COMMENT ⊗   VALID 00006 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	(fasload traverse)
C00006 00003	Travere
C00007 00004	 NIL
C00009 00005	 FRANZ
C00012 00006	 SCORE Dec 11, 1983
C00014 ENDMK
C⊗;
(fasload traverse)
(init-timit)
(timit)

Timing performed on Monday 04/04/83 at 12:51:18.
Cpu (- GC) Time = 6.689
Elapsed Time = 391.633335
Wholine Time = 90.15
GC Time = 45.174
Load Average Before  = 1.34567666
Load Average After   = 5.16910934
Average Load Average = 3.257393
NIL 

Timing performed on Monday 04/04/83 at 12:57:51.
Cpu (- GC) Time = 24.014
Elapsed Time = 241.833334
Wholine Time = 49.2166667
GC Time = 0.0
Load Average Before  = 5.1550678
Load Average After   = 4.9717611
Average Load Average = 5.06341445
NIL 

Timing performed on Monday 04/04/83 at 13:01:54.
Cpu (- GC) Time = 23.989
Elapsed Time = 118.866667
Wholine Time = 48.016667
GC Time = 0.0
Load Average Before  = 4.95022345
Load Average After   = 3.74252367
Average Load Average = 4.34637356
NIL 

Timing performed on Monday 04/04/83 at 13:03:54.
Cpu (- GC) Time = 23.993
Elapsed Time = 95.2
Wholine Time = 49.766667
GC Time = 0.0
Load Average Before  = 3.68303668
Load Average After   = 3.02455235
Average Load Average = 3.35379452
NIL 

(TIMIT)
Timing performed on Monday 06/13/83 at 12:19:04.
Cpu (- GC) Time = 6.687
Elapsed Time = 186.533333
Wholine Time = 81.316667
GC Time = 45.135
Load Average Before  = 3.12555504
Load Average After   = 3.16852546
Average Load Average = 3.14704025
NIL 

Timing performed on Monday 06/13/83 at 12:22:13.
Cpu (- GC) Time = 23.96
Elapsed Time = 84.7
Wholine Time = 43.9333334
GC Time = 0.0
Load Average Before  = 3.1394682
Load Average After   = 2.77139103
Average Load Average = 2.9554296
NIL 

New CPU
(fasload traverse)
(init-timit)
(timit)
Timing performed on Thursday 01/26/84 at 13:02:13.
Cpu (- GC) Time = 6.687
Elapsed Time = 89.466666
Wholine Time = 73.333333
GC Time = 45.128
Load Average Before  = 0.74832678
Load Average After   = 0.99613249
Average Load Average = 0.872229636
NIL 
Timing performed on Thursday 01/26/84 at 13:03:58.
Cpu (- GC) Time = 23.963
Elapsed Time = 78.666667
Wholine Time = 43.85
GC Time = 0.0
Load Average Before  = 0.97275448
Load Average After   = 1.40655911
Average Load Average = 1.1896568
NIL 
;;;Travere

D3
7/5/83
INIT:
	Elapsed	20.5
	Swap	.015
	GC	.491
	CPU	20.0
TRAVERSE:
	Elapsed	63.9
	CPU	63.9

D2
7/5/83, without interrupts
INIT:
	Elapsed	102.0
	GC	1.25
	CPU	100.0
	pgflts	2
TRAVERSE:
	Elapsed	299.0
	CPU	299.0
	pgflts	1

D1
1/25/84 with interrupts
INIT:
	Elapsed	49.2
	CPU	48.0
	GC	 1.16
TRAVERSE:
	Elapsed	181.0
	CPU	181.0
;;; NIL
TRAVERSE

Init:
cpu=85.81,elapsed=85.83,pagefaults=645
Traverse:
cpu=626.17,elapsed=626.26,pagefaults=1 (yow!)

Perturb the defstruct so that the structure is typed (as will be the
default for common-lisp).
Init:
cpu=86.7,elapsed=87.38,pagefaults=650
Traverse:
cpu=669.78,elapsed=669.88,pagefaults=5
[Hmmm.  I didn't think it would be any slower than the previous one
on traverse...  I should look at the xref code.  Creation of such a
structure should not be significantly slower than creation of just a
simple vector, however the structure has one more pointer-slot than
the vector does.]

Inline carcdr.  Make the above typed structure refernece inline-coded
without error checking.  (The above did not do type checking on the
structure, only that it is an "extend" and that the index is within
bounds;  essentially the same as doing a generic vector reference on a
simple vector.)

Init:
cpu=38.93,elapsed=39.72,pagefaults=664
Traverse:
cpu=273.26,elapsed=273.32,pagefaults=4
;;; FRANZ
∂06-Jun-83  1035	RPG   	traverse benchmark
 ∂31-May-83  1116	jkf%UCBKIM@Berkeley 	traverse benchmark 
Received: from UDEL-RELAY by SU-AI with TCP/SMTP; 31 May 83  11:16:38 PDT
Date: 31 May 83 11:01:08 PDT (Tue)
From: jkf%UCBKIM@Berkeley (John Foderaro)
Return-Path: <jkf%UCBKIM@Berkeley>
Subject: traverse benchmark
Received: by UCBKIM.ARPA (3.340/3.5)
	id AA13717; 31 May 83 11:01:08 PDT (Tue)
Received: from UCBKIM.ARPA by UCBVAX.ARPA (3.341/3.29)
	id AA11025; 31 May 83 11:00:19 PDT (Tue)
Received: from UCBVAX.ARPA by udel-relay.ARPA ; 31 May 83 14:02:52 EDT (Tue)
Message-Id: <8305311801.13717@UCBKIM.ARPA>
To: rpg@su-ai

 Here is the traverse benchmark.  Two data points are missing.  After two
days of trying to get an unloaded 750 which is up for sufficiently long
to measure it, I gave up.  Perhaps I'll fill it in someday.

780cpu,gc/750cpu,gc [seconds]  

>>> traverse: init-tlimit
					translinks 
			on				off

localf		18.47,11.8/32.98,19.3		18.7,12.73/33.18,20.75

compiled	20.42,11.77/35.45,19.72		29.61.68,12.82/51.23,21.03

interpreted	1059.13,13.58/1758.35,21.22	1068.2,13.88/1780.75,22.05

>>> traverse: timit
					translinks 
			on				off

localf		    82.98/132.63		    83.13/132.6

compiled	    156.21/244.01		    559.58/911.2

interpreted	    5225.58/			    5268.46/


;;; SCORE Dec 11, 1983

LISP
LOAD(TRAV.NEW)
MAKEFILE(TRAVERSE)
BCOMPL(TRAVERSE)
ST
∧
←
BCOMPL(TRAVERSE)
listing? 

listing? STore and redefine 
πππ

*****(FILEDATES not compileable)
u.d.f. "18-Oct-83 23:39:52" {in EVAL} in ("18-Oct-83 23:39:52" . 
<RPG>TRAVERSE..2)
("18-Oct-83 23:39:52" broken)
:
ββ
LOGO

Not block compiled:
Initialize:

36796 conses
36.149 seconds
6.246 seconds gc

1860.125 seconds
0.0 GC

(TIMIT-10)

collecting lists
9732, 10244 free cells

collecting large numbers
1503, 3039 free cells

collecting large numbers
3039, 3039 free cells

collecting large numbers
3039, 3039 free cells

collecting large numbers
3039, 3039 free cells

collecting large numbers
3039, 3039 free cells
36796 conses
37.62 seconds
6.098 seconds, garbage collection time
{NODE}#170776
0 conses
85.862 seconds
0.0 seconds, garbage collection time
100
100
←