Source Code Cross Referenced for Jips.java in  » Science » Cougaar12_4 » org » cougaar » qos » qrs » sysstat » Java Source Code / Java DocumentationJava Source Code and Java Documentation

Java Source Code / Java Documentation
1. 6.0 JDK Core
2. 6.0 JDK Modules
3. 6.0 JDK Modules com.sun
4. 6.0 JDK Modules com.sun.java
5. 6.0 JDK Modules sun
6. 6.0 JDK Platform
7. Ajax
8. Apache Harmony Java SE
9. Aspect oriented
10. Authentication Authorization
11. Blogger System
12. Build
13. Byte Code
14. Cache
15. Chart
16. Chat
17. Code Analyzer
18. Collaboration
19. Content Management System
20. Database Client
21. Database DBMS
22. Database JDBC Connection Pool
23. Database ORM
24. Development
25. EJB Server geronimo
26. EJB Server GlassFish
27. EJB Server JBoss 4.2.1
28. EJB Server resin 3.1.5
29. ERP CRM Financial
30. ESB
31. Forum
32. GIS
33. Graphic Library
34. Groupware
35. HTML Parser
36. IDE
37. IDE Eclipse
38. IDE Netbeans
39. Installer
40. Internationalization Localization
41. Inversion of Control
42. Issue Tracking
43. J2EE
44. JBoss
45. JMS
46. JMX
47. Library
48. Mail Clients
49. Net
50. Parser
51. PDF
52. Portal
53. Profiler
54. Project Management
55. Report
56. RSS RDF
57. Rule Engine
58. Science
59. Scripting
60. Search Engine
61. Security
62. Sevlet Container
63. Source Control
64. Swing Library
65. Template Engine
66. Test Coverage
67. Testing
68. UML
69. Web Crawler
70. Web Framework
71. Web Mail
72. Web Server
73. Web Services
74. Web Services apache cxf 2.0.1
75. Web Services AXIS2
76. Wiki Engine
77. Workflow Engines
78. XML
79. XML UI
Java
Java Tutorial
Java Open Source
Jar File Download
Java Articles
Java Products
Java by API
Photoshop Tutorials
Maya Tutorials
Flash Tutorials
3ds-Max Tutorials
Illustrator Tutorials
GIMP Tutorials
C# / C Sharp
C# / CSharp Tutorial
C# / CSharp Open Source
ASP.Net
ASP.NET Tutorial
JavaScript DHTML
JavaScript Tutorial
JavaScript Reference
HTML / CSS
HTML CSS Reference
C / ANSI-C
C Tutorial
C++
C++ Tutorial
Ruby
PHP
Python
Python Tutorial
Python Open Source
SQL Server / T-SQL
SQL Server / T-SQL Tutorial
Oracle PL / SQL
Oracle PL/SQL Tutorial
PostgreSQL
SQL / MySQL
MySQL Tutorial
VB.Net
VB.Net Tutorial
Flash / Flex / ActionScript
VBA / Excel / Access / Word
XML
XML Tutorial
Microsoft Office PowerPoint 2007 Tutorial
Microsoft Office Excel 2007 Tutorial
Microsoft Office Word 2007 Tutorial
Java Source Code / Java Documentation » Science » Cougaar12_4 » org.cougaar.qos.qrs.sysstat 
Source Cross Referenced  Class Diagram Java Document (Java Doc) 


001:        /*
002:         * <copyright>
003:         *  
004:         *  Copyright 2002-2007 BBNT Solutions, LLC
005:         *  under sponsorship of the Defense Advanced Research Projects
006:         *  Agency (DARPA).
007:         * 
008:         *  You can redistribute this software and/or modify it under the
009:         *  terms of the Cougaar Open Source License as published on the
010:         *  Cougaar Open Source Website (www.cougaar.org).
011:         * 
012:         *  THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
013:         *  "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
014:         *  LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
015:         *  A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
016:         *  OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
017:         *  SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
018:         *  LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
019:         *  DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
020:         *  THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
021:         *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
022:         *  OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
023:         *  
024:         * </copyright> 
025:         */
026:
027:        package org.cougaar.qos.qrs.sysstat;
028:
029:        import java.util.Arrays;
030:        import java.util.Map;
031:
032:        import org.cougaar.qos.qrs.DataValue;
033:
034:        /**
035:         * 
036:         * This is a JVM+Host microbenchmark facility designed to determine the CAPACITY
037:         * of a specific Java Virtual machine running on a specific host. We have found
038:         * that the version of JVM makes a big difference in the application
039:         * performance, orders of magnitudes, as well as the MIPS of the host. This
040:         * benchmark tries to unify the hosts BogoMIPS measures and JVM version.
