly noteworthy efforts. Modeling the cost of communications in computer clusters is an important and challenging problem. It provides insights into the design of the communication pattern of parallel scientific applications and mathematical kernels and sets a clear ground for optimization of their deployment in the increasingly complex high-performance computing infrastructure. The survey provides background information on how different performance models represent the underlying platform and shows the evolution of these models over time from early clusters of single-core processors to present-day multi-core and heterogeneous platforms. Prospective directions for future research in the area of analytic communication performance modeling conclude the survey.

A static analytical performance model for GPU kernel

Graphics processing units (GPUs) have shown increased popularity and play an important role as kind of coprocessor in heterogeneous co-processing environment. Heavily data parallel problems can be solved efficiently due to tens of thousands threads collaborative work in parallel GPU architecture. The achieved performance, therefore, depends on the capability of multiple threads in parallel collaboration. This paper, a static analytical kernel performance model (SAKP) was proposed to estimate the execution time of GPU kernel. Especially, a set of kernel and device features for target GPU is generated in the proposed model. We determine the performance limiting factors and generate an estimation of kernel execution time with this model. Matrix multiplication (MM) and histogram generation (HG) in NVIDIA GTX680 GPU card were performed to verify our proposed model and showed an absolute error in prediction less than 6.8%.

A static analytical performance model for GPU kernel

Graphics processing units (GPUs) have shown increased popularity and play an important role as kind of coprocessor in heterogeneous co-processing environment. Heavily data parallel problems can be solved efficiently due to tens of thousands threads collaborative work in parallel GPU architecture. The achieved performance, therefore, depends on the capability of multiple threads in parallel collaboration. This paper, a static analytical kernel performance model (SAKP) was proposed to estimate the execution time of GPU kernel. Especially, a set of kernel and device features for target GPU is generated in the proposed model. We determine the performance limiting factors and generate an estimation of kernel execution time with this model. Matrix multiplication (MM) and histogram generation (HG) in NVIDIA GTX680 GPU card were performed to verify our proposed model and showed an absolute error in prediction less than 6.8%.

Power Performance Tradeoffs Using Adaptive Bit Width Adjustments on Resistive Associative Processors

Associative processors, while ideal for vector-based applications, have been limited in their use mainly due to area and energy considerations. This reality is changing due to the advent of ultra-dense resistive memories (RRAMs), allowing for area efficient implementations of in-memory associative processors. However, with the high levels of integration, issues related to power density become the major bottleneck. In this paper, we investigate approximate computing in RRAM-based associative processors as a means of enhancing energy efficiency for in-memory associative processors. A case study of a wireless communication system is considered, where fast Fourier transform (FFT) modules are widely used. The proposed manager adjusts the bit width based on the channel SNR, aiming at achieving the desired performance at a reduced energy consumption. An analytical model of system performance under the reduced bit width induced noise is presented. Based on this model, an adaptive bit width adjustment algorithm is presented that utilizes the received SNR estimates to find the optimal bit width that achieves the desired performance goals. Simulation results show that the proposed algorithm can achieve up to 45% energy savings as compared to wireless communication systems using conventional RRAM FFT associative processor.

Moment-Based Importance Analysis of Structure Performance Functions Under Mixed Uncertainties

In this paper, a novel strategy for the importance analysis of structural performance models under both aleatory and epistemic uncertainties is presented. Random variables and fuzzy numbers are adopted for the representation of the two types of uncertainty respectively. Based on the statistical moments of the model outputs, two categories of importance measures are proposed under a hybrid framework composed of probability theory and fuzzy logic. The first category is for the input factors with aleatory uncertainty, while the other is for the input factors with epistemic uncertainty. In order to depict the credibility of the important measures of the random factors, a stability indicator is further introduced. Under the hybrid framework, the statistical moments of the performance outputs are fuzzy membership functions instead of deterministic values. Therefore, the importance measures are defined based on the area differences between conditional and unconditional membership functions. For the estimation of the proposed importance measures and stability indicator, a uniform discretization of the fuzzy membership function is first performed to combine the fuzzy factors with the random samples. Then, the Monte Carlo simulation (MCS) and Gorman and Seo’s three-point estimates (GSP) are employed as uncertainty propagation methods to address the statistical moments of the performance outputs. Finally, the proposed importance measures and stability indicator are studied through two numerical examples by MCS and GSP comparatively for demonstrating their benefits in stability, applicability, and efficiency.

Cluster Analysis of Customer Satisfaction at PT Esta Dana Ventura, Indonesia

The purpose of this study is to determine the level of customer satisfaction with service attributes at PT. Esta Dana Venture (physical evidence, reliability, responsiveness, assurance, and empathy). This study involved 220 respondents, where research involved PT Esta Dana Ventura consumers in the category of landing credit debtors. Data analysis methods use the Important Performance Analysis (IPA) model. Based on the results of primary data recapitulation, in general, consumers are quite satisfied with the services provided, where the number of the existing performance is greater than the expectation given by consumers. The results of the study show that the variables are the Empathy and Tangible are the first quadrants, which means very good, while the responsiveness and reliability are the second levels which mean they are not too optimal, while the assurance variable as in the third quadrant means that it is still bad. With the results of this study, it is expected that PT Esta Dana Ventura can further optimize its marketing strategy in order to compete with similar companies.

A Novel Dynamic Performance Analysis and Evaluation Model of Series-Parallel Connected Battery Pack for Electric Vehicles

Due to the low voltage and small capacity of a Li-ion battery cell, large numbers of cells are connected to construct a battery pack to satisfy the voltage and capacity requirement of the power system of an electric vehicle. Focusing on parallel and series connection mode of battery packs, the main contributions include the following. First, in order to increase the utilization rate of cells and enhance the performance of the battery pack, a method that makes the battery pack achieve their maximum initial capacity has been proposed. Second, a dynamic modeling and analysis method for the battery pack based on the equivalent circuit model has also been proposed. The results show that the battery pack in parallel and then in series has a better performance on charge/discharge capacity, efficiency, and utilization rate of cells. Third, the performance analysis and evaluation platform of the battery pack has been built, and the battery pack test results of different groups supported the conclusion of the simulation. The battery packs in parallel and then in series can achieve superior performance and greatly increase capacity and energy utilization.