(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 3, 2011
FUZZY HRRN CPU SCHEDULINGALGORITHM
1
Bashir Alam,
1
M.N. Doja,
2
R. Biswas,
3
M. Alam
1
Department of Computer Engineering, Jamia Millia Islamia, New Delhi, India
2
Department of Computer Science and Engineering, Manav Rachna University, Faridabad, India
3
Department of computer Science, Jamia millia Islamia,New Delhi India
Email: - babashiralam@gmail.com,ndoja@yahoo.com,ranjitbiswas@yahoo.com, mansaf@lycos.com
Abstract
— There are several scheduling algorithms like FCFS,SRTN, RR, priority etc. Scheduling decisions of these algorithmsare based on parameters which are assumed to be crisp.However, in many circumstances these parameters are vague.The vagueness of these parameters suggests that scheduler shoulduse fuzzy technique in scheduling the jobs. In this paper we haveproposed a novel CPU scheduling algorithm Fuzzy HRRN thatincorporates fuzziness in basic HRRN using fuzzy Technique FIS.
Keywords: - HRRN, CPU Scheduling, FIS, Fuzzy Logic
1.
INTRODUCTION
When a computer is multi programmed, it frequently hasmultiple processes competing for the CPU at the same time.When more than one process is in the ready state and there isonly one CPU available, the operating system must decidewhich process to run first. The part of operating system thatmakes the choice is called short term scheduler or CPUscheduler. The algorithm that it uses is called schedulingalgorithm. There are several scheduling algorithms. Differentscheduling algorithms have different properties and the choiceof a particular algorithm may favor one class of processes overanother. Many criteria have been suggested for comparingCPU scheduling algorithms and deciding which one is the bestalgorithm [1]. Some of the criteria include the following:-(i)Fairness(ii)CPU utilization(iii)Throughput(iv)Turnaround time(v)Waiting time(vi)Response timeIt is desirable to maximize CPU utilization and throughput, tominimize turnaround time, waiting time and response time andto avoid starvation of any process. [1, 2]Some of the scheduling algorithms are briefly describedbelow:(i)
FCFS: In First come First serve scheduling algorithmthe process that request first is scheduled forexecution. [1,2,3](ii)
SJF: In shortest Job first scheduling algorithm theprocess with the minimum burst time isscheduled for execution.[1,2](iii)
SRTN: In shortest Remaining time next schedulingalgorithm , the process with shortest remainingtime is scheduled for execution.[3](iv)
Priority: in Priority Scheduling algorithm the processwith highest priority is scheduled for execution.(v)
Round-robin: In this the CPU scheduler goes aroundthe ready queue allocating the CPU to eachprocess for a time interval of up to one timequantum. [1,2,3](vi)
Multilevel queue scheduling: In this the ready queueis partitioned into several separate queue. Theprocesses are permanently assigned to one queuegenerally based on some property of the processsuch as memory size, process priority or processtype. Each queue has its own schedulingalgorithm. There is scheduling among thequeues, which is commonly implemented asfixed-priority preemptive scheduling. Eachqueue has absolute priority over low priorityqueues.[1](vii)
Multilevel feedback-queue scheduling:-This allows aprocess to move between queues.[1](viii)
Fair share Scheduling: Fair share schedulerconsiders the execution history of a related groupof processes, along with the individual executionhistory of each process in making schedulingdecision. The user community is divided into afair- share groups. Each group is allocated afraction of CPU time. Scheduling is done on thebasis of priority of the process, its recentprocessor usage and the recent processor usagesof the group to which the process belongs. Eachprocess is assigned a base priority. The priorityof a process drops as the process uses theprocessor and as the group to which processbelongs uses the processor.[3]
120http://sites.google.com/site/ijcsis/ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 3, 2011
(ix)
Guaranteed scheduling:-In this a ratio of actual CPUtime a process had and its entitled CPU time iscalculated. The process with this lowest ratio isscheduled[2](x)
Lottery Scheduling:-The basic idea is to giveprocesses lottery tickets for CPU time. Whenevera scheduling decision has to be made , a lotteryticket is chosen at random and the processholding the ticket gets the CPU.[2](xi)
HRRN: - In this response ration is calculated for eachprocess. The process with the highest ratio isscheduled for execution. [3]In all the above scheduling algorithm the parameters used arecrisp. However, in many circumstances these parameters arevague. [9] To exploit this vagueness we have used fuzzy logicin our proposed scheduling algorithm. We have also donesimulation for comparing the performance of basic HRRNscheduling algorithm and Fuzzy HRRN scheduling algorithm.
