CHAPTER 5 STAFFING LEVEL AND COST ANALYSES FOR SOFTWARE DEBUGGING ACTIVITIES THROUGH RATE- BASED SIMULATION APPROACHES
|
|
- Nora Simmons
- 8 years ago
- Views:
Transcription
1 101 CHAPTER 5 STAFFING LEVEL AND COST ANALYSES FOR SOFTWARE DEBUGGING ACTIVITIES THROUGH RATE- BASED SIMULATION APPROACHES 5.1 INTRODUCTION Many approaches have been given like rate based approaches for the reliability of the software and to analyze the reasons for the failure of the software. Criteria for the reliability of the software, number of debuggers or developers available are not taken into account. Newly detected faults have to wait for some time since all the debuggers will be busy in detecting the faults which they found previously. Time taken to remove the fault is taken into consideration and the main fact relies in it is that less number of faults been removed when compared to the number of faults detected. 5.2 IMPORTANCE OF DEBUGGING Debugging is the process of locating or fixing the error in software development coding or hardware. Debugging performance is mainly based on debugger s performance. So the debugger behavior should be analysed.
2 QUEUING SYSTEM Queuing system has important three characteristics. The characteristics are as follows: Arrival Process : The probability density distribution that determines the fault arrivals in the system. Service Process : The probability density distribution that determines the fault correction process. Number of Servers : Number of servers available in the correction process. 5.4 PROPOSED ARCHITECTURE This approach proposes a rate-based simulation method incorporated with the queuing theory for the debugging behavior during the development of the software. G/G/ and G/G/m models have been used in our proposed method. This method is used for the real software failure. This approach helps to predict the debuggers performance and the cost effectiveness. In the figure 5.1, the proposed simulated model analyzed the influence on the performance, and the cost, related to software debugging when the number of allocated debuggers changes. It is important information to guide the project manager in the estimation and adjustment of the staffing needs and cost benefit for debugging the systems.
3 Figure 5.1 Increasing Debugging activities through Queuing Theory 103
4 Simulation-Based Approaches Although productive research has used neural network approaches to predict the increment of the failure process, most training algorithms for neural network approaches suffer from the overfitting problem. That is, the fitting bias of the training set is very slight regarding known data, but the bias is unpredictably large when new data are presented to the network. Determination of the proper number of neurons is another common problem in the field of neural network research. Moreover, Tausworthe and Lyu (2004) argued that most SRGM only focus on the failure observation during the test phase, or the operational phase. They reported that the assumptions of most SRG lead to the over-simplification of the failure process. Thus, general simulation techniques have been developed to relax certain unreasonable assumptions. For ease of discussion, we let {N(t), t 0be the stochastic failure process that represents the number of failures observed in an execution interval (0, t). The failure behavior is modeled by a failure rate, N(t). That is, the behavior of the stochastic process purely depends on the rate function for each state of the software system. The state is represented by the number of occurrences of the event. Tausworthe and Lyu (2004) proposed a set of simulation procedures. The simulation algorithm can be applied to each individual activity during SDLC. Later, Gokhale and Lyu (2004) proposed a simulation technique to analyze structure-based software reliability. They believed that the time required by fault repair should be considered explicitly. They also extended the simulation to the reliability assessment on the application level. Recently, Gokhale et al (2004). Further considered the possibility of imperfect debugging in the
5 105 simulation approach Gokhale and Lyu (2005). However, the existing system published simulation techniques seldom consider the limitations of debugging resources, and this oversight may not be reasonable. In practice, the number of qualified debuggers will be controlled during SDLC Inoue and Yamada (2003). In the next section, System uses queuing theory to model software fault correction activities through simulation procedures Simulation Procedures with Queuing Management Mechanism The FDP, and FCP are related but distinct. If {H(t),t 0denotes the stochastic process representing the number of faults corrected during (0,t), then H(t) N(t). The output of the fault identification system is the input to the fault correction system. The amount of lag is tightly related to the number of personnel dedicated to the fault correction activities. The related activities of testing and debugging are depicted. In the past, some researchers Gokhale (2007) have attempted to predict software debugging behavior using a queueing model. Musa reported that there are three resources that govern the pace of debugging, and each resource represents an important part of the software debugging process. He used a type of queueing model in the FCP. The queueing constraints on each debugger will be modeled as a classical queueing system with random arrivals, and random service. Both the identification, and correction of faults are considered to be time-invariant Poisson processes Huang et al (2003) in calendar time. Procedures 1, and 2, were developed using a Java programming language, and are discussed in detail.
