Date post: | 25-Mar-2016 |
Category: |
Documents |
Upload: | alexander-pospishniy |
View: | 237 times |
Download: | 13 times |
I
3
., ., . -…….
7
., . -……………………………………………………………………………...
14
., ., . ………………………….
23
. …………………….. 29 ., ., . ……………………………………………………………………………….
33
. …………………………………………………………………….
41
., . OPENGPSS GPSS/PC………………………………………………………
47
., ., . ……………………………………………………………………………...
54
. …………………………….. 62 ., . 67
. ………………... 73 ., ., ., ., .
, ………………………………
77
. . FOREX…………………………………………………………………..
88
., ., . ……………………………………………………………………………….….
94
., . …………………………………………………………………………………...
108
. ……………………………………………………………..
112
. GRID MPLS….. 117 ., .
……………………………………………………………………...………. 122
., . ……………………………………………………………….
127
. 3D ……………………………………………………………..
134
., ., . -………………………………………………………………………………….
140
., . ………. 146 , ., .
…………………….. 152
. . ……………………………….
. ., ” ”. E-mail: [email protected] . . ” ”
. ., ” ” . ., ” ”
. , ” ” . ” ”
. “ ” . . ” ”
. ” ”
. . , « » . . ” ”
. . , ” ” « » . . ” ”
. ., ,
. ., -. -mail: [email protected]
. . . . . E-mail: [email protected]
. ., « » . ., ” ”. E-mail: [email protected]
. ” ” ., . . ., ” ”
. ., . . ., ” ” . ., . . ., ” ”
. ” ” . ” ” « » . ., . . ” ”
. ” ” . ” ”
. ., ” ” . ., ” ”
. ., , . ., ” ”
. ., - « ». -mail: [email protected]
. ., . . ., ” ” . ., ” ”
. . ” ” . , ” ”
. ” ” « » E-mail: [email protected] . . , “ ”
. , ” ” . ” ”.
. ” ”. . ., . . ., ” ”
. ” ” . ” ”
004
.
. , -
. One of the first expert systems was medical. In article the review in a historical retrospective show of this
class of expert systems contains, and some prospects of their application and the subsequent development in applied medicine are defined.
-
. -
1960- , -
, [2, 28].
-
80- XX . -
, , , ----
.
: " "; -
, ;
; -
.
.
MYCIN .
"... ,
, , -
-" [2, 33].
. [3, 19] -: " – -
,
-".
– -, -
-.
[4, 406]. –
,
[1, 39]. [6, 32] : -
,
, , -, ,
-, " "
– -.
. 1 [2, 41].
. 1.
4,
,
: 1)
; 2) ,
; 3) -
. -
,
1970- .
, , -
, -
. -, ,
, -.
, -, , -
, - – -
, -,
[6, 8].
MYCIN
. -
. MYCIN 500 -
, --
. -
. MYCIN
EMYCIN . , NEOMYCIN,
, ,
. -
, ( ). -
-, -
. MYCIN NEOMYCIN – -
.
INTERNIST, -, -
. -
, ,
, -.
500), , 3500
, -
. INTERNIST-I, – CADUCEUS.
, .
[7], 34 , 84 .
-. -
1000 -
.
. - ( -
, ), ,
.
. : ,
. , -
-
. -
: 1) -
, -
« » , 49 5,
(ABEL); 2) (MYCIN, HEME, -
: AI/COAG, CLOT);
3) (AI/MM, DIALYSIS THERAPY ADVISOR, EEG ANALYSIS SYSTEM);
4) (AI/RHEUM, ARAMIS);
5) (ANGY, ANNA, DIAGNOSER, DIGITALIS ADVISOR, GALEN, HEART IMAGE IN-TERPRETER, HT-ATTENDING, MECS-AI, MI, );
6) (MECS-AI, THYROID MODEL);
7) (BLUE BOX, HEADMED, NEUREX);
8) (CAS-NET/GLAUCOMA, MEDICO, OCULAR HERPES MODEL, PEC);
9) (CENTAUR, PUFF, WHEEZE, );
10) , - (EMERGE, MED1);
11) (MDX, PATREC, RADEX).
1
-
. -
, -
, -.
, - [5].
, , -
. ,
-
, , -. , -
Express -, -
[8]. Express Test
. 30 ,
- 1.
- [8]:
-;
-;
;
, ;
; -
.
. , -.
-, -
:
,
, -, ;
,
, -,
1, 2
0-20 4 . . 21-30 3 . . 31-40 2 . . 41-49 1 . . 50-65 66-90 . .
91-100
3-30
0-20 4 . . 21-30 3 . . 31-50 2 . . 51-77 1 . . 78-86 87-90 . .
91-100
6 , -
;
-,
;
. , , -
, -.
1. / . . , . . – : , 2000. – 384 .
2. ., . : , 4- .: . . – .: " . ", 2007. – 1152 .
3. . .: . .: . . – .: ", 2001. – 624 .
4. ., . : . . – .: , 1993. – 608 . 5. . ., . .
// " " (24-25 2008 .). – .: , 2008. – . 108-109.
6. . ., . . / . . , . . . – .: +, 1998. – 320 .
7. . . – / . . , . , . . // . – 1995. – 1. – . 30-32.
8. . . ATM Express: ( ): - / . ., . ., . . – .: , 2007. – 27 .
004.93(015.7)
., .,
.
,
. . -
. -.
. -.
This paper presents investigation of nonlinear Boolean transformations inverse for which are ambiguous and
its application in cryptographical algorithms. A new method for designing such class of Boolean trans-formations is suggested. The method deals with the procedure form representation of Boolean transformations. It allowed to buil Boolean transformatiom from hundreds Boolean variables. Boolean transformation on such class can be use for accelerate of user identification based on “zero-knoledge” conseption. The relationship between transformation building time and procedure form parameters is established.
-
-.
-.
, , -
. -
, , -
. ----
.
, ,
, . -
. ---
, .
, --
.
,
-
: --
, .
-
[1]. -
, - Y=F(X) , -
, - F(X)
(Y)
X= (Y) F(X).
, X - Y
F(X) . , -
--
… 8
, . , -
,
( -), ,
RSA, El-Gamal, EEC, -, DSS [1].
-
– -
. -
, DES, IDEA, Rijndael, ,
, - RC-5, SHA RIPEMD-160 [2].
---
. RSA, El-Gamal, EEC, , -
-.
, , – .
-
--
, . 1978 . “ -
” – RSA, -
MAXF X mod)( , “ --
” [2]. , --
, ,
, , , X1 X2
).()( 21 XFXF -, -
-
, -
, .
-, -
-. -
--
, -
. , -
- 3-4
[3]. , -
, --
, . -
, ---
, , - “ ”.
, --
[3]. -
-,
[2 ].
, -
, , ,
, .
--
, - [3]. ,
, -
-
” [4].
, -
« » , 49
9
.
-
-
F(X), - 2n n -
X={x1,x2,…,xn}, i {1,…,n}: xi {0,1} - n -
Y=F(X), Y={y1,y2,…,yn}, i {1,…,n}: yi {0,1}. i -
yi Y -, -
fi(X), - X. , ,
F(X) n f1(X),f2(X),…,fn(X) : F(X)={
f1(X),f2(X),…,fn(X),}. , -
F(X) -, ,
f1(X),f2(X),…,fn(X), F(X)
n - X={x1,x2,…,xn}.
, F(X) -, -
, -,
F(X) : Q,G , Q G: F(Q) = F(G) = U.
,
: - ; - - , -
Y X.
-, ,
,
, . -
, -.
- F(X)
, .1.
.1.
- F(X) k-
. , n k 2, h -
h = n/k. -, -
, ,
.1, -
. c h -
- 1, 2,…, h. j , j {1,…,h}, -
j(V) 2k k-
V={v1,v2,…,vk}, l {1,…,k}: vl {0,1}. -
2 -
. , -
Y
… 10
X,
h – . Zqj k -
, j n– Wq={Zq1,Zq2,…,Zqh} -
q , W0 = X, Wh = Y, -
, .1 :
1,1,1
1,1,1
2)1(1)1(,0
)()(:}1,...,1{
:},...,1{};,...,,{:},...,1{
qhqhqh
gqgqjqg
kjkjkjj
ZZZZZZgg
hqxxxZhj
(1)
Wq(Q) q
X=Q. -, Zqj(Q) j
Wq(Q).
F(X) , .1, -
– 1, 2, h,
={X1,X2,…,Xm}
, (2.10).
j , j {1,…,h}, - j(V) k -
j ={ j1(V), j2(V),…, jk(V)} , -
0 2k. ,
---
. - ,
1, 2, h
, -
Xt. --
: 1. h -
1, 2, h 2k --
–1, -. = .
2. n - 1, .
3. U = F(X1) - (2.10). , q
h , j- - (j=1,…h), -
zq-1,j(X1) --
, j(zq-1,j(X1))=-1, 0 2k-1.
4. Xt. - W0(Xt)=Xt. -
: = . Nc (Nc={1,…,h} ), -
(“ ”) - Xt
t. q : q=1. Nc=1, .7.
5. q<Nc-1, - j
: j(zq-1,j) -1, -
(1). - j
zq-1,j : j(zq-1,j)= -1, -
0 2k-1 j zq-1,j, -
(1). j - j -
zq-1,j : <j,zq-1,j>.
h - q Xt
, q=q+1
. 5. 6. q=Nc-1, j(zq-1,j)= -1, -
j zq-1,j, -, zq,j = j(zq-1,j)
zq-1,j+1 j , : j(zq,j)= -1.
j - j zq-1,j
: <j,zq-1,j>. j(zq-1,j) -1, zq,j -
(2.10). h
q Xt
« » , 49
11
, - q=q+1.
7. j : j(zq-1,j) = -1, ,
d {1,…,t-1}, -.8. j -
: j(zq-1,j) -1, d {1,…,t-1},
j(zq-1,j) zq-1,j+1 = zqj(Xd). -, .8.
-,
- Xt .
Xt -
: <j,z> : j(z)= -1. .4.
8. j : j(zq-1,j) = -1, -
-: j(zq-1,j(Xt)) =
zq-1,j+1(Xt) zq-1,j(Xd), j<h j(zq-1,j(Xt)) = zq-
1,1,(Xt) zq-1,j+1(Xd), j=h. 9. t < m,
t : t = t +1. . 4. 10. h -
1, 2,…, h, , ,
, , ,
0 2k–1: j {1,…,h}, z {0,…,2k-1}: j(z)= -1 : j(z)=
Random(0,2k-1).
h -
1, 2, h , , -,
.1, F(X).
, -
, U 2-n. , n
, -, -
, -
, -.
---
. n – F(X)
16, n=16. , , - F(X) -
4 (h=4), 4 (k=n/h=4). , -
4- . -
1, 2, 3, 4 16- 4- -
4 , , -
0 15- . 3-
16- , 16-
: X1=5E96h, X2 = B1C6h, X3 = 7A48h. -
1, 2, 3, 4
, 1. -
---
. --
1, 2, 3, 4. 3- 1 2 3
1, 2, 3, 4
64% -. , 36% -
.10 ,
.
… 12
.1. 1, 2, 3, 4
X
X1
1 2 3 4 1 2 3 4 0 11 12 4 11 8 1 3 14 9 2 2 0 0 6 7 3 4 4 8 4 11 4 8 8 12 6 5 5 10 10 7 3 8 6 2 3 10 7 2 7 12 15 13 15 12 15 8 12 7 15 6 9 3 6 3 9 6 8
10 15 5 15 15 10 11 13 14 15 13 10 14 12 12 8 9 12 11 13 6 0 5 6 10 0 14 2 8 2 12 9 15 0 3 15 3
1, 2, 3, 4
: ,
1, 2, 3, 4 -
, -,
, .7 . -
, , ---
.
,
--
. , g
Xg . , , -
,
, h/2.
, g Xg h -
, , h g/2 .
( .7) , ,
j zq,j q -
j zq,j, j(zq,j) zq-
1,j+1 j<h j(zq,j) zq-1,1 j=h. , ,
, , -
j . , , -
j 2k, , -, h g/2 j
, , j -
(h g)2/22 (k+1). -, -
-,
Pg , h
1, 2, , h
:
hkgghP )
21( )1(2
22
(2)
, tg , --
g : tg = 1/Pg .
, Tm , -
m -
, , -, -
m -:
m
j hk
m jhT
1)1(2
22
)2
1(
1 (3)
VT --
:
« » , 49
13
hn
T hV 2 (4)
-
m
. --
, (3) , , -
, -. ,
n=256 h=16 k=16,
m=4000 40000 , -
. - 4000 ,
F(X) -. -
47% .
.10 .
, . -
, - ( 2-3 ) -
. -
, -
. , -
-.
1. ., . .: ” ”, 2001.– 368 .
2. ., ., , ., SAC- . //
”K I”. , .–1998.– 31.– .131-140.
3. Seberry J., Zhang X., Zheng Y. Nonlinearity and propagation characteristics of balanced Boolean func-tions.//Information and Computation Academic Press. 1995.–Vol. 119, 1 –P.1-13.
681.3.07
., .
. .
, , . Investigation of neural network use for medical diagnostic problem on cancer detection example in
gynaecology is conducted in this work. Investigation of Takagi-Sugeno-Kang’s fuzzy neural network opportunities for this problem solving is conducted. Furthermore, posibility of test’s (which are necessary for diagnostics process) number decrease is conducted in this work.
-,
-. -
. [1]. -
,
- ( ).
-
(TSK- ) ([2]) , --.
( ) 32 1 ( – -
-). 32 TSK, – .
- 50:50, 60:40, 70:30,
80:20, 90:10 . . -
-, -
.
. 32 – .
– , -. ,
( ) 32 1 . -
: 50:50, 60:40, 70:30, 80:20, 90:10.
- TSK
-
. -.
, -
, -.
1.
TSK , -
[2]. TSK
M N - jx -
: 1R : if 11
221
11 ;...;; nn AxAxAx then N
jjj xppy
11101 ;
MR : if Mnn
MM AxAxAx ;...;; 2211
then N
jjMjMM xppy
10 ,
« » , 49
15
kiA – -
ix kR -
MkNixik
A ,1;,1, . TSK
kR -, :
N
jj
kA
kA xx
1
. (1)
M --
( -):
M
kk
M
kkk xy
xy
1
1 . (2)
N
jjkjkk xppy
10 .
k :
xkAk ,
(1). (
TSK , , )
, , -.
, , ,
TSK ( , TSK ).
, ( )
. , « »- . -
-, , .
: -?
-, .
, [2].
x , -
ic , .
-, -
.
[2]. –
-, , -
jx , -.
ixD :
N
j a
bji
ir
xxxD
12
2
2
exp . (3)
ar . ixD -
jx , .
ix ixD . -
ix , x ,
xD -.
1c . 2c
, .
-:
2
21
1
2
expb
bi
iinewr
cxcDxDxD . (4)
D - br -
, -. -
, ab rr .
x, maxxDnew . -
. -,
16
. -
, .
-
x , -, -
.
- ii dx ,
, ,
id , -: iii xdx , . , -
- ix , -
ic . ,
- x d ,
p , x ( N )
q , d . -,
, - Kiqpc iii ,1,, .
:
? .
-. -
-. ( -
) ,
. »- .
« »- . -
: « »- , .
.
, . ?
. -
. , d .
, i iq .
« » ( ), Back Propagation,
[2]. , . -
( ). -
, , , ,
–
. ,
[2]. , -
, .
kjp – « »- , – « -
»- . .
-
( – , ), -
, kjp TSK.
kjp : N
jjkjk
M
kk xppy
10
1
. (5)
M
r
N
jj
rA
N
jj
kA
k
x
x
1 1
1 , Mk ,1 . (6)
ll dxL , , ),1( Ll -
ld , - L :
« » , 49
17
L
MN
M
M
N
LNLM
LLMLM
LNL
LLL
NMMMN
NMMMN
d
dd
p
pp
p
pp
xxxx
xxxxxxxx
...
...
...
...
....................................
.........
.........2
1
1
0
1
11
10
11111
22
2122
221
212121
11
1111
111
111111
(7)
lj ( ) - i l
lx . :
dAp (7) A MNL 1 .
L
MN 1 . p -.
kjp
Lly l ,1, , -:
Apy (8)
dy : 2
121 L
l
ll dxyE (9)
Back Propagation, ( ),
-. -
.
-
TSK ( ) ( ). -
, .
2.
: -
.
, . -
, , , – . , -
-. -
, -,
0,98001 , 0,98, -
– ,
. ,
, ---
.
– A , ija – i j .
, , - A – ija .
i j ( ) ,
– . , -
i . A
i . , -
, -,
. 0,005.
,
18
005,0max iiiyy , iy – , -
; iy – , -.
K .
3.
. , , -
, , , « »- .
, (193 -) .
ba rr . RI (range influence).
RQ (rules quantity).
-
n
iii yy
nRMSE
1
21.