041:         * Obviously to be valid the benchmark must be run under the same JAVA VM as the
042:         * application.
043:         * 
044:         * The goal is to measure JVM+host Capacity. We need to remove all the possible
045:         * sources of latency variation, such as clock granularity, OS scheduling,
046:         * memory swapping, and garbage collection. Given Java's clock granularity of 1
047:         * msec, we should run the test for at least 100ms to get 1% accuracy.
048:         * Unfortunately 100ms is longer than the OS scheduling increment, such as a
049:         * Linux Jiffy which is 10ms. NOTE RUNNING FOR A LONGER TIME DOES NOT HELP,
050:         * because OS scheduling other tasks, and garbage collection will then be
051:         * "averaged" into the measurements.
052:         * 
053:         * The scheme is:
054:         * 
055:         * 1) Estimate the number of benchmark iterations that will run a test for
056:         * around 100ms.
057:         * 
058:         * 2) Run COUNT tests with a SLEEP of msec between the tests. The sleep is to
059:         * help the benchmark process from being marked as greedy by the OS. Hence
060:         * sleeping reduces the probability of being preempted during the test by some
061:         * other job.
062:         * 
063:         * 3) Pick the fastest time as best estimator for the maximum Capacity. Used to
064:         * pick the second fastest because the fastest might be a bad measurement (the
065:         * best seems fine)
066:         * 
067:         * 4) Return JIPS (Java Instructions Per Second) Jips = (benchmarkIterations /
068:         * BestTime) * INST_PER_ITER
069:         * 
070:         * We have run benchmark on hosts with a load average of 10 and have got the
071:         * same Jips as when run with a unloaded host. This work despite an under
072:         * estimate of the number of iterations for 100ms.
073:         * 
074:         * Jips comes with with 3 built in benchmarks, You are encouraged to experiment
075:         * with others: 0 measures the overhead of the loop. 1 measures time to add two
076:         * integers (Default) 2 measures time to add two floats
077:         * 
078:         * Problems:
079:         * 
080:         * What if the application excapes to C code, such as array copy, sockets, or
081:         * compression?
082:         * 
083:         * 
084:         */
085:
086:        public class Jips extends SysStatHandler implements  Runnable
087:
088:        {
089:            public static final double PERIOD = 0.1;
090:            public static final int COUNT = 5;
091:            public static final int TYPE = 3;
092:            public static final int SLEEP = 100;
093:            public static final int MEM_SIZE = 1000000;
094:            public static final int MEM_SIZE_MINUS_1 = MEM_SIZE - 1;
095:
096:            // estimate the number of "instructions per iteration"
097:            public static final int[] INST_PER_ITER = { 19, 20, 20, 91, 148 };
098:            private double fieldD = 0.0;
099:            private int fieldI = 0;
100:            private int[] fieldM = null;
101:            private final int type;
102:            private String key;
103:            private DataValue theValue;
104:
105:            private double iterationsPerSecond;
106:
107:            public Jips(int type) {
108:                this .type = type;
109:            }
110:
111:            public Jips() {
112:                type = TYPE;
113:            }
114:
115:            public void initialize(String host, int pid) {
116:                if (type >= 3) {
117:                    fieldM = new int[MEM_SIZE];
118:                }
119:                // cache value here
120:                iterationsPerSecond = runTests(PERIOD, COUNT, SLEEP);
121:                fieldM = null; // let memory be garbage collected
122:                key = "Host" + KEY_SEPR + host + KEY_SEPR + "CPU" + KEY_SEPR
123:                        + "Jips";
124:                theValue = new DataValue(iterationsPerSecond,
125:                        HOURLY_MEAS_CREDIBILITY, "", PROVENANCE);
126:            }
127:
128:            // backward compatibility
129:            DataValue getData() {
130:                return theValue;
131:            }
132:
133:            public void getData(Map<String, DataValue> map) {
134:                map.put(key, theValue);
135:            }
136:
137:            public void printResults(double[] results) {
138:                for (int i = 0; i < results.length; i++) {
139:                    System.out.println(i + " " + results[i]);
140:                }
141:            }
142:
143:            public void run() {
144:                switch (type) {
145:                case 0:
146:                    break;
147:                case 1:
148:                    incI(1);
149:                    break;
150:                case 2:
151:                    incD(1.0);
152:                    break;
153:                case 3:
154:                    incM(1);
155:                    break;
156:                // Need better Cougaar benchmark, memory good enough for now.