6 Queuing and Reliability Concepts A general description of the queuing system Kanoun and Laprie (1994) can be given as customers arriving for service, waiting for service if it is not immediate, and leaving the system after being served. As seen from Figure. 5.1, if a queue model is used to sketch the debugging system, the intensity function of N(t), i.e., (t), can be viewed as the arrival rate of the queueing system. To estimate and predict (t), apply SRGM to calculate the failure intensity. Further, the probabilities for the increments of the failure process N(t) can be assigned as N(t+ t) N(t) = 0 with probability 1 (t) t + o( t) 1 with probability (t) t + o( t) 2 with probability o( t) (5.1) where the function o( t) is defined o( t)/ t=0, Limit t 0 This function indicates that the probability of multiple failures during a time interval (t, t) is negligible for small values of t. On the other hand, the number of service channels can be considered as the number of allocated debuggers, and the service rate can be used to depict a debugger s efficiency in the debugging system. The negative exponential distribution is a common assumption for service time, and it has the probability function P(0 T s t s ) = G(t s ) = 1-e -µ t s, t s 0 (5.2)
7 107 Also, P(T s > t s ) = 1- G(t s ) = e -µ t s (5.3) Thus, the probability that a debugger finishes the fault correction in the time interval (t s,t s t s, is t), given that it has already been in progress for time P(t s T s t s t T s > t s ) = = µ t (5.4) Procedure 1 Queueing and software reliability theories are incorporated to simulate FDP, and FCP. To simplify the problem, the constraint on the number of available debuggers is first constrained. void Simulation_Procedure (double stop_time, double dt) { double current_time = 0; int working_server = 0, max_server = 0; struct fault_info correction[max_size]; int cumulative_arrival = 0, cumulative_departure = 0; while ( current_time < stop_time) { Figure 5.2 (Procedure1 Continued)
8 108 DETECTING: if( occur() ){ working_server++; if (working_server > max_server) max_server = working_server; correction[cumulative_arrival].arrival_time = current_time; correction[cumulative_arrival++].state = CORRECTING; CORRECTING: for(int i = 0; i < cumulative_arrival; i++) { if (correction[i].state == CORRECTING && leave(current_time - correction[i].arrival_time)) { working_server--; correction[i].departure_time= current_time' correction[i].state = CORRECTED; cumulative_departure++; current_time += dt; Figure 5.2 Procedure 1
9 Procedure 2 void Simulation_Procedure (double stop_time, double dt, int staffing_level) { double current_time = 0; int working_server = 0; struct fault_info correction[max_size], waiting_queue{max_size]; int num_correction = 0,cumulative_arrival = 0, cumulative_departure =0; int queue_head = 0, queue_tail = 0; while (current_time < stop_time) { Figure 5.3 (Continued)
10 110 ALLOCATION: while (working_server < staffing_level && queue_head!=queue_tail) { waiting_queue[queue_head].state = OUT_OF_QUEUE; current_time; waiting_queue[queue_head++].departure_time = correction[num_correction].state = CORRECTING; DETECTING: if(occur()){ cumulative_arrival++; if (working_server >= staffing_level){ waiting_queue[queue_tail].state = ENQUEUE; waiting_queue[queue_tail++].arrival_time = current_time; else { working_server++; Figure 5.3 (Continued)
11 111 if (working_server > max_server) max_server = working_server; correction[num_correction].arrival_time = current_time; correction[num_correction++].state = CORRECTING; CORRECTING: for (int i=0; i<num_correction; i++) { if (correction[i].state == CORRECTING && leave(current_time - correction[i].arrival_time)) { working_server--; correction[i].departure_time = current_time; correction[i].state = CORRECTED; cumulative_departure++; current_time += dt; Figure 5.3 Procedure 2
12 112 The following assumptions are made: 1) The software system is subject to failures at random times caused by the manifestation of the remaining faults in the system. 2) All faults are independent, and equally detectable. The probability that a failure will be experienced during (t, t+ t) is approximately, and the probability that two or more failures will occur during (t, t+ t) is negligible. 3) The correction of faults takes non-negligible time, i.e., explicit repair. The probability that a fault is corrected in time interval (t s, t s t) is µ t. Further, fault removals do not affect the ongoing activities of fault detection. 4) No new faults are introduced during the correction process. 5) Available, and qualified debuggers are always sufficient. The debugging system is modeled by a queue system (G/G/ ). Each time a failure occurs, there is no lag to allot a debugger to the detected fault. Based on these assumptions, Procedure 1 was developed, and is depicted in Figure Procedure 1 accepts two parameters as inputs: the total number of execution time units, defined as stop_time; and the consumed time of each run, denoted by dt. The length of each time unit should be consistent with the failure data collection form. Further, each time unit is divided into a large number of runs, and the length of each run should be short enough that multiple events in a run are rare. That is, the variations of failure
13 113 rate in (t,t+dt) should be insignificant. In addition to the two inputs, certain variables are also used in the simulation to represent the major components of the debugging system, and to gather useful statistics. The variable current_time represents a clock, which also indicates the cumulative execution time to the present. It increases with the completion of each run. As this variable s value is increased, the simulator updates the state of the debugging system to reflect the activities of each component. Construct a structure, fault_info, which is given as struct fault_info { arrival_time; // the time to occupy a resource departure_time; // the time to release a resource state; // the current status of the fault The array correction, each element of which contains a fault_info, is used to keep track of the status of each fault. Further, working_server denotes the current number of busy debuggers, while max_server logs the number of utilized debuggers at peak time. Finally, cumulative_arrival, and cumulative_departure are integers used to count the numbers of cumulative detected, faults and cumulative removed faults, respectively. There are several ways to derive the simulation. This framework adopts a random-number generator, which is common. Following mathematical probability distributions, the generator is programmed to
14 114 generate arrivals, departures, and so on. During simulation, actions taken in each run consist of two steps: detecting, and correcting Detecting Process Following this FDP Gokhale et al (2006). At the outset of each run, the function occur () will be invoked to determine whether the testers detect a fault in this run. From the procedure, the probability that a failure occurs during (t,t+dt) is approximately (t) dt. Hence, each time the occur function is called, a (0.0, 1.0)-uniform random number x will be generated, and compared with (t) dt. This means that the testers may detect a fault if x is greater than (t) dt. Once the occur function returns 1, working_server is increased, the value of max_server is updated, and the state of the detected fault will be recorded. Lines in Procedure 1 show the activities taking place as a result of each failure occurrence Correcting Process Departing from the detection step, it commences diagnosing the status of each detected fault by checking all elements of the array correction Hu et al (2007). If an open-remaining fault (a detected but uncorrected fault) is found, the function leave(t s ) determines whether this fault will be corrected in this run. Similar to the occur function, the success of fault removal relies upon the comparison between µ dt, and the random number y. If y > µ dt, then the dedicated debugger successfully corrects this fault in this run. The necessary actions taken due to this successful repair are given in Lines Otherwise, this fault cannot be corrected at this time, and will be re-examined in the next run. From Assumption 3), we know that the necessary correction
15 115 time is non-negligible. Hence, the return of leave (0) is given as 0, ensuring that the fault detected in the current run will not be removed immediately. The above two steps will be repeated until the end of the entire interval, when current_time = stop_time. After simulation, the total number of detected faults, the cumulative number of corrected faults, the maximum number of utilized debuggers, and the FDP are obtained. The FCP can also further be described. During software testing and debugging, project managers can collect all failure data prior to the arbitrary time t i, estimate the failure rate posterior to t i, exercise the simulation at time t i, and predict possible behaviors in the future, i.e., the behaviors in the time period t(t i ). The frequency of making a prediction (i.e.,how far ahead the prediction should be made) depends upon the policies of management. 5.5 RESULTS The first data set (DS1) Musa (1975) was from system T1 of the Rome Air Development Center project. The failure data were carefully collected under strict supervision. System T1 was applied to real time command and control, including 21,700 delivered object instructions. Over the course of 21 weeks, 9 programmers detected and removed 136 faults.