-
. -
. , « »
. 1
50:50 60:40 RI RQ RMSE train RMSE test RQ RMSE train RMSE test 0,1 64 1,8315 10-17 0,0332 64 1,5072 10-17 0,0414 0,2 64 2,6832 10-17 0,0312 64 3,1898 10-17 0,0389 0,3 64 3,2746 10-17 0,0297 64 3,7296 10-17 0,0371 0,4 64 5,9620 10-17 0,0282 64 4,5869 10-17 0,0351 0,5 62 6,0560 10-17 0,0277 61 5,4587 10-17 0,036 0,6 55 2,3649 10-16 0,0623 57 3,6669 10-16 0,0387 0,7 47 2,1801 10-16 0,0552 47 1,6546 10-16 0,047 0,8 32 4,3250 10-16 0,0376 33 2,2483 10-16 0,0637 0,9 21 2,7109 10-16 0,0504 27 4,2174 10-16 0,0776 1 19 6,3821 10-16 0,0549 20 7,7099 10-16 0,0912
2 70:30 80:20
RI RQ RMSE train RMSE test RQ RMSE train RMSE test 0,1 70 1,4754 10-17 0,0435 89 1,2581 10-17 0,0331 0,2 70 2,1681 10-17 0,0418 89 2,0689 10-17 0,0302 0,3 70 3,3746 10-17 0,0388 89 2,9218 10-17 0,0293 0,4 69 4,7760 10-17 0,0365 88 4,1680 10-17 0,0286 0,5 68 5,2207 10-17 0,0358 85 5,0317 10-17 0,0269 0,6 64 5,6841 10-17 0,0386 74 8,3585 10-17 0,0290 0,7 51 2,0565 10-16 0,0534 57 2,3107 10-16 0,047 0,8 37 4,0681 10-16 0,0967 34 6,5636 10-16 0,0956 0,9 28 4,5956 10-15 0,1212 25 9,5112 10-16 0,0865 1 6 1,2788 10-15 0,1691 5 2,3033 10-15 0,1637
– 90%, – 10%. 3
RI RQ RMSE train RMSE test 0,1 96 1,1710 10-17 2,9216 10-18
0,2 96 2,1647 10-17 5,9195 10-17 0,3 96 2,6863 10-17 8,1233 10-17 0,4 96 3,3784 10-17 9,6381 10-17 0,5 92 4,4979 10-17 1,3867 10-16 0,6 80 1,6441 10-16 3,0583 10-16 0,7 61 2,2497 10-16 4,5655 10-16 0,8 44 5,5951 10-16 1,3725 10-15 0,9 31 1,0292 10-15 3,4575 10-15 1 5 1,7631 10-15 3,3432 10-15
« » , 49
19
, , -
. ,
( ). , ( . 2, . 3),
.
« – -
», :
0,00E+005,00E-161,00E-151,50E-152,00E-152,50E-153,00E-153,50E-154,00E-154,50E-155,00E-15
0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1
50:5060:4070:3080:2090:10
. 1 RI – RMSE train
1 – 3
00,020,040,060,080,1
0,120,140,160,18
0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1
50:50
60:40
70:30
80:20
90:10
. 2 RI – RMSE test
1 – 3 . 1, -
5,01,0RI -
-, RI
; - ( . 2) – -
7:3 -.
, - « »- .
.
( -).
RI ). .
4 50:50 60:40
RI RQ RMSE train RMSE test RQ RMSE train RMSE test 0,1 64 2,3976 10-8 0,033091 64 1,7003 10-8 0,041256 0,2 64 2,3981 10-8 0,031066 64 1,7008 10-8 0,038731 0,3 64 2,411 10-8 0,029487 64 1,7095 10-8 0,036763 0,4 64 2,5551 10-8 0,027863 64 1,8073 10-8 0,034738 0,5 62 2,235 10-6 0,027419 61 1,6779 10-6 0,035122 0,6 55 5,51 10-6 0,028531 57 3,931 10-6 0,03407 0,7 47 8,112 10-6 0,030635 47 6,314 10-6 0,043796 0,8 32 8,5395 10-6 0,03862 33 7,1297 10-6 0,041018 0,9 21 1,9289 10-5 0,038845 27 1,0663 10-5 0,046861 1 19 3,6898 10-5 0,045584 20 2,8545 10-5 0,047587
5 70:30 80:20
RI RQ RMSE train RMSE test RQ RMSE train RMSE test 0,1 70 1,4761 10-8 0,043158 89 1,4916 10-8 0,033651 0,2 70 1,4767 10-8 0,041357 89 1,4919 10-8 0,030443 0,3 70 1,4817 10-8 0,038156 89 1,4985 10-8 0,0293 0,4 69 5,7658 10-8 0,035841 88 5,5153 10-8 0,0288 0,5 68 1,3224 10-6 0,035571 85 6,0647 10-8 0,0291 0,6 64 3,4702 10-6 0,043127 74 3,7547 10-6 0,031 0,7 51 4,9418 10-6 0,047605 57 6,8552 10-6 0,0359 0,8 37 7,3412 10-6 0,048571 34 1,0559 10-5 0,0148 0,9 28 1,0643 10-5 0,046314 25 1,7685 10-5 0,0376 1 6 0,00022524 0,095813 5 6,9307 10-4 0,1178
20
– 90%, – 10%. 6
RI RQ RMSE train RMSE test 0,1 96 1,3748 10-8 7,3358 10-8
0,2 96 1,3750 10-8 7,3361 10-8 0,3 96 1,3806 10-8 7,3317 10-8 0,4 96 1,4470 10-8 7,3233 10-8 0,5 92 5,4135 10-7 1,3909 10-6 0,6 80 3,2039 10-6 9,8092 10-6 0,7 61 6,0106 10-6 2,4068 10-5 0,8 44 9,8820 10-6 3,0978 10-5 0,9 31 1,6437 10-5 6,3682 10-5 1 5 7,0728 10-4 0,003 ,
.
. -
, .
0,00E+00
5,00E-06
1,00E-05
1,50E-05
2,00E-05
2,50E-05
0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9
50:50
60:40
70:30
80:20
90:10
. 3 RI – RMSE train
4 – 6
0
0,02
0,04
0,06
0,08
0,1
0,12
0,14
0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1
50:50
60:40
70:30
80:20
90:10
. 4 RI – RMSE test
4 – 6 ( . 3) -
-. ( . 4) -
- 7:3
.
4.
-, 1-3
4-6. « »- , -
,
, ( . ).
, Back
Propagation, [3], , -
TSK . –
« »-,
. , ,
. ( 1-6), -
, - ( -
), ).
-
7:3 -. , -
, -,
- « »-
5,01,0RI , – -.
5. ,
( 2),
« » , 49
21
, -.
9:1.
-
. n – , -
. RI 0,1. K – -, .
7. 7
n RQ RMSE train RMSE test K 2 76 9,1466 10-15 2,5951 10-14 3,6710 10-13 3 88 9,8243 10-16 3,5033 10-15 6,1173 10-14 4 95 2,9052 10-17 7,9815 10-17 7,7716 10-16 5 95 3,2477 10-17 7,6187 10-17 6,6613 10-16 6 96 2,1861 10-17 6,4230 10-17 6,6613 10-16 7 96 2,0654 10-17 6,5948 10-17 7,77168 10-16 8 96 3,5415 10-17 6,5741 10-17 6,6613 10-16 9 96 1,7497 10-17 3,9496 10-17 4,4409 10-16 10 96 1,7743 10-17 6,2081 10-17 8,8818 10-16 11 96 1,5131 10-17 3,8456 10-17 4,4409 10-16 12 96 2,0499 10-17 5,4912 10-17 6,6613 10-16 13 96 1,6495 10-17 6,60624 10-17 6,6613 10-16 14 96 1,8812 10-17 4,3088 10-17 4,4409 10-16 15 96 1,5806 10-17 3,1263 10-17 3,3307 10-16 16 96 1,4932 10-17 3,5029 10-17 4,4409 10-16 17 96 1,7960 10-17 3,9873 10-17 4,4409 10-16 18 96 1,4720 10-17 3,3978 10-17 3,3307 10-16 19 96 1,7069 10-17 3,5090 10-17 3,3307 10-16 20 96 1,2195 10-17 3,0116 10-17 3,3307 10-16 21 96 1,4985 10-17 3,0080 10-17 3,3307 10-16 22 96 1,4824 10-17 3,5931 10-17 4,4409 10-16 23 96 1,5059 10-17 2,8923 10-17 4,4409 10-16 24 96 1,3239 10-17 3,2038 10-17 4,4409 10-16 25 96 1,2587 10-17 2,3005 10-17 2,2204 10-16 26 96 1,5864 10-17 2,2263 10-17 2,2204 10-16 27 96 1,3964 10-17 2,9947 10-17 3,3307 10-16 28 96 1,6132 10-17 1,7530 10-17 2,2204 10-16 29 96 1,2920 10-17 1,4681 10-17 2,2204 10-16 30 96 1,2898 10-17 2,9216 10-18 5,5511 10-17 31 96 1,1404 10-17 2,9216 10-18 5,5511 10-17
, , K , -
, -.
RI
. ,
: 9 :
8 RI RQ RMSE train RMSE check K 1 2 0,0233 0,0672 0,5364
0,6 11 5,6436 10-15 1,0704 10-14 1,2401 10-13 7 :
9 RI RQ RMSE train RMSE check K 1 3 0,0218 0,0621 0,6588
0,7 6 0,0164 0,0647 0,7306 0,5 32 5,784 10-15 1,922 10-14 1,1446 10-13
, -.
, -. , -
-, . ,
, -, , -
( -
– , ). , , -
-, ,
, -. -
-. .
22
-
TSK -. -
, .
, --
. , , -, -
« »- -,
. , --
50:50
70:30 ,
« »- 5,0;1,0RI . ,
( ) .
. --
TSK , , -
-. -
.
1. Steve Saggese, Trevor Johnson, Ivy Basco, Yalew Tamrat. Automatic Segmentation of Uterine Cervix for in vivo localization and Identification of Cervical Intraepithelial Neoplasia.– Apogen Technologies 7545 Metropolitan Dr San Diego, CA 92108.
2. . . . – .: - « », 2004. – 352 .
3. . -. :
X (20-24 2008 ., ).– .: « » 2008.– 235 .
004.052.42
., .,
.
-
. -, -
.
. The paper is dedicated to solving the efficiency increasing problem for data transmission error detection in
spectrum modulation channel by properties such errors appearance accounted. For the guaranteed errors detec-tion in one and more channel symbols the approach based on Chinese Reminder Theorem has been proposed. In the course of the theoretical researches of proposed approach the guaranteed error detection has been proved for errors quantity less or equal to number of the check symbols in modular representation.
-
. -
(EDGE, CDMA2000, 1xEV-DO), -
(DVB-T/H/C/S/S2, ATSC), -
(Wi-Fi, Wi-MAX, Bluetooth), - [1]. -
( -PSK, --QAM)
.
– . : QPSK – -
2 , 8-PSK – 3 , 16-,32-,64-,128- 256-QAM – 4,5,6,7 8 -
. ---
. -
. , ,
.
-,
.
. , -
--
.
, .
, m : B={b1,b2,…,bm}, bl {0,1}, l=1,…,m
t = m/k -: B={X1,X2,…,Xt}. j
j, j {1,2,…,t} k
},...,,{},...,,{ 2)1(1)1(21 kjkjkjkj bbbxxxX -.
-
CRC (Cyclic Redundancy Check –
... 24
). , CRC .
CRC , B={b1,b2,…,bm}
P(B) m+k : mk
mmk
mkkk xbxbxbxbxbBP 1
12
31
21 ...)(. R(B)
P(B) Q(X) k CRC.
– , -
. -
, ), --
Q( ) CRC . - [3], ), -
Q(X), , -:
1. b1,b2,…,bm , Q( )
-: )()1()( XSxXQ ;
2. b1,b2,…,bm , -
k
k xqxqxqqXQ ...)( 2210
. 3. , k
. [3],
R(B)
2k . -, PCRC ,
- CRC
Q( ) k kCRCP 2 .
, CRC
-,
k , k -
CRC. -
, CRC .
, CRC -,
-
, -,
. ---
. --
(Weighed Check Sum – WCS) [4]. - CRC,
WCS -
, -. WCS -
, ,
2- . WCS
- CRC ,
)log1( 2 tk . ,
, 2- .
--
, .
---
. , -
– .
n - {p}n ={p1,p2, …,pn} P,
: n
jjpP
1
(1)
« » , 49
25
, - [1] ,
[0…P-1] -
, {p}n.
njpMrMrrrPM
jj
n
,...,1,mod,},...,,{:}1,...,0{ 21 (2)
, -.
(2) ( ). -
-, 1,…, n
: k
jpnj 2:},...,1{ (3) {p}n
: n>t {p}t = {p1,p2,…,pt} {p}n . , (2) (3)
X1,X2,…,Xt, -
:
ii
ntt
pAXtiAXXXXB
mod:},...,1{,},...,,,...,{ 11 (4)
{p}n:
jj
ntt
pAXnjAXXXX
mod:},...,1{,},....,,,...,{ 11 (5)
-
, .
:
1. {p}n ,
k , -, -
t -.
2. (5) B {p}n
. X1,X2,…,Xt -
, Xt,…,Xn -. -
, -
Z : ZBT || . 3. -
. 4. -
n- E={e1,e2,…,en},
ej – k : ej = { 1j, 2j,…, kj}, {0,1}:
ETYYR n},...,{ 1 (6) 5.
, -,
1 2 [2]: 221121 },....,,{,},...,,{ CYYYCYYY nt (7)
, -, 1 2: 1= 2.
,
(n-t) , :
a). {p}n -.
). pt+1,pt+2,…,pn -
:
},...,1{,1mod1
tjpp j
n
tii (8)
, -
, - Z, -
. .
: -.
, , , -
: ACYYYXXX tt 1,2121 :},...,,{},...,,{ (9)
-,
, , -
:
ACYYXXAXX nttn
2
111 },....,,,...,{,},...,{ (10)
(9) (10) -: 21 CAC . , 1 2,
... 26
-, -
n-t. :
,
. , ,
, d q
.
},,,...,,,....,,,...,{
},...,,,...,{},....{},...,{
111
1112
1111
1
1
nnqnqnqnt
ttdtdtdt
ttdtdtdt
n
t
eXeXXXeXeXXXC
eXeXXXCXXAXXA
(11)
:
iiiii
n
n
iiii
pcgpPc
MrrrPrgcM
mod1;/
,},...,.{,mod,, 211 (12)
(11) , - (12), :
PXgcA
PXgcA
ii
t
ii
tt
iiii
mod)''(
mod)(
1
1
tt
dtjjjj
t
iiii PegcXgcC mod)(
111
(13)
t
ii
t pP1
, 21
1
CCtn
tjj
t PPPpPP ,
qn
tll
C pP1
1 – , -
, -.
n
qnhh
C pP1
2 – , -
, .
, -
, :
ACACCC 2121 (14) (13) (14) :
Pegcegc
Pegc
n
qnllll
t
dtjjjj
tt
dtjjjj
mod)''''(
mod)(
11
1 (15)
(13) (14) :
},...,1{,'
},...,1{,'
/212
1
21
21
ntlPPPp
PPPc
tjPPccPPPP
lCCt
l
CCt
l
CCjj
CCt
(16)
lCP /2 – , -
, -, l
.
, -, Ptr , -
,
- Pte:
jtetr
j
tetr
j
CCtetr
tetrt
PPpPPc
PPPPPPPP
/
21 (17)
Pte/j – , -
, ,
j .
(15) – (17) :
212
21
mod))'
'((
mod)(
1
/
1
/
1
/
CCtetrn
qnlll
lCte
t
dtjjj
jteCC
tetrt
dtjjj
jtetr
PPPPegPP
egPPP
PPegPP
(18)
(18) -
tr, -
tr :
« » , 49
27
21
21
mod))'
'((
mod)(
/
1
/
/
CCtell
ltete
t
dtjjj
jteCC
tejj
jte
PPPegPP
egPPP
PegP
(19)
(19) PC1,
PC1. , (19) PC1.
(19)
. , (19) , (12) , :
1..1
1..1
jj
jj
pepg
(20)
, (19)
:
},...,1{,'
},...,1{,'
},...,1{,
tqnlpegtdtjpegtdtjpeg
lll
jjj
jjj
(21)
(21) :
},...,1{,' tdtjgg jj . -:
jC
jjjjj pPcpcpc modmod'mod 1 . -, 1mod1
jC pP .
(7). , (19)
. (19) , Pte.
, - (19) :
1Cte PP (22) ) -
, , - (22) ,
(22)
(22). , -
, u<n-t -
n-t-u -. , -
h -
: tnh (23)
-
, -,
. ,
--
, , -. -
–
. -
, .
, --
(8). -, ,
, -
, -. -
j < 2 , j=1..n, – .
-
, --
, . - j 2k, j=1..n.
h ,
. , , h=2, k=4, =8
53 , 2 (16
), , - 51 . -
51*4=204 . ,
. ,
(32 )
... 28
26156 . -
--.
. , -
k. -,
k , --
. ,
, , CRC,
--
.
-,
-.
-
, -. -
, -.
-
- – , -,
.
-.
1. . . – .: .–2001.–153 . 2. . . .: . ., –1999.–340 . 3. . . . .:
”, 2004.– 1104 . 4. ., ., .
// -. : 3(14).– ., .– 2008.– .121-128
004.415.2.043
.
. , .
. : , .
: . . ,
, . -.
