29:
230:
process and perhaps one that should always be carried out (instead of simply accepting that a program is performing reasonably). After eliminating all extraneous processing (just by removing all the embedded comments for instance), a new runtime analysis would more accurately detect the "genuine" hot
201:
to retain control, so you could execute the program at fairly high speed instead of interpreting each instruction one at a time and record in a file just where a program diverged from sequentiality. By processing this file you could figure out where the program was spending most of its time. So the
251:
view of hot spots (from an instruction step perspective) since most instructions have different timings on many machines. It nevertheless provides a measure of highly used code and one that is quite useful in itself when tuning an algorithm.
231:
spots in the translation. If no hot spot detection had taken place at all, the program may well have consumed vastly more resources than necessary, possibly for many years on numerous machines, without anyone ever being fully aware of this.
148:
to the next instruction to be executed) is frequently found to contain the address of an instruction within a certain range, possibly indicating code that is in need of optimization or even indicating the existence of a 'tight'
137:
where a high proportion of executed instructions occur or where most time is spent during the program's execution (not necessarily the same thing since some instructions are faster than others).
243:
can be used to count each time a particular instruction is executed and later produce either an on-screen display, a printed program listing (with counts and/or percentages of total
46:
93:
65:
72:
112:
261:
161:
79:
298:
214:
145:
50:
61:
247:) or a separate report, showing precisely where the highest number of instructions took place. This only provides a
39:
240:
157:
210:
17:
244:
86:
156:. This simple technique can detect highly used instructions, although more-sophisticated methods, such as
185:
173:
198:
193:
In the '60s, someone invented the concept of a 'jump trace'. This was a way of altering the
153:
134:
130:
141:
217:! The reason was that it was translating from one code system into another into another.
292:
226:
The example above serves to illustrate that effective hot spot detection is often an
194:
176:
28:
227:
206:
203:
280:
213:. We found out it was spending 87 percent of its time reading
150:
22:
202:
first day we had this software running, we applied it to our
179:
described his first encounter with what he refers to as a
281:
Jack Woehr: An interview with Donald Knuth, April 1996.
197:of a program so it would change the next branch or
53:. Unsourced material may be challenged and removed.
235:Instruction set simulation as a hot spot detector
164:, achieve this more accurately and consistently.
209:supplied by, I suppose it was in those days,
8:
140:If a program is interrupted randomly, the
133:is most usually defined as a region of a
113:Learn how and when to remove this message
18:Hotspot (disambiguation) § Computing
273:
7:
62:"Hot spot" computer programming
51:adding citations to reliable sources
14:
262:Profiling (computer programming)
27:
38:needs additional citations for
1:
168:History of hot spot detection
315:
158:instruction set simulators
15:
241:instruction set simulator
211:Control Data Corporation
245:instruction path length
219:
299:Software optimization
191:
162:performance analyzers
183:in an interview for
47:improve this article
16:For other uses, see
186:Dr. Dobb's Journal
174:computer scientist
189:in 1996, saying:
123:
122:
115:
97:
306:
283:
278:
199:jump instruction
195:machine language
135:computer program
131:computer science
118:
111:
107:
104:
98:
96:
55:
31:
23:
314:
313:
309:
308:
307:
305:
304:
303:
289:
288:
287:
286:
279:
275:
270:
258:
237:
224:
170:
142:program counter
119:
108:
102:
99:
56:
54:
44:
32:
21:
12:
11:
5:
312:
310:
302:
301:
291:
290:
285:
284:
272:
271:
269:
266:
265:
264:
257:
254:
236:
233:
223:
220:
169:
166:
121:
120:
35:
33:
26:
13:
10:
9:
6:
4:
3:
2:
311:
300:
297:
296:
294:
282:
277:
274:
267:
263:
260:
259:
255:
253:
250:
246:
242:
234:
232:
229:
221:
218:
216:
212:
208:
205:
200:
196:
190:
188:
187:
182:
178:
175:
167:
165:
163:
159:
155:
152:
147:
143:
138:
136:
132:
128:
117:
114:
106:
95:
92:
88:
85:
81:
78:
74:
71:
67:
64: –
63:
59:
58:Find sources:
52:
48:
42:
41:
36:This article
34:
30:
25:
24:
19:
276:
248:
238:
225:
192:
184:
180:
177:Donald Knuth
171:
139:
126:
124:
109:
100:
90:
83:
76:
69:
57:
45:Please help
40:verification
37:
268:References
181:jump trace
73:newspapers
228:iterative
222:Iteration
293:Category
256:See also
249:relative
215:comments
207:compiler
127:hot spot
103:May 2009
204:Fortran
146:pointer
87:scholar
89:
82:
75:
68:
60:
144:(the
94:JSTOR
80:books
172:The
154:loop
66:news
239:An
160:or
151:CPU
129:in
49:by
295::
125:A
116:)
110:(
105:)
101:(
91:·
84:·
77:·
70:·
43:.
20:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.