16 116 Table 5.1 Debugger s utilization Limitation of Available Debuggers Throughput (by the end of 21 weeks) Openremaining Avg. removals faults Time to remove faults (weeks) 2,3 Avg. waiting time (weeks) 3 Avg. response Avg. queue time lgth 3 (weeks) 3 Debugger utilization (%) Unlimited * * * * * * * The average of fault removals per week. 2 The faults detected during the period of 21 weeks. 3 The average measure of all faults. Note that the waiting time means the time of the fault pending in waiting queue, and the response time indicates the total time spent in the queuing system. Besides, because some detected faults may not be removed yet at the end of 21 weeks Huang et al (2006), the simulation of correction processes, and the statistics are continued until all pending faults are addressed.
17 117 Figure 5.4 Performance comparisons between different staffing levels as the value of varies from 1 to 8
18 Debuger's salary Penalties of Open-Remaining faults Number of Debuggers Figure 5.5 Relationship between debuggers and expected cost According to the above analyses, the limitation on the debugging team size is not the bottleneck to enhance performance when the number of personnel is more than 7. Increasing the number of debuggers can only increase the number of simultaneous working debuggers, but it does not reduce the consumed time taken for a debugger to fix a fault. Therefore, when the staffing level can bear the load, more debuggers cannot improve the throughput. If the manager wants to ameliorate the performance, it is necessary to improve the debuggers skills, i.e. to increase the value of Figure 5.4 shows some performance comparisons between different staffing levels as the value of varies from 1 to 8.The utilization of 6 debuggers is very low at the beginning. However, because the number of detected faults grows rapidly from the 8th week to the 18th week, 6 personnel seem unable to bear the load. Re-staffing may be reasonable in both conditions. Using the pro- posed framework, project managers can easily estimate the influence
19 119 caused by re-staffing the debugging system, and further decide whether to make an adjustment. In practice, the resources of a project are often estimated by expert judgment. That is, several experts on software development techniques estimate the needs individually, consult with each other, and reach an agreed estimate. When the experts discuss the allocation of staff to specific Work Breakdown Structure tasks, they may also benefit from the outputs of the proposed approaches. Moreover, the information on completed projects (the source cases) is valuable as input for preliminary planning of a similar new project (the target case). This common technique is called estimation by analogy (also called case-based reasoning). For the completed projects, we can also use the developed framework to retrace and reconstruct the possible debugging processes under different amounts of debugging personnel. Such analyses provide guidance for estimating the staffing needs of new projects with similar characteristics. 5.6 CONCLUSION The proposed framework helps to understand about the on-going project and also used to infer the current and future situations in the project or reconstruct the project. The applications of the proposed procedures are explained through two real data sets. This studies show that the proposed simulation procedures can analyze the influence on the performance, and the cost related to software debugging when the number of allocated debuggers changes. This useful, important information can guide project managers in the estimation and adjustment of the staffing needs for debugging systems.