The traditional analysis of program language operators if with laying out of them on separate lexemes as equal
in rights units of language complicates enough the syntactic analysis. The natural languages of intercourse are had at their perception of separation to the subject, to the predicate and other parts of language for the awareness of sense of suggestion. Something similar is offered on the stage of lexical analysis of program language operators. All lexemes are here divided into three groups: lexemes-objects, lexemes of action at al. A feature as programming if consists in the obligatory presence of pair of lexemes: lexeme-object and lexeme of action. Thus from this pair all operators begin and in this pair it is always the lexeme of action is a mandatory member. Therefore, if a lexeme-object appear at the lexical analysis, the simultaneous search of the proper lexeme of action is appropriate. Such approach allows considerably simplify the syntactic analysis of operators of language and accelelerate his imple-mentation.
-.
. -
--
. ,
-.
.
, .
, ,
.
– , LL- LR-
[1, 2]. -
. --
. , , -
, .
G = f(T, N, P, Ns), Ns – - ( ),
. ( ,
, , .). N – (
, , ). -
. ,
,
, .
-, -
. -
,
. -
30
-.
. -
[1, 2, 3]. - ( ,
, -), , -
.
. ,
, , -
. - ( -
) [2], -,
. , .
-,
, , -
.
, , , -.
-, -
-
.
,
, , :
. -, –
, : -, , . , L Ld + La, L – , Ld – -
, La – . -
. , , , – -
. --
, , , .
-. ,
, -
, , -, . -
. --
, .
. --
. -
, . – Ld -
. -
– La. .
. ,
. -
. , - “
”, -. Ns, -
, , -
, .
, ,
, .
: , .
( , , ), . -
, , , .
-, -
, .
, , . -
« » , 49 31 – , –
. -. -
, . -
. , ,
. 1. . 1
, -. -
-, -
.
-.
- “ ” -
. ,
, -
. ,
-. -
-.
. -
. -
- ( -
). .
,
) -, .
-, ,
-,
.
-. ,
IF -
. -
, --
. --
: .
, -, -
. -
. . -,
( , , )
. -.
.
, -
. -.
MNEM
- TYP
CODE
-
PRIOR . 1
-
MNEM. TYP , - –
-
, -
: ,
,
,
32
.
-. CODE
PRIOR -
. , , -
.
, (IF), - (FOR, WHILE, REPEAT), -
(READ, WRITE) [3, 4] (
). -,
. ,
CODE . -
-,
. , -, , -
.
-
. --
, : ---
. -
Ns – .
AVR, --
.
--
, , -,
. -,
– . -
, ,
. -
, --
.
-.
1. ., , . , . – .: “ ”, 2002. – 528 .
2. ., , . : , -. – .: “ ”, 2003. – 768 .
3. . . – .: ”, 2001.– 672 .
4. . . . – ; , 2000. – 366 .
004.074.32
.,
., .
,
. . . .
.
In article the new hash-searching organization such that keys may storing in one of dual hash-address has been
proposed. It is allowed to limit the searching time by dual memory access. The procedure of recursion recording keys into the hash-memory has been developed. The analytical evaluation of collision probability has been obtained. The possibilities of proposed organization for hash-searching in static and dynamic arrays of data has been analyzed
--
. ,
. -
, -
. -
,
. - —
,
, -. -
: -
.
, ,
-.
, -,
. --
. , -
, , ,
. , -
, --
.
, ,
, . X A -
- h(X): A=h(X). -
, , , ,
-. -
– ,
m : =m/M [1]. -
,
– ,
.
: .
34
-.
-.
perfect - [2].
[1]. -,
save smax -, -
. -. -
–
, , --
: h1(X), h2(X), h3(X),… -
. , , -
: hi(X) = (h1(X)+i) mod M. , -
– , -
. -
[3] - save
, -:
11
aves (1)
-, - smax -
. , ps = s+1 -
, s -.
-,
, smax.
( ) ,
-
, - .
- [1,2,4], -
Cichelli R.J. Czech Z.J. -
. , , perfect
[4]:
21)1(
)1(mmeT
(2)
(2), ---
.
-
-.
.
, .
- (
), 1 -) 2 ),
/2 . -
1 -2. -
h1(X) T1 h2(X) T2, , -
h1(X) h2(X) X.
-. ,
X , h1(X) 1.
, - h2(X) T2.
h1(X) h2(X) X
« » , 49
35 ,
-. ,
h1(X) 1, h2(X),
h1(X) . , -
2 h2(X) h1(X).
, -, -
.
, , -, ,
-,
" " : h1(X) T1, h2(X)
2. -
:
1. - h1(X) h2(X)
2. , h1(X) .
( ), -
. 3. h2(X).
, -.
4. , h2(X) .
( ), -
.
5. h1(X). , -
. - – .
, , -, ,
( ).
-
. , -, -
, .
.1. a ,b -
1, v,z w – 2. 4 -
: 1, h1(X1)=a, h2(X1)=z, a -
T1. X2, - b w (h1(X2)=b,
h2(X2)=w) b T1. X3, (h1(X3)=b,
h2(X3)=v), b v. b T1
2, X3 v T2. -
1, 2 3 .1.
X1 X3
X2
a
b
w
v
z
X4(a,v)
X3(b,v)
X1(a,z)
X2(b,w)
. 1. 1, 2, 3
X4, h1(X4)=a h2(X4)=v, , -
– a T1 v T2 .
--,
. ,
4 1,
a, 4 . -
1 z 2 , 1 ( -
a=h1(X1) h2(X1), z=h2(X1)
36
2, a=h1(X1) ). a,
X4 ( a h2(X4)).
4 .2.
X4 X3
X2 X1
a
b
w
v
z
X4(a,v)
X3(b,v)
X1(a,z)
X2(b,w)
. 2. 4
--
. t ( t {1,2} ),
, -. -
, -
. L, R -
nb .
-:
1. h1( ) h2( ), t=1.
2. h1( ) T1 d
. : =0, ,
h2( ) . -.
.4. 3. h2( )
T1 g . : =0,
, h1( ) . -
.
4. h1( ), h2( ) t L,R, nb . d
, R , -.
5. h2( ): h2( )=d. - t : t = 3-t.
6. h2( ) T2 g
. : =0, ,
h1( ) . -.10.
7. h1( ), h2( ) t L,R, nb . g
, L , -.
8. h1( ): h1( )=g. - t : t = 3-t.
9. h1( ) d. .5. 10. L,R nb
h1( ), h2( ) t. 11. t=1, h2(t)
h1(t) -1. . 13.
12. t=2, h1(t) h2(t) -
2. 13. ,
t : t = 3-t -.8. .
--
, . TW
:
11
2
)12()1()1(21j
jW jT
(3)
(3) , : 0.7
0.9 -.
--
, , .
« » , 49
37
. , w
w. , -
.
, . -
.3. -
, – - ={X1,X2,…,X6}, h1(X1)=a, h2(X1)=v,
h1(X2)=a, h2(X1)=z ; h1(X3)=b, h2(X3)=v; h1(X4)=b, h2(X4)=z; h1(X5)=c, h2(X5)=v; h1(X6)=c, h2(X6)=z.
. 3.
6- 5- , .3,
6- . ,
v z, ,
3- a,b,c .
6- . , - , 6-
. , ,
, , u
u, , - u+1.
, ,
6- . , 9 ,
6- -, .4.
. 4.
9- 6- ---
. m ,
M. p6 -
, .3, 6- -
5 , -.
-, v z.
>0.5, m M/2
, m>M/2. ,
, v -: h2(X)=v. X
a, h1(X)=a. , ,
Y, , h2(Y)=z h1(Y)=b.
c -, ,
G, h1(G)=c h2(G)=v M-1. ,
--
M-2. , Q1,Q2 Q3, -
, - ( , -
Q1 : h1(Q1)=a h2(Q1)=z; Q2 : h1(Q2)=b h2(Q2)=v; Q3
: h1(Q3)=c, h2(Q3)=z) M-6.
38
, , -, .3 -
Q1,Q2 Q3 -: c, v, z M-7. -
, , .3 ,
c v,
z. Q1,Q2 Q3 :
6)2()1(
)!3(!3! mmm
mm
m, m3/6. ,
v z M2/2. , q6 , -
, .3 -:
1276
33
)1(Mm
Mq 6 ,
m 6- -,
.3 :
12766
33
)1(11Mm
Mqp (4)
m M (4) -
, -
5% :
MMmMMmp
1212)
121(1
3
4
37
33
6 (5)
-, , -
(5) -6 6- .
, -.4,
9- , , 9 -
:
3
3
9 36 Mp (6)
(5) (6) ,
: 6>>p9. , ,
,
. , , -
, ,
6- , - (5).
-, -
. -
, - save smax -
. , -
. , - s max=2. -
, , ps , s
ps = s+1. ,
=0.75 - 0.01 -
28 . , , -
14 .
s ave -
:
RL
RLaves 2' (7)
L – , R –
. ,
( 0.7) , -
, 1.5. , =0.6 -
L -
« » , 49
39
0.7, R = 0.5. , , s ave ,
(7) 1.42. 0.7
-
1.5. ,
s ave -
save -. 0.4 (
) s ave<save, , --
, . -, -
=0.75 - 2.7 -
-.
,
. , --
, --
.
– ,
,
, , .
-
. -,
. , -
, -
, -
.
-
, -
.
h1(X)
h2(X). --
. - –
-, h1(X)
h2(X) , --
. p - h1(X) h2(X)
(5) :
M
MMjT
jjj
j
p
121
1
)12
1(12
3
3
111
)1(3
(8)
(2) (8) , --
. , , -
,
, , -
( ) -,
. , -
, (3),
Nw , --
:
)11
2(
121
3
M
mN w (9)
(9) , ---
-
. -
,
40
, , : -
, ---
, -, .
-
. , -
. --
. -
, 2-3 – -
. -
. -
-.
, -
-, -
, -
-.
1. . .- .: ,1980.- 198 . 2. ., , . -
// ”. , .- 1998,- 31,- C.14-23.
3. . -. // ”K I”. ,
.- 2003.- 40.- .131-140. 4. Czech Z.J., Havas G., Majevski B.S. An Optimal algorithm for generating minimal perfect hash func-
tions.//Information processing letters. – 1997. – Vol.43.- 5. - P.257-264.
681.3.06
.
-.
( ) , , , . ,
.
The approach to the complex task decision for analytical support of software analysis and synthesis is
considered. Key elements of such approach use representation formats for specification headers contents of task decision resources (TDR) and knowledge base for analogues search, a choice of best components, use of transformation rules and decomposition mechanisms for TDR. It is shown, that application of such knowledge bases allows using the most of the previous software, and making values of target semantic variables extremely close to restriction requirements and criteria of processing speed.
. -
( ) ( )
[1] -. ,
-
, - [3, 6, 7].
--
( ) [2] -
.
, -. -
– -
[1, 3]. ( )
: , -
) ; -
; - -
; , ,
; , -
; , .
,
. --
[4]. -
-.
-
, . -
. ( ) - ( ) -
, , .
rsk, rtk, rrk, rkk,
rdk, rlk, rmk, adk, ask, atk, aek, ark, alk, ack, afk(xk1,…,xkimax, k1,…, kjmax, ak1,…, aknmax)) kafk( k1( k1), ..., kj( kjmax))|xk1, ..., xkimax, k1,..., kjmax, k – ; xk1, …, xkimax – ; k1,…, kjmax –
; ak1,…,aknmax – .
. 1. -
42 rtk, rrk, rkk, rlk, rmk, adk, atk, aek, ark, ack, avk -
, , , -
, -.
rsk
rtk
rrk
REF REF REF
… REF … …
rkk rdk
…
… rmk
rlk
… … …
-
adk ask
… …
atk aek
… …
ark alk
… …
ack avk
… … …
. 1.
-.
, -. 2, rcm, rsm, rtm,
rlpm, rdm, rlrm, rlm, rfm, , , ,
.
, . -
« » , 49 43
atm, apm, asm, alm, a m, avm, adm, advm
, .
, ,
.
-
( )
( ) rcm
rsm
rtm
REF REF REF REF
… REF … …
rlpm rdm
rpm …
… REF rfm
rlrm
REF … … …
atm apm … …
asm alm
… …
a m avm
… …
adm advm
… … …
. 2.
.
. -, .
3, ren, rpn, rsn, rtn, rrn, ran, rdn, rin, ,
, , . -
44
. - apn, azn, asn, ann, a n, adn, amn, afn -
-.
( )
ren
rpn
rsn
REF REF REF REF
… REF … …
rtn rin
ran …
… rrn
rdn
… … …
apn azn
… …
asn ann
… …
acn adn
… …
amn afn
… … …
. 3.
-, -
, [7] --
. -
[4],
« » , 49 45
[4, 6] -.
[5] -
[6] -
.
: - -
; - -
-, -
[6]; -
; - -
-;
- - ,
; -
-;
- - -
.
. -
. --. -
. --
: -
--
[2]; - -
, --
;
- -
, -;
- -, .
, 4 --
, . -
krfk, rck - j kj( kjmax),
.
. -, ,
, -
, . , -, -
-.
-,
[4]. --
Ppn -.
- Spn Ppn,
- Ppn= f(y, Spn - Ppn). -
-, -
-.
-.
, -. -
, -
, -.
- NP-
[6]. -
, -
46 , -
, .
nc nk -
, .
ni(Nik), no(Nok) nw(Nwk),
nk = nk(ni(Nik), nw(Nwk), no(Nok)). (1)
, -
,
. -
. - ( )
-
. -
, . -
nt ,
.
, . , nt -
-.
, , --
, --
. 1.
) .
2. -.
3. -
, -, .2.
4. -
, . 3. -
, [4].
, -
,
-
. -
.
-. -
,
[2].
1. A., ., . : , , : . . – .: , 2001. – 768 .
2. ., . . Oracle8. : – .: - ” ”. – 2000. – 800 .
3. ., . . . . .: , 1989. 424 .
4. . - // “ ”. , – .: « +».
2007, 47, . 269-279. 5. Hehner E.C.R. Practical theory of programming. Springer-Verlag, New York, 1993 – 243 p. 6. Metzger R.C., Zhaofang W. Automatic algorithm recognition and replacement: a new approach to program
optimization / The MIT Press, Cambridge, 2000. 219 p. 7. Woodcock J., Davies J. Using Z. Specification, Refinement, and Proof. C.A.R. Hoare Series editor, 1995 –
390 p.
004.94
.,
.
OPENGPSS GPSS/PC
OpenGPSS GPSS/PC, .
, .
The report deals with questions of computing experiment in distributed discrete-event simulation systems
OpenGPSS and GPSS/PC, their high-quality and quantitative job performances are compared. The problems of experiment distribution by independent part, deployment this part on cluster node, parallel execution and result assembles and here does not influence on the rightness of end-point also were laboured in the report.
.
– -
. -,
[1, 2].
, -, SPEEDES [3], PARASOL [4] Triad.Net
[5]. -, -
GPSS [6]. -
GPSS . -
OpenGPSS [7],
GPSS. -.
: --
GPSS/PC OpenGPSS. -
GPSS/PC - 2.0 Minuteman Software
(http://www.minitemansoftware.com/),
.
-
- GPSS, -
) ( . 1).
. 1. GPSS- GPSS
100 GENERATE 10,5 110 QUEUE QUE1 120 SEIZE PRIB1 130 DEPART QUE1 140 ADVANCE 15,5 150 RELEASE PRIB1 160 TERMINATE 170 GENERATE 5000 400 TERMINATE 1 500 START 1
GPSS- .
- GPSS/PC, -
3 , -
. OpenGPSS
( -),
, [8] -
). OpenGPSS -
. . 2 -.
« » , 49 48
. 2. GPSS/PC 2.0 OpenGPSS
) 5000 5000
PRIB1 ENTRIES 338 334 UTIL (%) 0,997 0,997 AVE.TIME 14,75 15,0
QUE1 MAX. 169 161 CONTENT 168 161 ENTRIES 506 495 ENTR (0)
2 3
AVE.CON 80,85 81,0 AVE.TIME
798,89 816,0
AVE.(0) -
802,06 821,0
-
, --
( ).
OpenGPSS - GPSS- . -
-.
, OpenGPSS , -
GPSS- – -
GPSS- ( . 3).
, --
START, CLEAR RMULT.
PRIB1 QUE1 RES.TXT RESULT.