2. FUZZY INFERENCE SYSTEMS AND FUZZYLOGIC
A fuzzy inference system (FIS) tries to derive answers from aknowledgebase by using a fuzzy inference engine. Theinference engine which is considered to be the brain of theexpert systems provides the methodologies for reasoningaround the information in the knowledgebase and formulatingthe results. Fuzzy logic is an extension of Boolean logicdealing with the concept of partial truth that denotes the extentto which a proposition is true. Whereas classical logic holdsthat everything can be expressed in binary terms (0 or 1, black or white, yes or no), fuzzy logic replaces Boolean truth valueswith the degree of truth. Degree of truth is often employed tocapture the imprecise modes of reasoning that play anessential role in the human ability to make decisions in anenvironment of uncertainty and imprecision. The membershipfunction of a fuzzy set corresponds to the indicator function of the classical sets. It can be expressed in the form of a curvethat defines how each point in the input space is mapped to amembership value or a degree of truth between 0 and 1. Themost common shape of a membership function is triangular,although trapezoidal and bell curves are also used. The inputspace is sometimes referred to as the universe of discourse [4].Fuzzy Inference Systems are conceptually very simple. AnFIS consists of an input stage, a processing stage, and anoutput stage. The input stage maps the inputs, such asdeadline, execution time, and so on, to the appropriatemembership functions and truth values. The processing stageinvokes each appropriate rule and generates a result for each.It then combines the results of the rules. Finally, the outputstage converts the combined result back into a specific outputvalue [4]. As discussed earlier, the processing stage, which iscalled the inference engine, is based on a collection of logicrules in the form of IF-THEN statements, where the IF part iscalled the "antecedent" and the THEN part is called the"consequent". Typical fuzzy inference subsystems havedozens of rules. These rules are stored in a knowledgebase. Anexample of fuzzy IF-THEN rules is: IF Remaining Time isextremely short then priority is very high, in which RemainingTime and priority
are linguistics variables and extremely short
and very high
are linguistics terms. The five steps toward afuzzy inference are as follows:• fuzzifying inputs• applying fuzzy operators• applying implication methods• aggregating outputs• defuzzifying resultsBelow is a quick review of these steps. However, a detailedstudy is not in the scope of this paper. Fuzzifying the inputs isthe act of determining the degree to which they belong to eachof the appropriate fuzzy sets via membership functions. Oncethe inputs have been fuzzified, the degree to which each partof the antecedent has been satisfied for each rule is known. If the antecedent of a given rule has more than one part, thefuzzy operator is applied to obtain one value that representsthe result of the antecedent for that rule. The implicationfunction then modifies that output fuzzy set to the degreespecified by the antecedent. Since decisions are based on thetesting of all of the rules in the Fuzzy Inference System (FIS),the results from each rule must be combined in order to makethe final decision. Aggregation is the process by which thefuzzy sets that represent the outputs of each rule are processesinto a single fuzzy set. The input for the defuzzificationprocess is the aggregated output fuzzy set and the output isthen a single crisp value [4]. This can be summarized asfollows: mapping input characteristics to input membershipfunctions, input membership function to rules, rules to a set of output characteristics, output characteristics to outputmembership functions, and the output membership function toa single crisp valued output. There are two common inferencemethods [4]. The first one is called Mamdani's fuzzy inferencemethod proposed in 1975 by Ebrahim Mamdani [5] and thesecond one is Takagi-Sugeno-Kang, or simply Sugeno,method of fuzzy inference introduced in 1985 [6]. These twomethods are the same in many respects, such as the procedureof fuzzifying the inputs and fuzzy operators. The maindifference between Mamdani and Sugeno is that the Sugeno’soutput membership functions are either linear or constant butMamdani’s inference expects the output membership functionsto be fuzzy sets. Sugeno’s method has three advantages.Firstly, it is computationally efficient, which is an essentialbenefit to real-time systems. Secondly, it works well withoptimization and adaptive techniques. These adaptivetechniques provide a method for the fuzzy modeling procedureto extract proper knowledge about a data set, in order tocompute the membership function parameters that best allowthe associated fuzzy inference system to track the giveninput/output data. The third, advantage of Sugeno typeinference is that it is well-suited to mathematical analysis.The block diagram of the proposed fuzzy inference system isgiven in figure1.