Supplement to Call Centers with Delay Information: Models and Insights
Supplement to Call Centers with Delay Information: Models and Insights Oualid Jouini 1 Zeynep Akşin 2 Yves Dallery 1 1 Laboratoire Genie Industriel, Ecole Centrale Paris, Grande Voie des Vignes, 92290
More informationStochastic Processes and Queueing Theory used in Cloud Computer Performance Simulations
56 Stochastic Processes and Queueing Theory used in Cloud Computer Performance Simulations Stochastic Processes and Queueing Theory used in Cloud Computer Performance Simulations Florin-Cătălin ENACHE
More informationSimulation Software 1
Simulation Software 1 Introduction The features that should be programmed in simulation are: Generating random numbers from the uniform distribution Generating random variates from any distribution Advancing
More informationProposed Pricing Model for Cloud Computing
Computer Science and Information Technology 2(4): 211-218, 2014 DOI: 10.13189/csit.2014.020405 http://www.hrpub.org Proposed Pricing Model for Cloud Computing Muhammad Adeel Javaid Member Vendor Advisory
More informationWaiting Times Chapter 7
Waiting Times Chapter 7 1 Learning Objectives Interarrival and Service Times and their variability Obtaining the average time spent in the queue Pooling of server capacities Priority rules Where are the
More informationCHAPTER 3 CALL CENTER QUEUING MODEL WITH LOGNORMAL SERVICE TIME DISTRIBUTION
31 CHAPTER 3 CALL CENTER QUEUING MODEL WITH LOGNORMAL SERVICE TIME DISTRIBUTION 3.1 INTRODUCTION In this chapter, construction of queuing model with non-exponential service time distribution, performance
More informationModelling the performance of computer mirroring with difference queues
Modelling the performance of computer mirroring with difference queues Przemyslaw Pochec Faculty of Computer Science University of New Brunswick, Fredericton, Canada E3A 5A3 email pochec@unb.ca ABSTRACT
More informationNetwork Design Performance Evaluation, and Simulation #6
Network Design Performance Evaluation, and Simulation #6 1 Network Design Problem Goal Given QoS metric, e.g., Average delay Loss probability Characterization of the traffic, e.g., Average interarrival
More informationHow To Manage A Call Center
THE ROLE OF SIMULATION IN CALL CENTER MANAGEMENT Roger Klungle AAA Michigan Introduction With recent advances in technology and the changing nature of business, call center management has become a rapidly
More informationQuantitative Analysis of Cloud-based Streaming Services
of Cloud-based Streaming Services Fang Yu 1, Yat-Wah Wan 2 and Rua-Huan Tsaih 1 1. Department of Management Information Systems National Chengchi University, Taipei, Taiwan 2. Graduate Institute of Logistics
More informationUNIT 2 QUEUING THEORY
UNIT 2 QUEUING THEORY LESSON 24 Learning Objective: Apply formulae to find solution that will predict the behaviour of the single server model II. Apply formulae to find solution that will predict the
More informationEstimating the Degree of Activity of jumps in High Frequency Financial Data. joint with Yacine Aït-Sahalia
Estimating the Degree of Activity of jumps in High Frequency Financial Data joint with Yacine Aït-Sahalia Aim and setting An underlying process X = (X t ) t 0, observed at equally spaced discrete times
More informationSimple Queuing Theory Tools You Can Use in Healthcare
Simple Queuing Theory Tools You Can Use in Healthcare Jeff Johnson Management Engineering Project Director North Colorado Medical Center Abstract Much has been written about queuing theory and its powerful
More informationSoftware reliability improvement with quality metric and defect tracking
Software reliability improvement with quality metric and defect tracking Madhavi Mane 1, Manjusha Joshi 2, Prof. Amol Kadam 3, Prof. Dr. S.D. Joshi 4, 1 M.Tech Student, Computer Engineering Department
More informationCHAPTER 6: Continuous Uniform Distribution: 6.1. Definition: The density function of the continuous random variable X on the interval [A, B] is.
Some Continuous Probability Distributions CHAPTER 6: Continuous Uniform Distribution: 6. Definition: The density function of the continuous random variable X on the interval [A, B] is B A A x B f(x; A,
More informationBasic Queuing Relationships
Queueing Theory Basic Queuing Relationships Resident items Waiting items Residence time Single server Utilisation System Utilisation Little s formulae are the most important equation in queuing theory
More informationMuse Server Sizing. 18 June 2012. Document Version 0.0.1.9 Muse 2.7.0.0
Muse Server Sizing 18 June 2012 Document Version 0.0.1.9 Muse 2.7.0.0 Notice No part of this publication may be reproduced stored in a retrieval system, or transmitted, in any form or by any means, without
More informationTraffic Behavior Analysis with Poisson Sampling on High-speed Network 1
Traffic Behavior Analysis with Poisson Sampling on High-speed etwork Guang Cheng Jian Gong (Computer Department of Southeast University anjing 0096, P.R.China) Abstract: With the subsequent increasing
More informationAS-D2 THE ROLE OF SIMULATION IN CALL CENTER MANAGEMENT. Dr. Roger Klungle Manager, Business Operations Analysis
AS-D2 THE ROLE OF SIMULATION IN CALL CENTER MANAGEMENT Dr. Roger Klungle Manager, Business Operations Analysis AAA Michigan 1 Auto Club Drive Dearborn, MI 48126 U.S.A. Phone: (313) 336-9946 Fax: (313)
More informationEvaluation of Complexity of Some Programming Languages on the Travelling Salesman Problem
International Journal of Applied Science and Technology Vol. 3 No. 8; December 2013 Evaluation of Complexity of Some Programming Languages on the Travelling Salesman Problem D. R. Aremu O. A. Gbadamosi
More informationNonparametric adaptive age replacement with a one-cycle criterion
Nonparametric adaptive age replacement with a one-cycle criterion P. Coolen-Schrijner, F.P.A. Coolen Department of Mathematical Sciences University of Durham, Durham, DH1 3LE, UK e-mail: Pauline.Schrijner@durham.ac.uk
More informationJustifying Simulation. Why use simulation? Accurate Depiction of Reality. Insightful system evaluations
Why use simulation? Accurate Depiction of Reality Anyone can perform a simple analysis manually. However, as the complexity of the analysis increases, so does the need to employ computer-based tools. While
More information4 The M/M/1 queue. 4.1 Time-dependent behaviour
4 The M/M/1 queue In this chapter we will analyze the model with exponential interarrival times with mean 1/λ, exponential service times with mean 1/µ and a single server. Customers are served in order
More informationAPPENDIX 1 USER LEVEL IMPLEMENTATION OF PPATPAN IN LINUX SYSTEM
152 APPENDIX 1 USER LEVEL IMPLEMENTATION OF PPATPAN IN LINUX SYSTEM A1.1 INTRODUCTION PPATPAN is implemented in a test bed with five Linux system arranged in a multihop topology. The system is implemented
More informationQueuing Theory. Long Term Averages. Assumptions. Interesting Values. Queuing Model
Queuing Theory Queuing Theory Queuing theory is the mathematics of waiting lines. It is extremely useful in predicting and evaluating system performance. Queuing theory has been used for operations research.