. 3. GPSS- GPSS
100 GENERATE 10,5 110 QUEUE QUE1 120 SEIZE PRIB1 130 DEPART QUE1 140 ADVANCE 15,5 150 RELEASE PRIB1 160 TERMINATE 170 GENERATE 5000 180 SAVEVALUE MSV01,FR$PRIB1 190 SAVEVALUE MSV02,FC$PRIB1 200 SAVEVALUE MSV03,FT$PRIB1 210 SAVEVALUE MSV04,Q$QUE1 220 SAVEVALUE MSV05,QA$QUE1 230 SAVEVALUE MSV06,QM$QUE1
240 SAVEVALUE MSV07,QC$QUE1 250 SAVEVALUE MSV08,QZ$QUE1 260 SAVEVALUE MSV09,QT$QUE1 270 SAVEVALUE MSV10,QX$QUE1 400 TERMINATE 1 410 CLEAR 420 RMULT 1,2,8,5,8,2,8 500 START 1 510 RESULT RES.TXT,MSV01,01;FR$PRIB1 520 RESULT RES.TXT,MSV02,02;FC$PRIB1 530 RESULT RES.TXT,MSV03,03;FT$PRIB1 540 RESULT RES.TXT,MSV04,04;Q$QUE1 550 RESULT RES.TXT,MSV05,05;QA$QUE1 560 RESULT RES.TXT,MSV06,06;QM$QUE1 570 RESULT RES.TXT,MSV07,07;QC$QUE1 580 RESULT RES.TXT,MSV08,08;QZ$QUE1 590 RESULT RES.TXT,MSV09,09;QT$QUE1 600 RESULT RES.TXT,MSV10,10;QX$QUE1 610 CLEAR 620 RMULT 5,1,6,6,7,1,1 630 START 1 640 RESULT RES.TXT,MSV01,01;FR$PRIB1 650 RESULT RES.TXT,MSV02,02;FC$PRIB1 660 RESULT RES.TXT,MSV03,03;FT$PRIB1 670 RESULT RES.TXT,MSV04,04;Q$QUE1 680 RESULT RES.TXT,MSV05,05;QA$QUE1 690 RESULT RES.TXT,MSV06,06;QM$QUE1 700 RESULT RES.TXT,MSV07,07;QC$QUE1 710 RESULT RES.TXT,MSV08,08;QZ$QUE1 720 RESULT RES.TXT,MSV09,09;QT$QUE1 730 RESULT RES.TXT,MSV10,10;QX$QUE1 740 CLEAR 750 RMULT 2,2,5,2,8,9,3 760 START 1 770 RESULT RES.TXT,MSV01,01;FR$PRIB1 780 RESULT RES.TXT,MSV02,02;FC$PRIB1 790 RESULT RES.TXT,MSV03,03;FT$PRIB1 800 RESULT RES.TXT,MSV04,04;Q$QUE1
49 OPENGPSS GPSS/PC
810 RESULT RES.TXT,MSV05,05;QA$QUE1 820 RESULT RES.TXT,MSV06,06;QM$QUE1 830 RESULT RES.TXT,MSV07,07;QC$QUE1 840 RESULT RES.TXT,MSV08,08;QZ$QUE1 850 RESULT RES.TXT,MSV09,09;QT$QUE1 860 RESULT RES.TXT,MSV10,10;QX$QUE1
RMULT -
. START -. GPSS-
- START, -
. CLEAR
. CLEAR , ,
. RESULT
, -.
, -.
- ( ) – -
, CLEAR RMULT, CLEAR, RMULT, START -
START RESULT. -
: . -
( ) , RMULT
CLEAR. --
. -
( ) – , -
CLEAR, RMULT, START. CLEAR, RMULT “ -
” , .
- ( ) –
, , -, -
. , ijS – j ,
i , ni ,1 , jinj ,1 ,
jin – i . -
i , },..,,{ 21 t
iiniii tttT .
CLEAR, RMULT, START,
, },..,,{ 21 u
iijnijijij uuuU .
, – },..,,{ 21 ciijnijijij cccC .
, ,
},..,,{ 21 uiijnijijij uuuU .
, -
, , -:
.,1,,1,,1},__
,|{}:|{ui
jiimif
ijififijijmijmijm
nmnjniuc
CccUumuK
-
. ijK – i
j :
ij
j
ni KK
ui
1
,
-ijR :
ijm
m
nij RR
ki
1
.
j :
ij
j
ni RR
ui
1
.
iR .
iA = { AgSimiA , pAg
iA Re , AgSpliA , AgSnc
iA , AgTrf
iA , AgPwriA , AgUsr
iA , AgGbriA } – -
i , ni ,1 , n – --
[9]. izB nn ,
i z :
;,0;
',1ziizB ,
ni ,1 , nz ,1 .
« » , 49 50
i ),,,( iiiiii R,KTSAP ,
– ),,,(1
iiiiii
i
nR,KTSAPP .
- –
-, .
OpenGPSS , -
. j GPSS- AgUsr
iA , [10] -
GPSS- -:
.}){},{(),(: ijiijiiiiiAgUsr
i kKsSPKSPA AgRep [11] GPSS-
:
,1:,),,(),(:Re
izizizz
iiipAg
i
ziKKSSPKSPA
B
-
. -
– -
. -: -
– –
.
,
[12] - (
) ( -
). AgSim -
i :
}),,..,,..,,({
}),..,,..,,({:`
21
21
tj
tj
inijiii
inijiiiAgSimi
ttttP
ttttPA
`ijij tt .
AgRep, PL/SQL [13, 14],
REPORT REPORT_DETAIL :
.1:,),()(:Reizziizz
pAgi ziRRPRPA B
AgRep – -
-.
. - GPSS- . 3 , -
, 4. 1 – ,
; 2 3 – ,
CLEAR, RMULT, ” .
. 4. GPSS- GPSS -
… 1 ( ) 1 410 CLEAR 420 RMULT 1,2,8,5,8,2,8 500 START 1 510 RESULT RES.TXT,MSV01,01;FR$PRIB1 520 RESULT RES.TXT,MSV02,02;FC$PRIB1 530 RESULT RES.TXT,MSV03,03;FT$PRIB1 540 RESULT RES.TXT,MSV04,04;Q$QUE1 550 RESULT RES.TXT,MSV05,05;QA$QUE1 560 RESULT RES.TXT,MSV06,06;QM$QUE1 570 RESULT RES.TXT,MSV07,07;QC$QUE1 580 RESULT RES.TXT,MSV08,08;QZ$QUE1 590 RESULT RES.TXT,MSV09,09;QT$QUE1 600 RESULT RES.TXT,MSV10,10;QX$QUE1 610 CLEAR 2 ( ) 2
51 OPENGPSS GPSS/PC
620 RMULT 5,1,6,6,7,1,1 630 START 1 640 RESULT RES.TXT,MSV01,01;FR$PRIB1 650 RESULT RES.TXT,MSV02,02;FC$PRIB1 660 RESULT RES.TXT,MSV03,03;FT$PRIB1 670 RESULT RES.TXT,MSV04,04;Q$QUE1 680 RESULT RES.TXT,MSV05,05;QA$QUE1 690 RESULT RES.TXT,MSV06,06;QM$QUE1 700 RESULT RES.TXT,MSV07,07;QC$QUE1 710 RESULT RES.TXT,MSV08,08;QZ$QUE1 720 RESULT RES.TXT,MSV09,09;QT$QUE1 730 RESULT RES.TXT,MSV10,10;QX$QUE1 740 CLEAR 3 ( ) 750 RMULT 2,2,5,2,8,9,3 760 START 1 770 RESULT RES.TXT,MSV01,01;FR$PRIB1 780 RESULT RES.TXT,MSV02,02;FC$PRIB1 790 RESULT RES.TXT,MSV03,03;FT$PRIB1 800 RESULT RES.TXT,MSV04,04;Q$QUE1 810 RESULT RES.TXT,MSV05,05;QA$QUE1 820 RESULT RES.TXT,MSV06,06;QM$QUE1 830 RESULT RES.TXT,MSV07,07;QC$QUE1 840 RESULT RES.TXT,MSV08,08;QZ$QUE1 850 RESULT RES.TXT,MSV09,09;QT$QUE1 860 RESULT RES.TXT,MSV10,10;QX$QUE1
,
, -
RMULT. GPSS- ,
. 4, , .
-
,
. -
, . :
, .
– -.
. 5. GPSS/PC ,
-. . 5 -
.
. 5. GPSS/PC 2.0 OpenGPSS
) OpenGPSS
1) OpenGPSS
2) -
, -
, 4 8 9 6
5000 5000 5000 5000
FR$PRIB1 ( -
)
997 997 998 998
FC$PRIB1 338 331 341 338
1 . 2 .
« » , 49 52
FT$PRIB1
14 15 14 14
Q$QUE1 168 161 162 158 QA$QUE1 80 79 82 77 QM$QUE1 169 162 162 159 QC$QUE1
506 492 503 596
QZ$QUE1
2 1 1 1
QT$QUE1 (
)
798 811 819 776
QX$QUE1 ( -
)
802 813 820 777
, ,
GPSS- 170 - «GENERATE 5000» - «GENERATE 5000», «GENERATE
10000», «GENERATE 20000»... «START 90000».
. .
1. -,
.
. 1 « » -
, GPSS- ( ) ,
« » .
. , ,
-, 90 . . -
OpenGPSS GPSS/PC.
0
5
10
15
20
25
30
10 20 30 40 50 60 70 80 90, . .
,
GPSS PC (1 )OpenGPSS (1 )OpenGPSS (2 )OpenGPSS (3 )
.1.
1. GPSS OpenGPSS
GPSS-
. -
-.
2. -
-
53 OPENGPSS GPSS/PC
( ),
. 3. -
-
.
.
(16-32).
-
.
1. Richard M. Fujimoto. Parallel and Distributed Simulation Systems. Wiley, 2000. 2. . : . – : ,
2007. – 119 . 3. SPEEDES. http://www.speedes.com. 4. Mascarenhas E., Knop F., Vernon R. ParaSol: A multithreaded system for parallel simulation based on mobile
threads. Winter Simulation Conference, 1995. 5. . , . , . . -
. Proceedings of XXII International Conference “Knowledge-Dialogue-Solution”.– FOI-COMMERCE, Sofia, 2006, pp. 280-287.
6. . GPSS. – .: , 1980. – 593 . 7. . http://www.simulation.kiev.ua. 8. . -
OpenGPSS. – – . 2007. – 5. . 49-53. 9. ., . -
OpenGPSS. . 4, 2006. – .: “ ”, 2006. .123–133.
10. , , . : , , . . . – .: " ", 2001. – 768 c.
11. . OPENGPSS. “
”, 2006. . 264–266. 12. ., . . . – .: , 2003. –
877 . 13. . Oracle . 2. : . . – .:
, 2003. – 848 . 14. . Oracle . 1. : . . –
.: , 2003. – 672 .
004.93(015.7)
., .,
.
. , . -
. -.
SmartBase. -
. In the given work the approach to clustering of documents collections with unknown quantity of clusters is
described. A method of finding matrix of similarity is improved. The method is based on the statistics of key terms occurrence in documents. For quality analysis and finding of limiting values of algorithm, there was used a function of competitive similarity improving. The approach is realized as the application server SmartBase’s application. Implementation details and results of the process are shown. Russian text set is used.
, ,
-
, , ,
( )
. --
, -. ,
, , -
--
.
, --
. -
,
. -
, .
-, -
-.
1.
. , --
, -.
, ,
, .
[1],
-
. , -
[2]. ,
-,
, [3].
,
55
( ),
.
, ,
, .
. -
. -
.
( ) - ( ) ,
, - ( ) -
( ) , , . – -
. -, " – -
( ) -,
".
-, -
, , , .
-,
, , .
, –
. ,
, -. , -
, .
. ,
, ART, SOM [4].
, , k-means, Fo-
rel, FRiS Cluster [5,6,11].
, . -
,
. .
,
. : ,
, ,
; ,
; .
– ,
, . -
. , ,
, , , .
.
-.
. -
[7,8]. ,
,
. -
. -
-.
2.
-
. -
: SRBDCTR ,,,,, (1)
: },...,,...,,{ l21 nTTTTT – ,
( );
« » , 49 56
,...},...,,{ j21 tttTl – , l ;
},...,,{ k21 CCCC – - ( ), k – ;
},...,,{ m21 DDDD – , ,
m – ; },...,,{ k21 bbbB –
( ), i = 1,2,…, k; R C D –
, : Ci C Dj D : (Ci, Dj) R,
; S – , : S1: T C – ,
, , -
Tl Dl Ci, Cj,
. ,
-; S2: C C C – -
; S3: C C C – -
.
3.
, , --
,
--
.
: 1. ,
T * .
2.
. -,
, , -
- [8].
3. - -
, -.
4. .
. 1) . -
Tl T -,
. --
tq. , -
, .
, -, .
2) -. -
. ti Tl -
wli, -
-:
ntf
i
lilr
ilrilt
iltili
ic
etp
TdtTf
tTftTf
tTftpw
)(5.1
1)(
,
350)(1),(
),(),(
)),()(log(
(2)
)t,T(f ilr – ti
Tl; )( lTd – Tl; )( ic tf – ti .
3) - -
, :
,),(22
21
21
21
ii
ii
iii
ww
wwTTsim
ii ww 21 , - ti
T1,T2 .
(3)
(3) , T1, -
57
T2. -. . -
: a) “
--
.” b) -
--
, , -”
“ ” -
. ,
(2) (3) - sim(T1,T2) ~ 1, -
. -
:
,2121),(
i 21
2122
21
21
21ii
ii
ii
ii
iii
wwww
ww
wwTTsim
(4)
ii ww 11 2,1 - , .
4) . FRiS- [10].
T n , k . i
( , ) b . T1 T
),( 1 ibTr .
),(min),(1 11 iii bTrbTr - -
i, ),(2 1 ibTr = ),(min* iii
bTr - -
. FRiS- [10], -
, :
),(2),(1),(2),(1
),(bTrbTrbTrbTr
BTF
(5)
T bi .
FR S- , (6) -
F(B), -:
TTii bTFmBF ),()/1()(
(6)
- FRiS- , -
- r2*. (5), (6)
:
),(*2),(1),(*2),(1
),(*bTrbTrbTrbTr
BTF
(7)
TTii bTFmBF ),(*)/1()(
(8)
4.
, -
: 1. k=1
Cfirst . 2. , -
, t (8)
FR S- F*(B).
3. 1 - m -
. b1 T*1, F* -
. 4. -
- FR S- .
5. 4 - m
, b1. b2 T*2, F* .
6. Cfirst - b1 b2.
7. , -, -
: , r1 -
. 8. k=1 Cfirst
m , k=2 -
. , C1
b1, -.
« » , 49 58
b1 . -, -
C1,C2, -, (6)
FR S- F(T1i,bi). , -
FR S- . C2 -
b2 F(T2i,bi). FR S- ,
.
9. , -
. - (9)
nk
iki
kk
nk
ikkik
bTsimn
middleSim
middleSimbTsimE
1
1
2
),(1
,)),(( , (9)
nk – .
Ek . 10. , 2,
2 -, 2 , -
, -.
, k .
5.
-
SmartBase.
- « » 3 .
26 . --
. ,
,
SnowBall. -
. stopword.xml,
, , " " -
Yandex.Server-FREE-020-3.8.3. 279
. HTML- HTML Parser, -
. web-
, http://www.zn.ua/ -
.
k, n,...,k 1 , n – -. -
( ),
. -
Intel(R) Xeon (TM) CPU 2,40 GHz 2,40 GHz 1GB RAM. -
http . : Intel(R) Core(TM) 2 Duo CPU
2,33 GHz 2GB RAM. 6.
-
, , , -, , ).
26. -
. .
.
(P) (R). -
, , .
-, -
,
. --
, , , , .
-. -
-
59
, -
. : – (
) ( -) ;
Cj – j ( -) ;
;,0
;1, CCCCM jiji
ij
;,0
;1,1 CCI ji
ij
;,0
;1, IIU i
iji
ij
j
.,1
;1, IIK j
ijj
ij
i
---
( ) --
:
iii
jj
ijij
KC
UMP
- ( ) -
-:
jj
jj
ijij
C
UMR
7.
-, :
29.12.2007 26.04.2008, 1515 ;
26.04.2008 27.00.2008, 1543 ;
31.05.2008 5.06.2008, 418 ;
2 : - Cluster
FRiS Cluster [11] ( FRiSCluster). ,
[9], . -
.
, - .
1 - 3 -
, 100 1600. -
.
. 1.
. 2.
, -
1 2 , -, -
, . -
« » , 49 60
- FRiSCluster
. 3.
3
( ) ( ). -
, Cluster , -
. -
. , -
-, ,
.
8.
-:
1. , .
2. , ---
, -
. 3. , -
. 4. , -
.
5.
--
,
. 6. -
, --
. 7. -
SmartBase -
. -
-,
.
1. . . .: - , 1979.- .557. 2. . -
// " - .: ,
2003. - . 211-215. 3. AllaZaboleeva, Yulia Orlova Computer-aided system of semantic text analysis of a technical specifi-
cation//Information Technologies and Knowledge. – 2008. – Vol.2. – P.139-145. 4. Vassilis G. Kaburlasos Unified Analysis and Design of ART/SOM Neural Networks. Heidelberg: -
Springer Berlin, Volume 4507, 2007.-p 80-93 5. MacQueen J. Some methods for classification and analysis of multivariate observations // Proceedings
of the 5th Berkley Symposium on Mathematical Statistic and Probability, University of California Press, 1967, Vol.1, p. 281-297.
6. Peter Grabusts A Study of Clustering Algorithm Application In RBF Neural Networks. //Information technology and management science. - Riga, - 2001. - 5.serija., p.50-57.
7. . ., // . « ». – . - 2008. – . 77-86.
61
8. . ., . ., . . , ,
, // . - ., 2005. - . 412-435.
9. . ad hoc . -.: IMAT, 2007. – 220 .
10. ., ., ., . FRiS- // -
« – – » ( –07), , 2007 . – . 2. – . 67-76.