157:                case 4:
158:                    incM(1);
159:                    incI(1);
160:                    incI(1);
161:                    incI(1);
162:                    break;
163:                }
164:            }
165:
166:            // Base Benchmarks
167:            // Interger overflow will happen when processors are about 10Ghz
168:            // (i.e. in the year 2006)
169:            // Floating Point Add
170:            public void incD(double value) {
171:                fieldD = fieldD + value;
172:            }
173:
174:            // Interger Add
175:            public void incI(int value) {
176:                fieldI = fieldI + value;
177:            }
178:
179:            // Memory Add
180:            public void incM(int value) {
181:                // sweep memory from both sides
182:                fieldI = fieldI + value;
183:                int dst = fieldI % MEM_SIZE;
184:                int src = MEM_SIZE_MINUS_1 - dst;
185:                fieldM[dst] = fieldM[src] + fieldI;
186:            }
187:
188:            // time Benchmark over n iterations
189:            public double time(int n) {
190:                fieldI = 0;
191:                fieldD = 0.0;
192:                // fieldM just keeps old values
193:                long start = System.currentTimeMillis();
194:                while (n-- > 0) {
195:                    run();
196:                }
197:                // return time in Seconds
198:                return (System.currentTimeMillis() - start) / 1000.0;
199:            }
200:
201:            // estimate the number of iterations to get the target period
202:            public int estimateN(int n, double target) {
203:                // exponentially increase n until time is above target
204:                double time;
205:                while ((time = Math.min(time(n), time(n))) < target) {
206:                    n = n * 2;
207:                    // sleep between tests
208:                    snooze(SLEEP + SLEEP);
209:                }
210:                return (int) (target / (time / n));
211:            }
212:
213:            // Sequence a number of tests and pick the second best time
214:            public double runTests(double target, int count, int sleepMsec) {
215:                // How many iterations to get about targetPeriod
216:                int n = estimateN(1, target);
217:                // Test results array
218:                double[] results = new double[count];
219:                for (int i = 0; i < count; i++) {
220:                    // run test
221:                    results[i] = time(n);
222:                    // sleep between tests
223:                    snooze(sleepMsec);
224:                }
225:                // sort results
226:                Arrays.sort(results);
227:                // Debug Dump array
228:                // printResults(results);
229:                // return the second fastest time
230:                // Seconds per iteration
231:                return n / results[0] * INST_PER_ITER[type];
232:
233:            }
234:
235:            private void snooze(int msec) {
236:                // sleep between tests
237:                try {
238:                    if (msec > 0) {
239:                        Thread.sleep(msec);
240:                    }
241:                } catch (InterruptedException e) {
242:                    e.printStackTrace();
243:                }
244:            }
245:
246:            public static void main(String args[]) {
247:                Jips benchmark = null;
248:
249:                benchmark = new Jips(0);
250:                benchmark.initialize("localhost", 0);
251:                System.out.println("Loop MJIPS = "
252:                        + benchmark.getData().getDoubleValue() / 1000000.0);
253:
254:                benchmark = new Jips(1);
255:                benchmark.initialize("localhost", 0);
256:                System.out.println("Int MJIPS = "
257:                        + benchmark.getData().getDoubleValue() / 1000000.0);
258:
259:                benchmark = new Jips(2);
260:                benchmark.initialize("localhost", 0);
261:                System.out.println("Double MJIPS = "
262:                        + benchmark.getData().getDoubleValue() / 1000000.0);
263:
264:                benchmark = new Jips(3);
265:                benchmark.initialize("localhost", 0);
266:                System.out.println("Memory MJIPS = "
267:                        + benchmark.getData().getDoubleValue() / 1000000.0);
268:
269:                benchmark = new Jips(4);
270:                benchmark.initialize("localhost", 0);
271:                System.out.println("Cougaar MJIPS = "
272:                        + benchmark.getData().getDoubleValue() / 1000000.0);
273:
274:            }
275:
276:        }
www.java2java.com | Contact Us
Copyright 2009 - 12 Demo Source and Support. All rights reserved.
All other trademarks are property of their respective owners.