More informationCommon Data Structures
Data Structures 1 Common Data Structures Arrays (single and multiple dimensional) Linked Lists Stacks Queues Trees Graphs You should already be familiar with arrays, so they will not be discussed. Trees
More informationDepartment of Mathematics, Indian Institute of Technology, Kharagpur Assignment 2-3, Probability and Statistics, March 2015. Due:-March 25, 2015.
Department of Mathematics, Indian Institute of Technology, Kharagpur Assignment -3, Probability and Statistics, March 05. Due:-March 5, 05.. Show that the function 0 for x < x+ F (x) = 4 for x < for x
More informationASSESSING AIRPORT PASSENGER SCREENING PROCESSING SYSTEMS
ASSESSING AIRPORT PASSENGER SCREENING PROCESSING SYSTEMS SAEID SAIDI, PHD CANDIDATE, SCHULICH SCHOOL OF ENGINEERING, UNIVERSITY OF CALGARY, CANADA (EMAIL: SSAIDI@UCALGARY.CA) DR. ALEXANDRE DE BARROS, ASSOCIATE
More informationExponential Distribution
Exponential Distribution Definition: Exponential distribution with parameter λ: { λe λx x 0 f(x) = 0 x < 0 The cdf: F(x) = x Mean E(X) = 1/λ. f(x)dx = Moment generating function: φ(t) = E[e tx ] = { 1
More informationDiscrete-Event Simulation
Discrete-Event Simulation Prateek Sharma Abstract: Simulation can be regarded as the emulation of the behavior of a real-world system over an interval of time. The process of simulation relies upon the
More informationProcess simulation. Enn Õunapuu enn.ounapuu@ttu.ee
Process simulation Enn Õunapuu enn.ounapuu@ttu.ee Content Problem How? Example Simulation Definition Modeling and simulation functionality allows for preexecution what-if modeling and simulation. Postexecution
More informationChange Management in Enterprise IT Systems: Process Modeling and Capacity-optimal Scheduling
Change Management in Enterprise IT Systems: Process Modeling and Capacity-optimal Scheduling Praveen K. Muthusamy, Koushik Kar, Sambit Sahu, Prashant Pradhan and Saswati Sarkar Rensselaer Polytechnic Institute
More informationSoftware Process for QA
Software Process for QA Basic approaches & alternatives CIS 610, W98 / M Young 1/7/98 1 This introduction and overview is intended to provide some basic background on software process (sometimes called
More informationHow to Plan a Successful Load Testing Programme for today s websites
How to Plan a Successful Load Testing Programme for today s websites This guide introduces best practise for load testing to overcome the complexities of today s rich, dynamic websites. It includes 10
More informationPull versus Push Mechanism in Large Distributed Networks: Closed Form Results
Pull versus Push Mechanism in Large Distributed Networks: Closed Form Results Wouter Minnebo, Benny Van Houdt Dept. Mathematics and Computer Science University of Antwerp - iminds Antwerp, Belgium Wouter
More informationSimulation of a Claims Call Center: A Success and a Failure
Proceedings of the 1999 Winter Simulation Conference P. A. Farrington, H. B. Nembhard, D. T. Sturrock, and G. W. Evans, eds. SIMULATION OF A CLAIMS CALL CENTER: A SUCCESS AND A FAILURE Roger Klungle AAA
More informationCloud Enabled Emergency Navigation Using Faster-than-real-time Simulation
Cloud Enabled Emergency Navigation Using Faster-than-real-time Simulation Huibo Bi and Erol Gelenbe Intelligent Systems and Networks Group Department of Electrical and Electronic Engineering Imperial College
More informationWindows Server Performance Monitoring
Spot server problems before they are noticed The system s really slow today! How often have you heard that? Finding the solution isn t so easy. The obvious questions to ask are why is it running slowly
More informationComputer Science, Informatik 4 Communication and Distributed Systems. Simulation. Discrete-Event System Simulation. Dr.
Simulation Discrete-Event System Simulation Chapter 3 General Principles General Principles Introduction Framework for modeling systems by discrete-event simulation A system is modeled in terms of its
More informationPerformance Modeling and Analysis of a Database Server with Write-Heavy Workload
Performance Modeling and Analysis of a Database Server with Write-Heavy Workload Manfred Dellkrantz, Maria Kihl 2, and Anders Robertsson Department of Automatic Control, Lund University 2 Department of
More informationMobile App Design Project #1 Java Boot Camp: Design Model for Chutes and Ladders Board Game
Mobile App Design Project #1 Java Boot Camp: Design Model for Chutes and Ladders Board Game Directions: In mobile Applications the Control Model View model works to divide the work within an application.