11. ., ., // « – –
» ( –07), , 2007 . – . 2. – . 77-86.
62-82:658.512.011.56
.
-
. , . -.
The way to increase cluster system efficiency using virtualization is considered. The short review of virtual-
ization types is made, and the most suitable is chosen. The structure of management system for virtual clusters over a physical cluster is proposed.
– - -
, , , -
. -,
, -
[1].
. -
-. -
- [2].
, --
, -
. -,
, --
, --
[3]. , -, -
, --
. -
, -,
. -
– .
-
5%. [4] , -
, , -
---
. -
: , -
, , -
. [5] -
, ---
, PBS Sun Grid Engine,
. -
. , -
-, -
[6].
, -,
)
, . ,
-
« » , 49 63
-:
; -
;
-;
, . -
, , -
, , -
.
--
. , , , -
, -.
: , ,
,
.
, -
-. -,
, . – Bochs, QEMU.
– -, -
, - « », -
--
. ,
, VMware Microsoft Virtual PC,
KVM (Kernel Virtual Machine) – Linux. -
--
-. -
, ,
. -
, .
, [7]. ----
. , , -
, -.
-,
, --
. -, , -
, « » ,
, -. -
. -
Xen. ---
Xen, -.
- –
.
, , Solaris Containers.
Virtuozzo/OpenVZ Linux. --
. ----
. , , ,
-
( -
64
-). -
, -
. , , -
, -, -
.
, . -
-,
, , -
. -
, -.
-. -
---
: - -
; -
.
, -. -
, -. -
, ( ,
Linux), ( , -
Windows Linux).
, ,
, --
. -, , , -
-, .
-, ,
, -
. . 1
.1
,
-, -
, -. . 2
--
,
, -,
, .
. ,
-. -
« » , 49
65
-, -
. , -
, , , -
.
( --
, -,
).
, -.
.2
( 3).
-, ---
, -, -
.
, -, , -
-,
. --
. -
. -. ,
. -.
-,
DHCP ,
.
66
.3
-
. -
, , -. -
, -
-, .
--
-.
, -
.
, , -
, -
.
1. Keahey, K., T. Freeman, J. Lauret, D. Olson. Virtual Workspaces for Scientific Applications, SciDAC
2007 Conference, Boston, MA. June 2007. 2. NAKADA, H., YOKOI, T., EBARA, T., TANIMURA, Y., OGAWA, H., AND SEKIGUCHI, S. The
design and implementation of a virtual cluster management system. In Proceedings of the first IEEE/IFIP International Workshop on End-to-end Virtualization and Grid Management, 2007.
3. Foster, I., T. Freeman, K. Keahey, D. Scheftner, B. Sotomayor, X. Zhang. Virtual Clusters for Grid Communities, CCGRID 2006, Singapore. May 2006.
4. Overhead Matters: A Model for Virtual Resource Management, Sotomayor, B., K. Keahey, I. Foster. VTDC 2006, Tampa, FL. November 2006.
5. Enabling Cost-Effective Resource Leases with Virtual Machines, Sotomayor, B., K. Keahey, I. Foster, T. Freeman. HPDC 2007 Hot Topics session, Monterey Bay, CA. June 2007
6. Combining Batch Execution and Leasing Using Virtual Machines, Sotomayor, B., K. Keahey, I. Foster. HPDC 2008, Boston. June 2008.
7. Jones T. An overview of virtualization methods, architectures, and implementations, IBM, 2006, http://www.ibm.com/developerworks/linux/library/l-linuxvirt.
004.22
.,
.
. , , . , ,
. . The class of scales of notation should be defined by the order of symbol’s calculation. In A class scales of
notation, to which positional scales of notation referred to, the order of symbol’s calculation is dependent and successive. In scales of notation of B class, to which scales of notation of vestigial classes referred to, the order of symbol’s calculation is parallel and dependent. The classification of scale of notation is given.
--.
, --
, . ,
, , -, -
, , -. -
, .
, --
. – ,
, -.
--
, , , -
-.
,
, ( )
, - ( ).
-:
1. - ( ( ))
– )1(n 2. ( ) – ),0( nii 3. -
il 4. i – -
],1[ ij
i lja 5. i- –
],...,,[ 21 iliiii aaaA
6. , -
– . 7. ( ) i- ( -
i – i – ) -
iQ 8. ( -
) - ],,...,,[ 110 nn QQQQB 9. – Q 10. , i – -
– ip 11. ],...,,[ 10 npppp
N -
jjjj
nj
n aaaaaN 0121...][ , ][N –
. ,
:
68
QQaN i
n
i
ji mod ][
0
, -
jia , -
iQ . j
ia --
. -, -
jna ja0 , -
) . -, ,
, ],...,,[ 10 npppp -
-, -
.
, iQ , ip , il -
iA ,
- Q.
- Q -
( , , 1).
,
. , -
,
il , -
iQ , -, . . -
, - ( -
iQ ), - il .
, IQ0 01
ia , ---
,
n0,i 1)1(1
0s
i
ssi lQQ .
, , -:
i
ii Q
Qp 1
, , , . 0|| QQi , -
, -
. - 1|| ip . -
-, ,
.
1nQ[1],A 1,1,1,1,1 iiii ajlpQ . “ ” -
, 1,5,10,50,100,500,1000
I,V,X,..,C,D,M. “ ” 1iQ j
1-iji a a
1iQ j1-i
ji a a .
],,,,,,[A 1 i MDCLXVIpi -
, -
( ). , ,
, -. , -
1ip .
, -
. ppi noi , .
, - IQo , i –
ii pQ .
, ,
-. , -
2p – , 3p – .
-, -
« » , 49
69
, . -
i -
: 1
00
i
kki pQQ
, ,
, ,
. ,
, ,
. , ,
, 121p , ,
2,3 ., 2p =2,3,…; ,
“ ” S- , -
)1()(
limn
nps
s
n )(ns – n S-
.
, )!1(iQi ,
,
)2()!1()!2( i
iipi ni ,0
-
. , ,
, . 1iQi , ,
12
iipi .
, - 2,1 0pQi , -
,
i
pi11 ni ,1 .
, iQ -,
0),1()1(,0,1,0,0
)(isii
ii
iQ
ss
si
S=1,2,3,…
)(
)1(i
ips
si
S=1, -
1,2,3,5,8,13, ., S=2, - 1,1,2,3,4,6,9,13
. -
, -
iQ , 1nQ .
jnQ nr :
nnj
n rQaN 1 , 10 nn Qr .
nr
nQ
11 nnj
nn rQar , nn Qr 10 111 nn
jnn
jn rQaQaN
1nr 1nQ . -,
1Qro . -
N j
ia , j
na . , N
ip , 0p ,
jia ,
ja0 . ,
il -. ip -
ii pl , (
), 01ia -
. 01
ia , -, -
.
70
, . -
12lp ],1,..,1,0,1,...,1,[ llllai -
. --
, .
[-1,0,1]. ip – , - il , ip . ii pl ,
. -
. -
(-1,0,1). - ip -
il , [] ii pl
ii pl , -.
1p 2p -
mpp 21 , . , ,
, -, ,
, a 2 2 2i -2, 2i 4, -2 -8
.
, 2
12,1,
s
Sjji pp ni ,0
1
0
1 ,1),1(
0,0i
js nin
iS
nkniknjnSi
j
i
sss ,0,,1,)1(;1)1(
0
1
02
, 1112 inSS ,
ipi 1, – jp j 1, – -
.
-
, --
. , -
. , --
8421 ,
: p0,2 =2; p1,2 =2; p2,2 =2; p3,2 =1,25; p4,2 =2; p5,2 =2; p6,2 =2; p7,2 =1,25;
. ;20;10;8;4;2;1 2,52,42,32,22,12,0 QQQQQQ
. -
2421 -
2,2,0,5,5; 2,2,0,5,5; . -
, , :
101
3
02, pp
jj
, -
, ip , ,
, -
ip , .
, iQ :
^
iii
ii pmppmQ , ii pm ..2,1
jiQQji pipi ,0||,1||
iQ
. .
- ( nxxx ,...,, 10 ) 11 mod pxX
11 || pxX , 22 mod pxX , nn pxX mod . --
ip .
« » , 49
71
, -,
-, .
-
[0..P-1], n
iipP
0.
PnnQxQxQxX |...| 2211 . , -
- [ .3].
,
. --
, , , -, -
, .
- 8421+3, - 3
8421. - “ ”
, -.
p>>2 ( 162p )
( - 162p
19,17,16,13 4321 pppp
=67184> 162 - 18 ,
16. --
, 5 . -
, -
. 1).
1. -
. .
, , -
-. , -
,
. 2. , -
, -
. 3. -
. , -
. 4.
, ,
.
72
B
,
.1.
1. . . – : . C. 82-96. 2. . . – : . – 1987. – C. 48. 3. . . . – 1974. – 680 . 4. . . : . – 1980. –
336 . 5. . . – : . . –1977. – 287 . 6. . . : – 1984. – 168 . 7. . .
. : – 1970. – 308 . 8. . . : . . , – 1978. – 260 . 9. ., . . . – 2002. – 176 . 10. , . .
3/ – 1987. . 81-86.
681.322
.
, , -
. -.
method of designing pipelined datapaths which are configured in FPGA is proposed. The method
provides the hardware minimization due to the wide utilization of the shift register components. The method is proven at the example of the zigzag scan reordering buffer design.
) -, -
, ,
-. -
, -, -
. , -
. --
, .
Virtex Xilinx - SRL16, -
16- , -
, -
. -
( ) --
. –
SRL16. -
SRL16 . , -
16 , -. -
, -,
. --
.
--
. - ( , ,
.), -
. -
. -.
, , .
, -
, . , , -
, . [1 3] -
-
( ), -
( ). -, -
( ), . -
-, , , -
, , .
. -
- Ki Dj
, [2,3]. Ki = (s,q,t)T -
s , , q t,
.
74
Ki . -
R(Dj) Dj = Ki Kl -,
Ki, Kl . -
|Ks,q L, , - s- , L, L -
, . , -
, [4]. -
, -,
Dj 0 1. -
--
, , [1]. --
, -.
-
s,q. -, -
, - Ki
-. -
, - VHDL [3].
, -.
Ki1 Kj1, Kl1 Dj1 = Kj1 Ki1, Dl1 = Kl1 Ki1,
Ki2 Kj2, Kl2 Dj2= Kj2 Ki2, Dl2= Kl2 Ki2,
( .1, ). , ,
.
-
. , -
, . , -
K
R(K ) ,
ot. - R(Dj1), R(Dl1), R(Dj2), R(Dl2) , -
( .1, ). -.1, .
, –
, -, -, -
.
. , -, .1, .
, .1, . SRL16 -
, -
. -, R(Dj)
. .2 -.1,
, Kl1, Kl2. -
Dj, ot.
s,
. -
, [5].
, --
. – , L .
, [1], , -
. -
, --
, -.
. -,
.
« » , 49
75
-,
, -
. -, -
.
-
[5], -
, - ( . .1, ).
Ki
1
Ki2 y
Kj Kj
Kl
Kl
Dj
Dj
Dl
Dl
o t
p
Kj Kj
Kl
Kl
Ki
Ki
o t
p
) )
.1.
.2. ,
-, -
, .
, , -
, -, -
. ,
VHDL , - [1],
, SRL16, -.
z-
, -264 [6]. 16
-,
-: 0,1,4,8,5,2,3,6,9,12,13,10,7,11,14,15. ,
, --
, -
.
, , .3.
VHDL .
.3.
Virtex4 Xilinx
– 4 , 9 - 12 SRL16 -
. - 900 .
-.
--
. -
--
.
y
Rg
Rg
Rg
Rg
MUX
x,y
)
Rg
Rg
Rg
Rg
MUX
x,y
)
Rg
Rg
y
Kj1 Kj2 Kl2 Kl1
Ki1 Ki2 t
p
76
entity ZZ4x4 is port(CLK : in STD_LOGIC; – START : in STD_LOGIC; – DI : in STD_LOGIC_VECTOR(11 downto 0); – DO : out STD_LOGIC_VECTOR(11 downto 0) ); – end ZZ4x4; architecture ZZ4x4 of ZZ4x4 is type TARR16 is array (0 to 15) of bit_vector(11 downto 0); type TA is array(0 to 15) of natural range 0 to 10; signal sr:TARR16; – SRL16 constant table :TA:=(5,5,3,0,4,8,8,6,4,2,2,6,10,7,5,5); – SRL16 signal fa, addr:natural range 0 to 15; begin FSM:process(CLK,RST) begin – if CLK'event and CLK='1' then if START='1' then addr<=0; else addr<=(addr+1) mod 16; end if; end if; end process; fa<=table(addr); – SRL16 SRL16:process(CLK) begin – SRL16 if CLK'event and CLK='1' then sr<=DI & sr(0 to 14); – end if; end process; DO<= sr(fa); – fa- SRL16 end ZZ4x4;
, -, -
, -
.
-, -
FIFO.
1. . VHDL . – : . – 2003. – 203 .
2. . // “ ”. , . – 2007. – 47. – .221 227.
3. ., ., . - // . .–2002.– .24.– 2.– . 46-59.
4. . . // “ ”. , -. – 2007. – 46. – .62 67.
5. ., . // . – .29. 2007. 2.– .49 62.
6. Richardson I.E.G. H.264 and MPEG-4 Video Compression. Video Coding for Next-generation Multimedia. –Wiley. –2003. –281p.
519.854.2
., .,
., .,
.
,
- ( ),
, , ,
. .
The realization scheme of the informational technology for planning and management in systems with the
network imagination of technological processes and the limited resources (NITPLR) is given. On the basis of the scheme the complexes of sequential interconnected mathematical models which are compatible with the hierarchy of decisions that made on every planning level, and the systems of new highly effective intercon-nected algorithms for the planning tasks solution in current conditions were created. This has let in the first time to solve the task of planning by different optimality criteria in complex.
-
-.
– -,
– , -
. ( 80% )
-
( ). ---
.
n -
J = {J1, J2, …, Jn} ( Ji, i = n,1 , -
). Ji --
.
. -
. , .
. j - lj – -
( , – , ,
), j i I (I – )
i = j i; -
Di. , -
( ,
) , .
-. -
, – ( – -
– -, , -
, ). -,
.
,
…
78
-:
) - ( 1);
) ( 2);
) - Di ( « »,
3); ) -
, i n,1 -
Di ( 4); ) -
-: i max(0, Ci – Di) min, Ci –
Ji, i = n,1 ( 5);
) --
, i n,1 Di -
( 6); ) --
: i |Ci – Di| min 7).
i ,
, - i -
-.
,
-.
1–7 :
;
-
;
. 1–7 NP-
.
--
[1],
. , -
-.
.
, . -
. -
-. -
. –
, ,
. , ,
-,
. -, ,
, . -
, -
-, -» ( ) -
, , ,
[2, 3]. --
, -,
-.
-
. --
, -.
-
« » , 49
79
( ).
: ) -
( 1); )
( 2); ) -
( 3);
) --
( 4); ) -
,
( 5); ) -
, -
( 6); ) -
-
( 7); ) -
( 8);
) -
( 9).
-
, .
. .
. 1).
--
. -
.
. 1. .
, ,
, , ,
( -, -
).
( ) ( .
2). -
, .
- ( -
), -, -
, ( . 4).
…
80
-
, ; -
( ). -
. ---
,
iCPj i
iiiji KR
NNpoLpoL ,
Li – i ; i – ( ); Lp ij – j i
; Ni – , - i ; KRi – -
( --
); CPi – ; )1,max(
iCPjiji LonLon
;
i
iii KR
NNpoNon –
.
, --
, -
. – ,
. . 2 -
, . – ( ) , – -
( – ). -, – -
, ( ).
. 4 -.
-
, , -
, --
. -
, --
( . 5).
. 2.
« » , 49
81
. 3. .
( .1)
. 4.
1-3(1)
D612
D6
D6
2
3
10
11
4
S6
S6
4
5
5
6
D7 213
D3
D3
2
1-3
3
4
D3 1-1- 12 S1
S1
S1
1-2
1-1- 1
37
8
9
D2
D2
D2
4
1-3(2)14
15
16
I2 51
. 5.
, ,
. -, -
, . 2. -
( . 3, 7) -
, -. -
. --
, , -.
. , -, – .
. , -, -
-.
, , .
« ». -
…
82
-, ,
. -,
.
-,
. - ( . 6)
, -
. .
,
. -
. , , ,
. - « » .
, -
. -
(
, -
). -
, ,
. --
, , -. ,
,
.
. --
« » . . ,
-
- [2].
-,
( ).
. 6.
(I1 I3) -
, -
.
- (
) , -
. , i -
i -,
3, 6, - ( 31, 32, 61,
62; ), -
, . 8.
« » , 49
83
. 7. .
( .2)
. 8.
6
.
. --
.
,
. -
( -,
-). -
-
. -. -
-. -
:
: -
…
84
, « » -
« »; , -
-
, , -.
- [3].
-, , -
. :
tP – t–1 -;
ktS – , -
k t; kSj –
ktS ( kSj -
--
); ikR – i R
k, , ;
ikR – i- ; ikR – i- ;
kj = max( ikR , kSj ) – - jk,
Ri k; kCj – -
jk, - Ri k: kCj =
kj + kjl .