121http://sites.google.com/site/ijcsis/ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 9, No. 3, 2011
In the proposed model, the input stage consists of threelinguistic variables. The first one is the static priority that isassigned to the process before its execution. The second is theexpected remaining time of the process. The third input is theResponse Ratio of the process. The output stage consists of one linguistic variable called Dynamic priority.
Figure 1: Block diagram of FIS
The input and out variables are mapped into fuzzy sets usingappropriate membership functions. The shape of themembership function for each linguistic term is determined bythe expert. Adjusting these membership functions in anoptimal mode is very difficult. However, there are sometechniques for adjusting membership functions [7,8]Thesetechniques cannot be covered in this paper. They can befurther studied in a separate paper.The membership functions for inputs and outputs are givenbelow
Membership Function for DP (Dynamic Priority)
Type- Triangular, Range:1-5, Very low-[-1,0,1], Low:-[0,1.5,3] medium:-[2,3,4] High:-[3,4,5] Very High:-[4,5,6]
Membership Function for SP (Static Priority)
Type- Triangular, Range: 1-5, low-[-2, 0, 2], medium-[1, 2.5,4] High [3, 5, 7]
Membership Function for RT (Remaining Time)
Type- Triangular, Range: 0-5, Extremely short:-[-2, 0, 2],Very Short:-[1, 2.5, 4] Short:-[3, 5, 7]
Membership Function for RR (Response Ratio)
Type- Triangular, Range: 0-10, Short:-[0, 1, 2], Medium:-[2,5, 8] Long:-[5, 10, 15]
Figure 2: Memebership Function for Static PriorityFigure 3: Membership function for Remaining TimeFigure 4: Membership Function for Response Ratio5Figure5: Membership Function for Dynamic Priority
Twenty seven rules are formulated and a Sugeno type fuzzyInference system is built. Some of the rules are listed below:
•
If the static priority is‘low’ and remaining time is ‘extremely short’ andResponse Ratio is ‘long’ then the dynamic priority is‘very high’.
•
If the static priority is‘low’ and remaining time is ‘ short’ and ResponseRatio is ‘short’ then the dynamic priority is ‘verylow’.
•
If the static priority is‘medium’ and remaining time is ‘ extremely short’and Response Ratio is ‘long’ then the dynamicpriority is ‘very high’.
•
If the static priority is‘medium’ and remaining time is ‘short’ and ResponseRatio is ‘short’ then the dynamic priority is‘medium’.ResponseRatioRemainingTimeFuzzyInferenceEngine(Sugeno)27 RulesStatic PriorityDynamicPriority
122http://sites.google.com/site/ijcsis/ISSN 1947-5500
Search History:
Result 00 of 00
00 results for result for
p.
Fuzzy HRRN CPU Scheduling Algorithm
There are several scheduling algorithms like FCFS, SRTN, RR, priority etc. Scheduling decisions of these algorithms are based on parameters which are assumed to be crisp. However, in many circumstances these parameters are vague. The vagueness…
(More)
1.1K
Reads
4
Readcasts
16
Embed Views
Recommended
More From This User
Notes
Load more