More informationAppendix: Simple Methods for Shift Scheduling in Multi-Skill Call Centers
MSOM.1070.0172 Appendix: Simple Methods for Shift Scheduling in Multi-Skill Call Centers In Bhulai et al. (2006) we presented a method for computing optimal schedules, separately, after the optimal staffing
More informationPerformance testing as a full life cycle activity. Julian Harty
Performance testing as a full life cycle activity Julian Harty Julian Harty & Stuart Reid 2004 Scope of Performance Performance What is performance testing? Various views 3 outcomes 3 evaluation techniques
More informatione.g. arrival of a customer to a service station or breakdown of a component in some system.
Poisson process Events occur at random instants of time at an average rate of λ events per second. e.g. arrival of a customer to a service station or breakdown of a component in some system. Let N(t) be
More informationBasic Multiplexing models. Computer Networks - Vassilis Tsaoussidis
Basic Multiplexing models? Supermarket?? Computer Networks - Vassilis Tsaoussidis Schedule Where does statistical multiplexing differ from TDM and FDM Why are buffers necessary - what is their tradeoff,
More informationReliability of a Commercial Telecommunications System
Reliability of a Commercial Telecommunications System Mohamed Kaâniche and Karama Kanoun LAAS-CNRS 7, Avenue du Colonel Roche 77 Toulouse, France Abstract We analyze data collected on a commercial telecommunications
More informationScheduling Policies, Batch Sizes, and Manufacturing Lead Times
To appear in IIE Transactions Scheduling Policies, Batch Sizes, and Manufacturing Lead Times Saifallah Benjaafar and Mehdi Sheikhzadeh Department of Mechanical Engineering University of Minnesota Minneapolis,
More informationDistributionally robust workforce scheduling in call centers with uncertain arrival rates
Distributionally robust workforce scheduling in call centers with uncertain arrival rates S. Liao 1, C. van Delft 2, J.-P. Vial 3,4 1 Ecole Centrale, Paris, France 2 HEC. Paris, France 3 Prof. Emeritus,
More informationPROJECT RISK MANAGEMENT
11 PROJECT RISK MANAGEMENT Project Risk Management includes the processes concerned with identifying, analyzing, and responding to project risk. It includes maximizing the results of positive events and
More information1. Simulation of load balancing in a cloud computing environment using OMNET
Cloud Computing Cloud computing is a rapidly growing technology that allows users to share computer resources according to their need. It is expected that cloud computing will generate close to 13.8 million
More informationMicrosoft HPC. V 1.0 José M. Cámara (checam@ubu.es)
Microsoft HPC V 1.0 José M. Cámara (checam@ubu.es) Introduction Microsoft High Performance Computing Package addresses computing power from a rather different approach. It is mainly focused on commodity
More informationAn Introduction To Simple Scheduling (Primarily targeted at Arduino Platform)
An Introduction To Simple Scheduling (Primarily targeted at Arduino Platform) I'm late I'm late For a very important date. No time to say "Hello, Goodbye". I'm late, I'm late, I'm late. (White Rabbit in
More informationFAULT TOLERANCE FOR MULTIPROCESSOR SYSTEMS VIA TIME REDUNDANT TASK SCHEDULING
FAULT TOLERANCE FOR MULTIPROCESSOR SYSTEMS VIA TIME REDUNDANT TASK SCHEDULING Hussain Al-Asaad and Alireza Sarvi Department of Electrical & Computer Engineering University of California Davis, CA, U.S.A.
More informationArena 9.0 Basic Modules based on Arena Online Help
Arena 9.0 Basic Modules based on Arena Online Help Create This module is intended as the starting point for entities in a simulation model. Entities are created using a schedule or based on a time between
More informationDetection of changes in variance using binary segmentation and optimal partitioning
Detection of changes in variance using binary segmentation and optimal partitioning Christian Rohrbeck Abstract This work explores the performance of binary segmentation and optimal partitioning in the
More informationCROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING
CHAPTER 6 CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING 6.1 INTRODUCTION The technical challenges in WMNs are load balancing, optimal routing, fairness, network auto-configuration and mobility
More informationA Uniform Asymptotic Estimate for Discounted Aggregate Claims with Subexponential Tails
12th International Congress on Insurance: Mathematics and Economics July 16-18, 2008 A Uniform Asymptotic Estimate for Discounted Aggregate Claims with Subexponential Tails XUEMIAO HAO (Based on a joint
More informationChapter 3 ATM and Multimedia Traffic
In the middle of the 1980, the telecommunications world started the design of a network technology that could act as a great unifier to support all digital services, including low-speed telephony and very
More informationOptimal parameter choice in modeling of ERP system reliability
Proceedings of the 22nd Central European Conference on Information and Intelligent Systems 365 Optimal parameter choice in modeling of ERP system reliability Frane Urem, Želimir Mikulić Department of management
More informationMinimize Wait Time and Improve the Waiting Experience
Improving the Customer Experience Minimize Wait Time and Improve the Waiting Experience www.lavi.com (888) 285-8605 Overview Waiting lines easily become the source of tension between customers and businesses
More informationTians Scheduling: Using Partial Processing in Best-Effort Applications
Tians Scheduling: Using Partial Processing in Best-Effort Applications Yuxiong He, Sameh Elnikety Microsoft Research Hongyang Sun Nanyang Technological University Abstract To service requests with high
More informationLoad Balancing and Switch Scheduling
EE384Y Project Final Report Load Balancing and Switch Scheduling Xiangheng Liu Department of Electrical Engineering Stanford University, Stanford CA 94305 Email: liuxh@systems.stanford.edu Abstract Load
More informationC. Wohlin, M. Höst, P. Runeson and A. Wesslén, "Software Reliability", in Encyclopedia of Physical Sciences and Technology (third edition), Vol.