0kj –
jk ktS : 0
kj = ktSjk
minmin
kj k, i: 0kj +
kjl ikR . 0
kjC – - jk k
tS : 0kjC =
ktSjk
minmin kCj k, i: ikR – kjl ikR .
) -
. ,
– 1 ). -
-.
-
). -
, .
-
, , ,
-. « »
. , -,
, -, -
, ( , 8-
). 1 mn
[3] . 1. t = 1, P1 = , i = 1, i – i ,
G. kS1 ,
k , mk ,1 .
2. kS1 - jk
. 3. .
: ) jk Pt , -, Pt+1; ) k
tS 1 , mk ,1 , - k
tS , jk, jk 1,
ktS ;
) k;
) t = t + 1. 4. -
1, .2. -
.2. – .
« » , 49
85
1 -. -
2 : . 2
,
kj = max( ikR , kSj ); 0
kj =
ktSjk
minmin kj k, i: 0kj +
kjl ikR .
3 -
, -. -
, . -
,
1 , -
. --
, . ---
( ). ---
, --
. -
, ,
, -
. « -» .
-,
, , . -
-
.
-, -
-. -
, -
, -,
, .
:
-,
,
, « ».
, , ,
( -
) ( . ).
-
. -,
. --
, -.
,
-. -
. ,
, --
, .
-
, , -
. , -
, , .
…
86
. 9.
. 10. .
-
, , :
-
, -;
( -
); -
, ; -.
.
. 10) ,
--
, . -
.
, -. -
: -
1-, – 2- ,
( , ,
) – 3- . -
.
« » , 49
87
-
( 1- ). -
.
( ), -. -
,
. -
, ,
. , -
.
. 11.
. 11. -
. , -, -
-,
.
-
[1] - .
, -
. ,
-,
», -;
, -.
. -
-.
1. ., ., . / “ ”.
, . .: +, 2000.– 33.– .27-33 2. : . / , .;
. . .– . : .; 1991.– 367 . 3. ., . .– .: , 1975.– 359 .
683.519
. .
FOREX
( )
FOREX. , -. -
-.
The subject of the article is an application of the method of prediction with certainty coefficient to a task of
developement of FOREX market mechanical trading system. An example of usage of certainty coefficient is demonstrated and formal procedure for definition of its relevance scale is provided. Quantitative results of effec-tiveness of certainty coefficient usage are provided with regard to prediction of the currency pair trend.
1.
-
-
FOREX. -,
.
-.
2.
( -
- ( ), ( ))
-.
N -, N – -
, -
. N,
-.
--
, -. -
, --
( ).
---
( , .),
).
3 : 1. 2. 3.
--
. T W T ; W
, T W ) -.
N -,
0 N T W . -
( -).
--
( ), . -
, -
, . = 0,
- ( , = 1). -
« » , 49
89-
, A . ,
, A , -
,
, .. . -.
- ( , ]i W T :
1i b bi
bbv
D d Dd
,
biD B -
; bd
, ). biD ,
, --
, - i . 1
bi m mbimm
m
D r Dr
,
mbiD -,
m ; mr m , -
, m -
.
: 1
i mi mimmi
m
P q Pq
,
miP i , m ;
miq m ,
: 1
mi b bmibb
b
q d Dd
.
[1].
3. )
( -): - Mi XXXX ,...,,...,, 21 -
- Mi yyyy ,...,,...,, 21 , i -
; M ; iNiii xxxX ,...,, 21 -
, ( ) N
; iy - i.
i iY Y X ,
i iT T Y , iT , ,
: –1 ( , ), 1 ( ,
), 0 ( ). it -
: 1, 0;
1, 0;0, .
i
i i
yt y .
4.
, - i iY Y X , -
. -, -
, i iD D X , -
0;1 , 0;1iD ,
.
, -.)
. , -
. -, -
( -), ,
.
FOREX
90
, 6 7.
, -, T :
1, 0 ;
1, 0 ;0, .
wi i
wi i i i
Y D DT T Y D D ,(0.1)
0;1wD .
5. : -
, 0iT , . -
iZ iT -:
i i i iZ T t y . (0.2) ,
-: -
. [1.. ]i M :
M
ii
Z Z . (0.3)
6.
Z wD -, -
wD 0;1 , , -
-,
( kZ ) - (
). --
. -
, -:
max max, maxw wkk
D D Z Z .
(0.4)
7. , ,
, - ( .
0iT ), i iT T Y .
-,
. ,
- iY , -
.
( w wi iY Y D D ) -
- ( w w
i iY Y D D ). - T .
(0.1), T :
1, 0 ;
1, 0 ;0, .
w wi i i
w wi i i i
Y D D Y Y
T Y D D Y Y .(0.5)
- maxZ
, : wD wY . (0.4)
: max max ,
max max ,
, max
, max
w wklk l
w wklk l
D D Z Z
Y Y Z Z,
(0.6) k wD ;
l wY . wD , wY
-,
-, -
- wY . ( , )w w
k lZ D Y - ( -
USD/CHF):
91 FOREX
0,00
030
0,00
039
0,00
048
0,00
057
0,00
066
0,00
075
0,00
084
0,00
093
0,00
102
0,00
111
0,00
120
0,00
129
0,00
138
0,00
147
0,00
156
0,00
165
0,00
174
0,00
183
0,01
0,09
0,17-1,0000
-0,8000
-0,6000
-0,4000
-0,2000
0,0000
0,2000
0,4000
0,6000
0,8000 (Z)
. ( abs(Y) )
(D)
0,6000-0,80000,4000-0,60000,2000-0,40000,0000-0,2000-0,2000-0,0000-0,4000--0,2000-0,6000--0,4000-0,8000--0,6000-1,0000--0,8000
. 1. ( , )w w
k lZ D Y ( 1)
0,000300,00060
0,00090
0,00120
0,00150
0,00180
0,01 0,04 0,07 0,10 0,13 0,16 0,19
0,22
-1,0000
-0,8000
-0,6000
-0,4000
-0,2000
0,0000
0,2000
0,4000
0,6000
0,8000
(Z)
. ( abs(Y) )
(D)
0,6000-0,8000
0,4000-0,6000
0,2000-0,4000
0,0000-0,2000
-0,2000-0,0000
-0,4000--0,2000
-0,6000--0,4000
-0,8000--0,6000
-1,0000--0,8000
. 2. ( , )w w
k lZ D Y ( 2) ( 1), -
, wY , -
wD . 0 0w wD Y
, -, -
, . -.
2 , ,
. w
kD klZ wlY -
, , 0, -
- ( 0iT ). , -
FOREX
92
0, 21wD , -
, 1, klZ
[0,02;0,05]wkD , . .
,
. -
Z wD - max[0; ]wD , , wD
max[ ;1]wD . -
wD wY (0.6). -
Z , , . maxZ , (0.6)
, , « ».
-,
wD wY . -
-
, -.
8. :
: USD/CHF : 4 : 6
(2001 2006 .) : 5 ( 1-
- wD wY
:
Y -.
: . : 6N
. N .
: 5 ( . )
: 6 ( . )
T (0.5).
wD wY (0.6) 1 -
-.
( , )w w
k lZ D Y wD :
1 0,01w wk kD D , wY : 1 0,00003w w
l lY Y . :
. 1.
N
(%)
negZ 410
posZ 410
ZN
Z
410 maxZ
410 max
ZZ
(%)
DZ 410
YZ 410
GMDHZ410
2001 1698 -2374 2002 326 42 -16 21 2.82 919 1168 79 345 -712 -2534 2003 272 36 -14 21 5.74 1562 1562 100 1003 -1751 -5472 2004 226 37 -12 20 5.08 1148 1474 78 -780 -1660 -6312 2005 212 42 -14 20 2.76 585 866 68 -3040 -267 -5852 2006 162 52 -15 28 2.69 436 660 66 -4418 -229 -6503 2007 53 42 -12 21 4.36 231 287 80 -4244 -389 -5267
AVG 209 43 -14 22 3.91 814 1003 78 -1856 -835 -5323
N , . 0iT ;
Z ( - wD wY
maxZ , );
1
1 , 0M
negi i
iZ Z Z
M.
;
« » , 49
93
1
1 , 0M
posi i
iZ Z Z
M.
; ZN
;
max
ZZ
, -
wD , wY -
-, ;
DZ T , wD ( wY );
YZ T , wY ( wD );
GMDHZ T , wD , wY ( -
, .
). GMDH Group Method of Data Handly.
: Z maxZ -
3 . -. .
:
. - posZ
negZ . , 2006 -
50%, posZ negZ
- 436 . DZ , YZ , GMDHZ
, -
-.
max/Z Z -, -
- wD , wY , , ,
-.
N Z .
, ZN
( -
2003 .) - (3.9, 1.2). -
N , -
, -
. 9.
, -
, -,
. -
« » , .
-,
. -
. --
. -
, - (
).
1. . // . – 2007. – 2. – . 131-141.
2. ., ., . //
, 2, 2000 ., . 18-26 3. . . - : , 2004. –352 . 4. . . – :
, 2003. – 416 .
681.518.3
.,
., .
-
. . -
, . , .
This article is dedicated to the problems of increasing of efficiency of fault search and elimination in IT-
systems. Methods of threshold values determination for the three-threshold scheme are proposed and analyzed. Method of fault localization in IT-systems that incorporates passive symptom gathering and active probing is proposed. Fault management system which uses these methods was developed.
-
- ( ), -
, -
. ---
( ) [1]. , -
-, -
[2—4], --
, , -
, --
.
-,
-
, --
.
-
--
. , --, -
, ,
-,
, -
. ,
- ( ).
,
, :
, -.
-,
. -
-
-. -
« » , 49
95
, --
, . -
--
, , --
.
[5], -, -
, , SNMP, CMIP .,
-. -
-.
— -
.
-
. --
, -
, ----
. --
-.
----
, , , --
, . -
, , -
. , --
, --
, ,
-. -
,
. -
, . -
-
,
. , -
--
, -, -
.
-
, -
, , , ,
.
, -
. --
, -,
96
. . [6] -
-, -
.
[7] --
,
. , [8] [9],
-, -
, , -
-.
-
, - [10].
-.
- —
[11], , , .
-
.
: - [12, 13] [14, 15].
( , [5], [16]),
.
-, , -
. -
. -
, -
. [12]
-
, -. -
, -
-, -
.
[13] ,
-
. -
, .
[14] --
, -, c
.
--
, [15], --
. -
-, -
.
-
, . ---
.
) , -
-
. --
« » , 49
97
. --
, , -. -
-,
, . -
-
. -
-,
, .
-
, -
. -,
. -
. -, -
-, -
-.
---
. -, -
. -
, . -
-
, -. -
,
.
-, ,
, -.
,
, -
,
-,
, . -
--
.
, --
. -
-,
. --
, -,
-
, . --,
. , -, , -
, Microsoft, - Windows
Microsoft. -
[11]: -.
, -, -
-, -
[5, 16], -
98
. -
, -, ,
SNMP, CMIP . -
, . ,
, -.
--
. -, , SNMP.
-
.
- Si(t), i = 1,…, I — -
, , Li ( .
1, ). , -
, -
, - [6],
. Si(t) -
Li - Di(t) ( . 1, ),
, Si(t) Li.
Di(t) -.
- ( . 1, ) Di(t) Si(t)
Li ( . 1, .). -
--
Di(t) . Si(t) -
-, -,
-,
, , --
. -. -
-, -
-, -
.
. 1.
--
( . 2, ), -
( . 2, ).
. 2.
, 1,iL i I ,
« » , 49
99
— Li(+) Li
(-), -, Di(t) -
Si(t) Li
(+) -,
Si(t) Li(-). -
Li(-) Si(t) Li
(+), Di(t) .
. 2, ) --
, -
, -.
( )
Li(-) Li
(+). , -
--
, --
, - Li, Li
(-) Li
(+). ,
. , -
-
-.
---
, , . -
,
. -
. --.
. -
. -
, -,
-,
.
-
. -, --
, -. -
-
, Li, Li
(-) Li(+), -
-,
, -. -
, ,
. --
. , , -
, - — -
. -
--
, -
. -,
, -,
,
, -.
--
, -,
100
-.
-
. -
[18], -. ---, ---
, . -
, -
, -,
.
--
, -, ,
. --
, .
-. 3 -
iS , ,
.
. 3.
-.
- T t -
/M T t , -
Li i :
,1
M
i mm
i
SL
M,
M — - i ,
, , 1, ,i mS m M — i-
,mt m t -,
, -.
,mt m t
1, ,m M ,i mS , -
, , , , -
, .
. ( )
iL i -
: ( ), ,
( ) 1
( ),
1
,
M
i m i mm
i M
i mm
B SL
B
,( )
,
1, ;0, .
i m ii m
S LB
- ( )
iL i -:
( ), ,
( ) 1
( ),
1
,
M
i m i mm
i M
i mm
B SL
B
,( )
,
1, ;0, .
i m ii m
S LB
. 3 -
: 0,7iL , ( ) 0,79iL ,
( ) 0, 49iL ( . 4). ,
. 5.
« » , 49
101
. 4.
. 5.
--
. ,
Si, - .T R t R
, T ,
. T -, , - [18], -
0b 1b :
, ,1 1
0
2(2 1) 6,
( 1)
R R
i r i rr r
R S rSb
R R
, ,1 1
1
12 6( 1),
( 1)( 1)
R R
i r i rr r
rS R Sb
tR R R
0b – , 1b – , -
0 1( )S t b b t .
. 3) -. 6.
. 6.
iL , ( )
iL ( )iL . -
,
, , -
. 6, -
iS , -. . 6 -
0,58iL , ( ) 0,72iL , ( ) 0, 45iL ( . . 7).
. 8.
. 7.
. 8.
--
, -
. 7. , -
. --
( . 3) -.
- 0a , ka kb :
1
0 ,1
2 ,M
i mm
a SM
1
,1
2 cos( ),M
k i mm
a S km tM
1
1
2 sin( ).M
k im
b S km tM
ka kb , , . 3
0
1( ) ( cos( ) sin( ))
2
K
k kk
aS t a km t b km t
:
102
( ) 0,61 0,022cos(6 ) 0,0227cos(7 )0,079sin(6 ) 0,02sin(8 )0,036sin(9 ),
S t m t m tm t m tm t
-. 9.
. 9.
iL , ( )iL
( )iL ,
, , -
. 8. . 3 -
0,73iL , ( ) 0,81iL , ( ) 0,58iL ( . . 10), -
. 11.
. 10.
. 11.
- ( . 11), , -
-
-.
,
, -
— .
,
, --
, -, -
, -
. -
, -, -
,
, .
[17] -
, . -
-,
, --
. -,
, -, -
. --
. -
. , -
, -
. -
, -,
, , -, . .
« » , 49
103
-
, --
. -
.
. « — — ».
, , - — -
. -, -
. ,
, . . ( -
. .).
.
-
. « – – ». -
, , . ,
, .
. -
-.
. --,
. . « – – ».
, , --
, , .
-,
, .
, -. -
, -
. -
, -
.
, - N . n ,
1,n N , -
: – nC
. . , - n ,
, , 1, ,n n j nC c j J nJ – - n ;
– - ,nF nC ,
, ,n n iF f 1, ,n ni I I – n ;
– - ,nA
nC , , , 1, ,n n k n nA a k K K – - n ;
– n nI J -
, , ,|n n ij n j n iQ q c f ,
, , ,|n ij n j n iq c f
, , ,
1, - | - ;
0, n ij n j n i
jq c f i
– n nJ I -
, , ,|n n ji n j n iP p c f ,
, , ,|n ji n j n ip c f -, j i -
n -:
,
, , ,, , ,
, , ,
|| ;
|n i n
n ij n j n in ji n j n i
n ij n j n if F
q c fp c f
q c f
– n nK J -
, , ,, ,n n kj n j n kV v c a -
, , 0n kj nj nkv c a , , - k j
. , , ,,n kj n j n kv c a
104
( -) k -
j n . --
, , -
. -
,H mc
, ,H H mC c 1, ,Hm M HM –
, - N .
, « -»
, -.
- Pn -
, -
, . . n -
, , 1,n n l nH h l L , -,
n -
. , -
K , . . -
,
,
, ,
, , ,
,, , ,
|
|n j H
n j fn i
n ji n j n ic C
n in ji n j n i
c C
p c fK f
p c f,
,n iK f – -
, ,n if
,n ifC – ,
, .n if nH -
.
. -
Hn. , -
, -, HC H ,
- nF ,
nF nF n -. -
-, -
H nF .
, -.
n ,
H={H1, …, Hn, …, HN}, HC ,
, --
. -
. , -, -
,
. -
Vn - An,
1,n N . , -, -
, , , ,,n kj n j n kv c a .
-
, , -
, , ,
-.
--
, .
:
– – -,
« » , 49
105
, ;
– – -, -
,
; – – , -
; – – ,
, --
, ,
-.
-, , -
--
. ,
. , -,
.
--
, , -, -
-.
--
. 12. , -
, . -
, , .
. 12.
--
-.
-,
. ---
, , -
. --
( , ) -
.
, , -
, ,
. -
, . -
--
, -. -
, . . --
.
106
-, -
. , -
. , . .
,
-. -
, .
-
-,
. ---
, .
-
, . -
.
, . -
, . . , -
. -
, , -,
,
, -.
---
, . . .
. --
,
.
: ,
. , -
. -
.
. -. -
--
. --
,
, : --
. , -
-, , . --
.
.
1. ., ., ., . -// « ». ,
. —2006. — 45. — . 112—126.
« » , 49
107
2. . -// . .
. . — 2007. — 2 (20). — . 73—82. 3. ., ., .
// « ». , . — 2006. — 44. — . 234—239.
4. ., ., ., . // . .
. — 2006. — 3 (37). — C. 33—43. 5. ., . -
// « ». , . — 2007. — 47. — . 113—124.
6. ., . -// « ». ,
. — 2008. — 48. — . 113—120. 7. Cormode G., Muthukrishnan S., Yi K. Algorithms for Distributed Functional Monitoring// Proceedings of
the nineteenth annual ACM-SIAM symposium on Discrete algorithms. — San Francisco, California. — 2008. — P. 1076—1085.
8. Wuhib F., Dam M., Stadler R., Clemm A. Decentralized computation of threshold crossing alerts// Proc. 16th IEEE/IFIP International Workshop on Distributed Systems. — Barcelona, Spain. — 2005. — Vol. 3775. — P. 220—232.
9. Wuhib F., Stadler R., Clemm C. Decentralized service-level monitoring using network threshold crossing alerts// IEEE Communications Magazine. — 2006. — Vol. 44. — 10. — P. 70—76.
10. Dilman M., Raz D. Efficient reactive monitoring// IEEE JSAC. — 2002.— Vol. 20, 4. — P. 668—676.
11. Stallings W. SNMP, SNMPv2, SNMPv3, RMON1 and 2. — 3rd edition. AdisonWesley. — 1998. — 640 p.
12. Steinder M., Sethi A. S. Probabilistic Fault Diagnosis in Communication Systems Through Incremental Hypothesis Updating// Computer Networks.— July 2004.— vol. 45.— no. 4.— pp. 537—562.
13. Appleby K., Goldszmidt G., Steinder M. Yemanja – A Layered Event Correlation Engine for Multi-domain Server Farms// Integrated Network Management Proceedings, 2001 IEEE/IFIP International Symposium on. — 2001.— pp. 329—344.
14. Rish I., Brodie M., Odintsova N., Ma S., Grabarnik G. Real-time Problem Determination in Distributed Systems using Active Probing// Network Operations and Management Symposium. NOMS 2004. IEEE/IFIP.— April 2004.— Vol. 1.— pp. 133—146.
15. Guo J., Kar G., Kermani P. Approaches to Building Self Healing System using Dependency Analysis// Network Operations and Management Symposium. NOMS 2004. IEEE/IFIP.— April 2004.— Vol. 1.— pp. 119—132.
16. ., ., ., . // . .
. — 2006.— 5 (34).— C. 117—124. 17. Tang Y., Al-Shaer E. S. Boutaba R. Active Integrated Fault Localization in Communication Networks//
Integrated Network Management Proceedings. IM’2005. IEEE/IFIP International Symposium on.— May 2005.— pp. 543—556.
18. ., . . — .: « ».— 1989. — 526 .
004.3
.,
.
, -
. -.
The table and algorithmic method of calculation of polynomials based on preliminary coefficient processing
is offered. Possibility of acceleration of calculation of polynomials in comparison with realization of the well-known table and algorithmic methods is shown.
-
, , , -
. -
-. -
.
,
. -,
, , -
.
, , -
. ----
( , .). -
-, -
-.
-
, --
.
,
-, -
. -
.
K 2n , K – -, n –
. K n -
.
- [1-3], -
, ,
. --
. --
, , ,
CSo – Configurable System on Chip). -
Triscend, Xilinx, Altera. -
, FPGA (Field Programmable Gate Arrays),
« » , 49
109
-
. [4] -
-
, --
. , -
mm XXXxP ...)( 2
210 (1)
C tttT 2 , Ct – -, t –
, t – ( 2m ) (m –
). , ---
, , .
.
(1)
).()(...
)()()(
11
32
2101
XFXXXXP
mm
(2)
, )(,,..., 10 XFm
).()(1 XPXP
(2) 1)( jj X
mj ,1 -,
1
01
1
0
11
11
3
0
323
2
0
21201
).(...
...)(
j
i
m
i
iimm
im
iijj
ij
i
iii
i
iii
XFXCXC
XCXCXP
-
( 132 ... mXXX ) :
)....()(
......
)()(
13211
11
2
121
2
1
11
1
10101
mmm
mm
m
ji
jii
ji
jm
i
iii
m
i
iii
m
i
iii
XXXXFXC
CXCX
CXCXP
- 10
kC , 11 kC k
k , :
),...()( 1321
mXXXXF ),1/(mmm
),1/()(1
11 jC
m
ji
jij
jijj
,2/)(2
1111
m
i
iiiC
.1
100
m
i
ii
, (2), . 1. i ( mi ,1 ) -
(i + 1), m+1 )(1 XF . -
, - – ( 2m ) .
0
m
2
12
1 )( X
32 )( X
1)( mm X
X)(1 XF
. 1.
-
i
m
i
n nN i1
1
1
22 , (3)
110
in1 in2 – i.
-. -
. , iii nnn 111 ,
iii nnn 222 , in1 in2 – -, in1 in2 –
.
in1 in2 i ),1( mi :
[.)(log][,)(log]
max22
1max21
XnXn
ii
iii (4)
– -,
. 1m :
[.log]
[,log]
22
1
221
Xn
Xn
i
m
j
ji (5)
, in1 in2 -
i . -,
.
.
i :
)[/1(log] 121 iin ,
)[/1(log] 222 iin ,
i1 i2 – .
(2), )(1 XP - ( 2m ) -
. -, -
1 ,
).2/(2)2/(1 mm n (6)
i ( 1,1 mi )
)[.2(log] 21 mnn i (7)
1 , - i2 -
1 – m. 1
(i+1) )( XZ ii , i2 ,
....
)(1
21121
22
121
21
111
21i
iii
iii
iii
iii
iiii
ii
iii
CZCZC
ZCZZ
, i2 ,
iiii
iii ZiZC 22
111 )1( .
,
.max)( XZ ii (6)
).))(2)(1/((2 max2i
in
i Xmi
, -i
)[.))(2)(1((log] max22i
ii Xminn (8)
i ( mi ,1 )
(4), (7) (8) :
)[.))(2)(1((log]
)[)(log]))[,2(log]
)[)(log]
max2
max22
2
1max21
ii
ii
iii
Xmi
Xnnm
Xnn
(9)
m+1 (5) (7), -
, X m+1 - n . :
[.log]
)[,2(log][log]
2)1(2
2
1
22)1(1
Xnn
mXnn
m
m
j
jm (10)
, --
(3) (9) (10). -,
, - [4], -
.
« » , 49
111
- t ,
t ( 2m )- ,
tttT .
- [5]. -
, -
C ttmT )1)2(log( 2 .
, , [4],
.
-
. -
-
. -
. -.
-
. -, ,
, -, -
, -.
1. : / . . – .: , 1985. – 184 .
2. ., . . . – .: , 1984. – 600 .
3. . : 3- ., . 2. – .: , 1977. – 723 . 4. ., . //
“ ”, I , . – 46. – 2007. – . 206-210. 5. ., . . – .: . ,
1981. – 360 .
004.3
.
, . , -
. .
In the article the method of acceleration of calculations at the level of the simultaneous processing of com-
puter words in the systems, guided the flow of data is offered. The structure of the system, allowing the simul-taneous forming and execution of a few instructions, is considered. Possibility of automatic identification of words of actors and dates on the basis of graph of task is shown.
---
.
. -
, ,
[1]. --
.
, .
, -
. , ,
,
-.
, --
,
. -
. -,
, -
.
, -
.
,
-
, . [2-6].
--
. ---
. , ,
-,
, -
, . -
, -
, . -
-,
-
« » , 49
113
. -
, . -
( ), - ( ), ( )
( ) . --
.
, i , – -
. i -
, - (actor). -
iiiii TNFIA ,,, , (1)
iI – ( ) ; iF –
( ); iN – , i ,
iT – .
.
iiiii TNQID ,,, , (2) iQ – .
iA iD , -
-. -
[2-5].
. ,
, .
-
, -
, -. -
-,
-
-. -, . 1. -
, .
. 1.
-
. 2.
. -
. .
.2.
. 1 ,
, , .
-.
.1 ). ,
.
114 …
-.
-,
, . -
, ---
. , -
[6, 7]. , -. -
.
. -
. -
. , -
, -
.
, --
, -
.
---
, . 3, – -.
, .
jk 2 , ,...3,2,1j (1) (2) -
: iiiiii TNFIMA ,,,, , (3)
iiiiii TNQIMD ,,,, , (4)
iM – , -
.
...
1 ...
...
k2 ...
.3.
iM .
1. , -,
. 2. -
, )
. 3. -
- (
). , -
. -
. --
, , -.
4. -.
, - ( . .3), –
. 5. 0-
, – 1- . (3) -
(4) iM . , -
, ,
« » , 49
115
-.
, , , ,
-.
, ---
. ,
.
( . 4), ,
. .
21 3
754
8 9
10
11
6
a b c d e f
g h
. 4.
.1 - 4, 5, 6 9
-.
, 11, .2
abcdefgh 1425836107911
.3 .
1- . : badcfehg 1;2;3;7 .
: 7, 3, 2, 1. 2- .
: 4;5;6;9 . : 9, 6, 5, 4.
3- .
8 . 8.
4- .
10 . 10.
5- .
11 . 11.
, -
. .4 .5 -
iM , .
iM =0 1, 4, 8, 10, 11, iM =1 – 2 5, iM =2 –
3 6, iM =3 – 7 9. --
. , -
, – .
. , -
.
-,
, -
, -.
.
.
. ---
. , -
116 …
-.
, ,
. -
, , ,
-.
--
. , -.
-.
1. . ., . . . – .: , 2004. – 608 .
2. Dennis J. B., Missunas D. P. A preliminary architecture for basic data flow processor// Proc. 2nd Annual Symp. Comput. Stockholm, May 1975. N. Y. IEEE. – 1975. – P. 126 – 132.
3. Silva J.G.D., Wood J.V. Design of processing subsystems for Manchester data flow computer // IEEE Proc. N.Y. – 1981. – Vol. 128, N 5. – P. 218 – 224.
4. Watson R., Guard J. A practical data flow computer // Computer. – 1982. – Vol. 15, N 2.– P. 51 – 57. 5. Hogenauer E.B., Newbold R.F. Inn Y.T. DDSP – a data flow computer for signal processing/ Proc. Int.
Conf. Parall. Process. Ohio, August 1982. N.Y. // IEEE. – 1982. – P. 126 – 133. 6. Johnson D. Data flow machines threaten the program counter// Electronic Design. – 1980. – N 22. – P.
255 – 258. 7. / ., ., ., -
. / . . – .: . . , 1988. – 224 .
681.3.06
.
GRID
MPLS
GRID .
LSP MPLS. , DR, -
, A way to improve the effectiveness of the GRID systems is proposed. The analysis of changes in channel
bandwidth capacity based on the combined trees information delivery of LSP tunnels MPLS networks is adopted. It has been shown that load distribution of the DR as well as balancing bandwidth capacity of multicast information delivery depends not only on the physical but also on the logical organization of a distributed sys-tem
- GRID , -
-, -
-.
GRID – -, -
[1]. - GRID -
(MPLS – Multiprotocol Label Switching). -
MPLS ,
[2] ( 2 OSI), ,
[3] ( 3 OSI). MPLS -
. MPLS
– -. , -
- MPLS -
, MPLS. -
. -
FEC (Forwarding Equivalence Class),
- LER (Label Edge Router) -
LSR (Label Switching Router). -
, -.
--
MPLS. MPLS
: - (overlay) (peer) .
, UNI, ---
. IP/MPLS ,
. MPLS
--
, --,
« -», .
MPLS -
LSR. GRID
- (MPLS –
Multiprotocol Label Switching). MPLS -
, -
118 GRID MPLS
( 2 OSI), -, -
( 3 OSI). MPLS
--
. MPLS
– -. , -
- MPLS -
, MPLS. -
. -
FEC (Forwarding Equivalence Class),
- LER (Label Edge Router) -
LSR (Label Switching Router). -
, -,
MPLS -.
MPLS – IP-
. -,
LSP ,
.
MPLS - IP-
,
-, -
. -
FEC . -
LSR. - LSR
, , -
. MPLS ,
. , -
-. -
MPLS .
LSP -,
. GRID
,
« – » -. -
IETF (Internet Engineer-ing Task Force) LSP MPLS [4]: RSVP-TE (Resource Reservation Protocol – Traffic Engineering)
LDP (Label Distribution Protocol). -
LSP « – -», , RSVP-TE – »
, LDP . - IP- -
, LDP .
, RSVP-TE
VPN -VPN .
, RSVP-TE
LSP --
. RSVP-TE -
GRID LSP -
VPN. --
IP- . , -
(Root), ,
. , (PE)
« » , 49
119
Root LSP , -
-
(DR) VPN , VPN
. - DR ( . )
- VPN, .
1.
A B C
D
E
F
. : 8.
8
. 34
. : 4. : 7
. 1. VPN
-
. -
, ,
. -
« -» LSP RSVP-TE.
, - GRID
. -
, PE c -
VPN . -
, VPN . -
, GRID ,
VPN -,
». ,
(Root). -
, -
-.
, VPN -
DR, --
CR. b VPN
CR Root - PE DR.
- DR,
VPN , --
PE . (BWLSP)
LSP VPN - « -
». - VPN ,
-:
)12(22*)(log
1
12
iMMNKBWN
i
iLSP ,
(1) : K – - CR;
N – DR, – .
120 GRID MPLS
(1) -
LSP VPN » . -
, N*M*2 DR PE.
)12(2)(log
1
12
iMN
i
i
DR.
, « -
» LSP VPN,
(BWmLSP) :
NMNKBWN
i
imLSP
)(log
12
2
2)(log1 , (2)
: K – - CR;
N – DR; – .
1+log2(N)
PE - « ».
)(log
1
2
2N
i
i
- PE,
LSP VPN. VPN
, (2), Z * N,
Z ( MZ ) ,
DR « » LSP, K W
( KW ).
(BWabLSP) , - VPN A PE VPN B.
- PE .
, DR W * Z .
, LZM .
WZNKBWN
i
iabLSP
)(log
12
2
2)(log1 , (3)
: K – - CR;
N – DR; Z – ,
DR; W – .
- (BW(ab)), -
b VPN VPN -
(2) (3), -:
)**(*)( WZNMKBW ab (4)
, VPN
-. , -
-, -
VPN -. -
, , LSP
VPN -.
, -
, VPN 1000. 20
P - DR. VPN -
,
DR. -
- - LSP VPN
DR (VPN density). , VPN . -
, W 1 1000. -
, Z -
VPN , -
VPN DR.
. 2. 2.
.
« » , 49
121
, -
», – « -
» . 2.b -
DR -.
. 2. GRID
LSP MPLS.
GRID , -
-
LSP
MPLS. , -
DR, --
, -.
1. Ian Foster. What is the Grid? A Three Point Checklist, Argonne National Laboratory & University of Chicago, 2002, p.4
2. ., . 2- . – .: , 2003. - 783 . 3. ., .: . . - .: « », 2002.-384 . 4. A.Chaak, “Quality of service and Traffic Engineering in Consolidated Core and Metro Networks”, Uni-
versity of Toronto, September 2004
681.327.12.001.362
.,
.
-
. , .
, . This article deals with the problem of creating an effective software tool for simulating neural networks. A
brief overview of existing software solutions was given, revealed a list of the deficiencies of discussed means. An Description of proposed multithread neural networks simulator architecture was given and results of its work.
-
. -,
, DSP- , --
.
-, . -
. --
, -, . -
--
, . -
-
.
— ,
-, -
.
--.
: Neural Network Toolbox Matlab, SNNS Joone.
. Matlab [1]. Matlab
neural network toolbox.
, -. , Matlab
, , - -
, (
(PCA), -.). , Neural
Network Toolbox . ,
, -, -
, -.
, . -
. -
. SNNS. SNNS
, -.
— --
. , SNNS -
« » , 49 123
, --
. Joone [2]. Joone
-, ,
, . -
1 -.
, -
-
.
– -
, -
,
-.
PANN -
. -, -
. - ,
. ,
. --
. , ,
-. , -
.
, --
, -, -
-
. -, -
.
-, , -
( -
,
). -
-.
XML, ,
.
- – -
– pann_viewer. --
, -
. -,
-.
, , -
. --
. .
1. --
, -.
– 1 – 9 – 4 – 1.
: y =
1.7179*tanh(2/3 * x). 2 — y = x^2,
[-1;+1] y = sin(x), [-3;+3]. 20 -
. – . 1 -
3D 1 ( -
pann_viewer):
124
. 1. 3D
1
2 3 - ( ,
–
):
. 2.
y=x2 [-1;+1]
. 3.
y=sin(x) [-3;+3]
4 5 :
. 4.
y=x2 [-1;+1]
. 5. -
y=sin(x) [-3;+3]
« » , 49 125
,
. 2.