C. Wohlin, M. Höst, P. Runeson and A. Wesslén, "Software Reliability", in Encyclopedia of Physical Sciences and Technology (third edition), Vol. 15, Academic Press, 2001. Software Reliability Claes Wohlin
More informationA Model-driven Approach to Predictive Non Functional Analysis of Component-based Systems
A Model-driven Approach to Predictive Non Functional Analysis of Component-based Systems Vincenzo Grassi Università di Roma Tor Vergata, Italy Raffaela Mirandola {vgrassi, mirandola}@info.uniroma2.it Abstract.
More informationPerformance Evaluation of Software Architectures
Performance Evaluation of Software Architectures Exercises José Costa Software for Embedded Systems Departamento de Engenharia Informática (DEI) Instituto Superior Técnico 2015-03-17 José Costa (DEI/IST)
More informationSoftware Project Management Matrics. Complied by Heng Sovannarith heng_sovannarith@yahoo.com
Software Project Management Matrics Complied by Heng Sovannarith heng_sovannarith@yahoo.com Introduction Hardware is declining while software is increasing. Software Crisis: Schedule and cost estimates
More informationChi Square Tests. Chapter 10. 10.1 Introduction
Contents 10 Chi Square Tests 703 10.1 Introduction............................ 703 10.2 The Chi Square Distribution.................. 704 10.3 Goodness of Fit Test....................... 709 10.4 Chi Square
More information1: B asic S imu lati on Modeling
Network Simulation Chapter 1: Basic Simulation Modeling Prof. Dr. Jürgen Jasperneite 1 Contents The Nature of Simulation Systems, Models and Simulation Discrete Event Simulation Simulation of a Single-Server
More informationCS 451 Software Engineering Winter 2009
CS 451 Software Engineering Winter 2009 Yuanfang Cai Room 104, University Crossings 215.895.0298 yfcai@cs.drexel.edu 1 Testing Process Testing Testing only reveals the presence of defects Does not identify
More informationFunctional Optimization Models for Active Queue Management
Functional Optimization Models for Active Queue Management Yixin Chen Department of Computer Science and Engineering Washington University in St Louis 1 Brookings Drive St Louis, MO 63130, USA chen@cse.wustl.edu
More informationTelephone Traffic Queues in a Customer Call Center. A Thesis. Presented for the. Master of Science Degree. The University of Tennessee at Chattanooga
Telephone Traffic Queues in a Customer Call Center A Thesis Presented for the Master of Science Degree The University of Tennessee at Chattanooga Patrick Todd April 2009 Copyright 2009 by Patrick Todd
More informationComputing Load Aware and Long-View Load Balancing for Cluster Storage Systems
215 IEEE International Conference on Big Data (Big Data) Computing Load Aware and Long-View Load Balancing for Cluster Storage Systems Guoxin Liu and Haiying Shen and Haoyu Wang Department of Electrical
More informationPortfolio Using Queuing Theory
Modeling the Number of Insured Households in an Insurance Portfolio Using Queuing Theory Jean-Philippe Boucher and Guillaume Couture-Piché December 8, 2015 Quantact / Département de mathématiques, UQAM.
More informationCSE373: Data Structures and Algorithms Lecture 3: Math Review; Algorithm Analysis. Linda Shapiro Winter 2015
CSE373: Data Structures and Algorithms Lecture 3: Math Review; Algorithm Analysis Linda Shapiro Today Registration should be done. Homework 1 due 11:59 pm next Wednesday, January 14 Review math essential
More informationLoad balancing model for Cloud Data Center ABSTRACT:
Load balancing model for Cloud Data Center ABSTRACT: Cloud data center management is a key problem due to the numerous and heterogeneous strategies that can be applied, ranging from the VM placement to
More informationModeling Stochastic Inventory Policy with Simulation
Modeling Stochastic Inventory Policy with Simulation 1 Modeling Stochastic Inventory Policy with Simulation János BENKŐ Department of Material Handling and Logistics, Institute of Engineering Management
More informationWhat Is Specific in Load Testing?
What Is Specific in Load Testing? Testing of multi-user applications under realistic and stress loads is really the only way to ensure appropriate performance and reliability in production. Load testing
More informationAutomated Test Generation
Automated Test Generation (From a Behavioral Model) James M. Clarke Lucent Technologies 2000 Naperville Road Naperville, IL 60666-7033 (630) 979-1861 jmclarke@lucent.com Abstract The challenge for testers:
More informationAutomated Scheduling Methods. Advanced Planning and Scheduling Techniques
Advanced Planning and Scheduling Techniques Table of Contents Introduction 3 The Basic Theories 3 Constrained and Unconstrained Planning 4 Forward, Backward, and other methods 5 Rules for Sequencing Tasks
More informationHW3: Programming with stacks
HW3: Programming with stacks Due: 12PM, Noon Thursday, September 18 Total: 20pts You may do this assignment with one other student. A team of two members must practice pair programming. Pair programming
More informationBinary search tree with SIMD bandwidth optimization using SSE
Binary search tree with SIMD bandwidth optimization using SSE Bowen Zhang, Xinwei Li 1.ABSTRACT In-memory tree structured index search is a fundamental database operation. Modern processors provide tremendous
More informationPerformance Testing at Scale
Performance Testing at Scale An overview of performance testing at NetApp. Shaun Dunning shaun.dunning@netapp.com 1 Outline Performance Engineering responsibilities How we protect performance Overview
More informationAnalysis of a Production/Inventory System with Multiple Retailers
Analysis of a Production/Inventory System with Multiple Retailers Ann M. Noblesse 1, Robert N. Boute 1,2, Marc R. Lambrecht 1, Benny Van Houdt 3 1 Research Center for Operations Management, University
More informationBest Practices in Model Development and Maintenance Adam Rose (adam.rose@xpsolutions.com), Product Manager, XP Solutions
Best Practices in Model Development and Maintenance Adam Rose (adam.rose@xpsolutions.com), Product Manager, XP Solutions adopted from Best practices for software development projects Mike Perks (mperks@us.ibm.com),
More informationFIXED CHARGE UNBALANCED TRANSPORTATION PROBLEM IN INVENTORY POOLING WITH MULTIPLE RETAILERS
FIXED CHARGE UNBALANCED TRANSPORTATION PROBLEM IN INVENTORY POOLING WITH MULTIPLE RETAILERS Ramidayu Yousuk Faculty of Engineering, Kasetsart University, Bangkok, Thailand ramidayu.y@ku.ac.th Huynh Trung
More informationOptimizing Help Desk Size
Optimizing Help Desk Size John Sauter December 20, 2009 Abstract Any large organization that uses personal computers needs a help desk to help users deal with computer problems. There is an optimum staffing
More informationThe Exponential Distribution
21 The Exponential Distribution From Discrete-Time to Continuous-Time: In Chapter 6 of the text we will be considering Markov processes in continuous time. In a sense, we already have a very good understanding
More informationSLA-based Admission Control for a Software-as-a-Service Provider in Cloud Computing Environments
SLA-based Admission Control for a Software-as-a-Service Provider in Cloud Computing Environments Linlin Wu, Saurabh Kumar Garg, and Rajkumar Buyya Cloud Computing and Distributed Systems (CLOUDS) Laboratory
More informationScheduling. Monday, November 22, 2004
Scheduling Page 1 Scheduling Monday, November 22, 2004 11:22 AM The scheduling problem (Chapter 9) Decide which processes are allowed to run when. Optimize throughput, response time, etc. Subject to constraints
More informationPerformance Workload Design
Performance Workload Design The goal of this paper is to show the basic principles involved in designing a workload for performance and scalability testing. We will understand how to achieve these principles
More informationJANUARY 2016 EXAMINATIONS. Life Insurance I
PAPER CODE NO. MATH 273 EXAMINER: Dr. C. Boado-Penas TEL.NO. 44026 DEPARTMENT: Mathematical Sciences JANUARY 2016 EXAMINATIONS Life Insurance I Time allowed: Two and a half hours INSTRUCTIONS TO CANDIDATES:
More informationThe Joint Distribution of Server State and Queue Length of M/M/1/1 Retrial Queue with Abandonment and Feedback
The Joint Distribution of Server State and Queue Length of M/M/1/1 Retrial Queue with Abandonment and Feedback Hamada Alshaer Université Pierre et Marie Curie - Lip 6 7515 Paris, France Hamada.alshaer@lip6.fr
More informationMonitoring Large Flows in Network
Monitoring Large Flows in Network Jing Li, Chengchen Hu, Bin Liu Department of Computer Science and Technology, Tsinghua University Beijing, P. R. China, 100084 { l-j02, hucc03 }@mails.tsinghua.edu.cn,
More informationOptimal order placement in a limit order book. Adrien de Larrard and Xin Guo. Laboratoire de Probabilités, Univ Paris VI & UC Berkeley
Optimal order placement in a limit order book Laboratoire de Probabilités, Univ Paris VI & UC Berkeley Outline 1 Background: Algorithm trading in different time scales 2 Some note on optimal execution
More informationPerformance Targets for Developers
Performance Targets for Developers Nov 2011 Performance targets are usually given at the end user or overall system level, such as an end to end response time target or an overall throughput target. After
More informationHSR HOCHSCHULE FÜR TECHNIK RA PPERSW I L
1 An Introduction into Modelling and Simulation Prof. Dr.-Ing. Andreas Rinkel af&e andreas.rinkel@hsr.ch Tel.: +41 (0) 55 2224928 Mobil: +41 (0) 79 3320562 Goal After the whole lecture you: will have an
More informationQuality of Service versus Fairness. Inelastic Applications. QoS Analogy: Surface Mail. How to Provide QoS?
18-345: Introduction to Telecommunication Networks Lectures 20: Quality of Service Peter Steenkiste Spring 2015 www.cs.cmu.edu/~prs/nets-ece Overview What is QoS? Queuing discipline and scheduling Traffic
More informationPerformance Testing Process A Whitepaper
Process A Whitepaper Copyright 2006. Technologies Pvt. Ltd. All Rights Reserved. is a registered trademark of, Inc. All other trademarks are owned by the respective owners. Proprietary Table of Contents
More informationC. Wohlin, "Managing Software Quality through Incremental Development and Certification", In Building Quality into Software, pp. 187-202, edited by
C. Wohlin, "Managing Software Quality through Incremental Development and Certification", In Building Quality into Software, pp. 187-202, edited by M. Ross, C. A. Brebbia, G. Staples and J. Stapleton,
More informationThere are a number of factors that increase the risk of performance problems in complex computer and software systems, such as e-commerce systems.
ASSURING PERFORMANCE IN E-COMMERCE SYSTEMS Dr. John Murphy Abstract Performance Assurance is a methodology that, when applied during the design and development cycle, will greatly increase the chances
More information