,
-. -
, -
.
– - 1 – 1000 – 1000 – 1000 – 1. -
-
- 1 8 -
. 8
2.0 1) 4 -
3.2 ( 2). - 10 .
1 6.
. 1. )
1 2 3 4 5 6 7 8
1 89.8
67.9
55.7
50.8
46.7
45.9
47.4
49.2
2 68.6
40.9
36.5
33.8
35.2
36.0
34.2
34.4
. 6.
:
. -
. , -
, -
. ,
,
-,
. , -
, --
-. -
-
.
-
-. -
-:
(
), -
, -;
– -, , -
– -. -
, -., -
, ;
-(SMP). -
, -
. --
« » , – -
.
0 1 2 3 4 5 6 7 8 90
10
20
30
40
50
60
70
80
90
100
1 2
126
-;
– -,
, . --
. -
.
-, -
.
-, ., -
.
-
[3,4] -.
1. Hyun-Yeop Lee, Kyung-Il Moon. Fuzzy Logic and Neural Networks Using MATLAB. 2008. 2. Jeff Heaton. Using Joone for Artificial Intelligence Programming. Developer.com. 2008. – Vol.
2. – P.40-46. 3. . -
.// « -», , 2007.
4. . -.// « », « , -
», 46. – 2007. – . 135-149.
004.3
.,
.
IT – , -
, -. , , -
( ), ,
, . : , , , -
. Virtualization is a key technology which helps to unite applications on various platforms and hardware of the
previous generations with use of smaller number of modern, more powerful servers with low energy consumption. Now, the opportunities offered by this technology can be essentially expanded by its using for satisfaction of re-quirements of a multilevel data storage (storage virtualization), and also for so-called client place virtualization which provides to the user access to working materials from any terminal, including territorially remote. In the article it is considered virtualization of all levels of IT infrastructure: servers, systems of storage, workplaces of a client, an infrastructure of a data processing centre.
-. , , -
-.
, ,
(SOA) (SOI). ,
, -, – -
. SOI : -
; -; -
, , -.
SOI - (DDC). -
, -- -
.
, --
. SOA -
( ) .
, ,
. -,
. , --
. -
. -,
, --
, , , -
,
. -, -
-, « -
» .
, --
,
, , ,
.
128
« » --
, .
-
-. -
-
.
« » 41 .
[10]. , -
, -
. , .
-
. , -
, . ,
, , , -
, , -
5-10%. ( , -
Windows 10%, Unix- -
30%.)
, -,
, , ,
, -.
-, .
-
-. -
. – -
. -,
.
(Windows, Linux, Solaris)
, « ». -
, , .
( ) -,
, .
, -, ( -
) , , .
, -
. --
. , --
.
.
: 1.
( ), ,
, . 2. ,
-.
3. , , -
( -). 4. , -
, -
, -. ,
, ,
, . –
, -, ,
, .
5. ,
-,
« » , 49 129
.
, . ,
-:
Flash- USB - (SAN).
6. -,
. , -
, , SCSI) ,
-. -
- (Windows, Linux .)
. -
, , -, ,
-, -
. ---
. , », , -
-.
, -. ,
, Linux Windows .
, -, -
. ,
. ,
Linux, – Win-dows.
Solaris. , -
, , ,
, .
-, , -
-, –
. , ,
: , . -
.
– . – .
-,
-,
. – -
. ,
, - Web- . -
, ,
. -
, , .
-
Windows. , , -
, - –
. -,
-. -
-
. -
Linux Windows ( ).
. -
, -
MATLAB, --
. ».
130
-
« ».
, – . -
, --
. -,
. , -
, -.
--.
, ,
[1], --
, ,
, -. -
.
GRID-
– , -, -
[2]. -
. - 1.2 PetaFlops [3].
--
, . -
- 90% .
Folding@home, Seti@home, mFluids@home BOINC,
-
.
.
-
. -,
. , -
, , -
.
, -,
, IP. -
, -.
-, -
HTTP proxy, NAT WAN.
-
OPENVPN [4] -. -
, . -
- OPENVPN -
, , .
OPENVPN - MPI
, btl (MPI point-to-point byte transfer layer). -
OPENVPN IP-in-IP (RFC2003), -
. IP-in-IP
.
2)1(NN , – N 2N–1 IP-in-IP
, (MESH) – 2N . -
( ) . IP-in-IP [5]. -
shell- , SSH
.
« » , 49 131
5 -
: , -, , VPN,
. -
, . Windows -
HYPER-V, VMware, Virtual Server, MS VIRTUALPC. , VMware HYPER-V Intel VT,
-. Unix Xen, OPENVZ,
KVM, VIRTUALBOX . FREEBSD [6],
RHL 5, Debian [7] Unix- , -
. -
. -
. VPN -
, -, -
. OPENVPN.
HTTP proxy. - mpd
-. mpd -
, - MPI.
MPI, OPENMP, PVM.
MPICH OPENMPI - MPI.
-
». – Windows, – FREEBSD, 100Mbit , -
– OPENMPI. ,
WAN. mpi-bench-suite, -
[7]. . 1 2 -
MPI.
mpi-bench-suite - [8].
. 1.
MPI
Star 2,19 Reduce 1,697
Star2 2,19 Allreduce 4,744 Chaos 2,06 Broadcast 1,394
Chaos2 2,14 Gather 1,69 Ring 1,95 Allgather 6,23
Ring2 2,13 All-to-all 6,39 Isend & Wait 0,628 Send 0,627 Sendrecv 0,994 Send & Receive 1,25
132
. 1. . Star – « ».
MPI_SEND(RECIEVE). Star2 – « ».
MPI_ISEND(IRECIEVE). Chaos – . -
. Chaos2 – .
.
Ring – « ». -.
Ring2 – « ». -.
Difference – -.
. 2. MPI. ALLREDUCE – N -
-.
REDUCE – N ,
BROADCAST – N -
. GATHER – N
; ,
N (np–1) -.
ALLGATHER – N ; ,
N, N (np–1) ;
GATHER ( ). BCAST ).
ALL-TO-ALL – N ; , -
« » , 49 133
N (P–1) -, P – .
ISEND & WAIT – N -
. BLOCKING SEND – N -
-.
SENDRECV – N .
SEND & RECV – , N .
Difference – - MPI.
(MPI_Barrier) 2,775 ,
1,05 .
MPI [9]. -
WAN , , - OPENVPN. -
.
, -,
. , ----
, SSH -
mpirun - slurm, PBS .
, -
. -, , -
, -, -
-
. , -
, -
-, -
, .
1.
http://www.udec.ntu-kpi.kiev.ua 2. . . – .: «Knowledge», 1999. – 320 . 3. http://top500.org 4. http://openvpn.net 5. OpenNET. — http://www.opennet.ru/docs 6. FreeBSD manual. — http://www.freebsd.org/docs.html 7. The Linux documentation project. — http://tldp.org 8. . – http://paral-
lel.ru 9. ., . . – .: « », 2005. – 226 . 10. " " http://kpi.ua/ru/edu-
cation
004.057.6
.
3D
.
. -.
In article the approaches for optimization of results of three-dimension scanning processing has been proposed.
Those approaches are based on procedure of processing of sets of three-dimension points to different forms of object representations. Proposed approaches has been realizing on system for computer designing in mechanical engineering.
-. -
-,
. -
(CAD -Computer Aided Design)
, . -
(2D) - (3D) ,
CAD. -
- -
, , .
3D- .
--
.
(3D)
-
. 3D- -
. -, -
,
-
. --
. -
3D- --
, -
. -
3D–.
- 3D– -
.
: ,
( -).
. ( -
) -. ,
, - ( . 1) [3]. 1 2 , 3 . -
, 4 . ,
, , -
-. -
, , -
« » , 49 135
. , . -
- ( . 1), -
, .
. -, ,
, -,
. -
. --
.
.1. - .
, -.
-,
-. -
.2.
.2.
, S
X Y, S ,
VP p S , VP
, S.
, -.
.
, - li lj Bij = { R2 | d(x,li) = d(x, lj)},
li lj , -, vi = { x R2 |
d(x, li) d(x, lj), j}, d .
S , -
-. -
, , -
, ,
S. -
. -
, -
. -. ,
, -
. --
, .
-,
.
, , .
DT(p) -
X Y , -, p -
, . -
, -
[5]. -.3.
.3.
, -
3D …
136
„ “. -, -
, , ( .4).
.4. , ) ( )
,
. , -
, . -
.
, .
. -
,
, -, ( . 5) [5].
.5. (a)
) 3D
- S -
S , ,
-. S
S. . -
( ) S,
S -
. .6.
.6. 3D
--
, .
. , -,
[5]. , -
, , [6].
--
-, , , 3D
.
. 7
CAD .
. -
. - 3D -
. -,
.
3D -. -
-.
« » , 49 137
.7.
, . ---
. CAD -
. CAD 3D , -
. -
CAD (STL, DXF).
, .8.
, -
. (X, Y, Z), . -
S. -
S. -
, .
- S.
.
S. ,
, -
, - ( ).
-,
. --
, -
. p+. -
, , -
. p-.
S.
- S -
(p+ p-). -,
S.
, -
, -.
.
1 : -, .
-.
2 : , , -
. -, -
-. -
, .
, - 3D , . 3D
, -.
ASCII -
"endian" -
(little endian, big endian).
3D
CAD
CAD
3D …
138
VS := S.
CH := O
s := S
s CH
p+ := s
VS
n+ := sp+
n+ :=
p- := s VS,
n+
S
s := S
P := p+
p-
DT := S P
DT,
S
: VS: S
CH:
S: s: p+, p-:
P:
DT:
. 8 –
« » , 49 139
CAD .
: 1) – -
, – ; 2) -
. . 9 - CAD ( ).
.9. CAD
-
(MDI – Multiple Document Interface). , CAD
-.
. , CAD ,
.
CAD , -
. - -
( ).
. -
. , -
, .
, -. , -
, -,
. CAD -
. , --
. , -
OpenGL. , , -
. --
, , ,
. - –
, -,
, .
, -. -
– -, , , , -
, .
-.
- 3D- , -
- – . -
--
3D .
1. , . , „ . ”, 2008, , .
2. , . . 3D . Interna-tional Scientific Conference “Uniteh 07”, Volume I, 23-24 2007, , .
3. http://www.cse.ohio-state.edu/~tamaldey/ paper/tcocone/tcocone.pdf 4. http://www.cs.utexas.edu/users/amenta/ pubs/sm.pdf 5. http://interviso.openfmi.net/index.php 6. http://web.mit.edu/manoli/www/ publications/Amenta _Siggraph_98.pdf
519.85
.,
., .
-
, .
For the construction of fuzzy expert system the calculation method of membership functions of fuzzy regres-
sion, calculated on results measuring of fuzzy controlled parameters is offered.
,
, .
( ) [1,2]. n21 x,...,x,x -
– ,
. – , -
.By,Ax
........,..........,Ax,Ax
nn
22
11
(1)
--
, -, -
(1). , -,
(1) -
. , -
-,
[3-4].
– [5].
-.
pS,...,S,S 21
,n,...,,j,p,...,,k),x( j)k( 2121
, -.
-
)(n
)()( x,...,xX 001
0 -
n,...,,j,p,...,,k),X( )(j
)k( 21210 . , -
, -
). -
,)x(),...,x(),x(max)X(
)(n
)k()()k(
)()k()()k(
002
01
0
(2)
.p,...,,k 21
, -, , :
p
k
)()k(
p
k
)()k(
)x(
k)x(k
1
0
1
0
. (3)
(3) --
, -.
-.
« » , 49 141
1. --
nx,...x,x 21 . 2.
.
3. , -,
- ( , -
). 4.
.
, -.
5. , , , -
,
.
, , -
[6]. -
,xa...xaxaay nkkkkk n21 210 (4)
,p,...,,k 21 -
y nx,...,x,x 21 .
)a(jk (4) -
. , – -
)x( j)k( -
,p,...,,k,Wk 21
.p,...,,k,)x(minW j)k(
k 21 (5) -
p
kkp
kk
k .yW
Wk1
1
(6)
: -
(4) -
(5) -.
[7] .
-
j)k( x
- jx ,
kS .
)(X 0 j
)k( x -
)(k XSP 0 , p,...,,k 21 . ,
-. , [7] -
-:
, -.
-
. , -
-
) -
,n,...,,j,x j 21 , -
,S,...,S,S p21 )k(
n)k()k( m,...,m,m 21 , )k(
n)k()k( D,...,D,D 21 ,
p,...,,k 21 .
-,
)(
n)()( x,...,xX 00
10 , -
)0()0(2
)0(1 ,...,, r
kxxx
SP
,)(ˆ
)(ˆ
11
)0(
1
)0(
p
kkr
k
r
krk
r
SPSxP
SPSxP
142
.,...,2,1 ,,...,2,1
,,...,,)(ˆ)0(1
)0(2
)0(1
1
nrpk
xxxSPSP
r
kkr
(7)
(7)
- (7).
1r
),S(Pm)S(PSxPM k
)k(k
k
)(
1
01 (8)
),S(PD)S(PSxPD k
)k(k
k
)(2
1
01 (9)
p
kk
k
)()S(PS
xM1
01
,)S(Pmp
kk
)k(
11 (10)
p
kk
k
)()S(PS
xD1
01
,)S(PDp
kk
)k(
1
21 (11)
)(k
xSPM 0
1
p
kk
k
)(
kk
)(
)S(PSxP
)S(PSxP
M
1
01
01
.)S(Pm
)S(Pmp
kk
)k(
k)k(
11
1 (12)
)(k
xSP 0
1
-: X
Y
].X[D]Y[M]Y[D]X[M]Y[D]X[D]XY[D
2
2
, ,
)(k
xSPD 0
1
k
)k( SPD 21
2
121
2
1
1
4)k(
)k(k
)k(
DmSP
D
2
121
121
1
4)k(
)k(
)k()k(
Dm
Dm
.Dm
)k()k( 12
1
1 (13)
(12) (13) --
)p(A
)(A
)(A m,...,m,m 21 -
)p()()( D,...,D,D 21 --
. -
,D
m)S(Pexp)S(P )k(A
)k(Ak
kA 2
2
.p,...,,k 21 -
, ,
-
)k(jm , )k(
jD , n,...,,j 21 , p,...,,k 21 . , ,
-
-, -
.
-
(4) -
-.
-
, --
.
« » , 49 143
nx,...,x,xX 21 –
,
j)k( x , n,...,,j 21 , p,...,,k 21 ,
, .
,
n
j jjjjjkj
n
jjkkk xxaxaay
j11 1 12
21210
n
j jj jjjdjjj...jkj
ddd
x...xxa11 12 1
2121, (14)
p,...,,k 21 . ,
(14) --
, -
n
j jjjjjkj
n
jjkkk xxaxaay
j11 1 12
21210, (15)
p,...,,k 21 . kja
. - ky ,
p,...,,k 21 , (14).
, -,
j)k( x ,
n,...,,j 21 , p,...,,k 21 . -,
--
. , , - (L-R)- [8-10], -
,ax,axR
,ax,xaL
x ,
L R --
, 00, . a
,
. -, (L-R)- -
L R - ,,a . -
: ,,aBLR .
(L-R)-
--
[4, 5], -.
- uuuLR ,,aU vvvLR ,,aV -
(L-R)- wwwLR ,,aW , vuw aaa , vuw ,
vuw .
uuuLR ,,aU - c (L-R)-
wwwLR ,,aW , aa uw ,
uw , uw .
uuuLR ,,aU - c (L-R)-
wwwLR ,,aW , aa uw ,
uw , uw . -
uuuLR ,,aU vvvLR ,,aV
(L-R)- wwwLR ,,aW ,
vuw aaa , uvvuw aa , uvvuw aa . --
ky , p,...,,k 21 , - (15). -
-.
ix , -
k (L-R)-
144
.xx
R
,xx
L
x
kj
)k(jj
kj
j)k(
j
j)k(
-.
jkjkj xau
,a
xauR
,a
uxaL
u
kjkj
)k(jkjkj
kjkj
kj)k(
jkj
kj)k(
n
jkjk uu
1
.a
xauR
,a
uxaL
u
n
jkjkj
n
j
)k(jkjk
n
jkjkj
k
n
j
)k(jkj
k)k(
1
1
1
1
,
2121 jjjj xxv
.xx
xxvR
,xx
vxxL
v
kj)k(
jkj)k(
j
)k(j
)k(jjj
kj)k(
jkj)k(
j
jj)k(
j)k(
j
jj)k(
1221
2121
1221
2121
21
212121 jjjjkjj vaw
,xxa
xxawR
,xxawxxa
L
w
kj)k(
jkj)k(
jjkj
)k(j
)k(jjkjkjj
kj)k(
jkj)k(
jjkj
kjj)k(
j)k(
jjkj
kjj)k(
122121
212121
122121
212121
21
n
j jjkjjk ww
11 1221
.xxa
xxawR
,xxa
wxxaL
w
kj)k(
jkj)k(
j
n
j jjjkj
n
j jj
)k(j
)k(jjkjk
kj)k(
jkj)k(
j
n
j jjjkj
k
n
j jj
)k(j
)k(jjkj
k)k(
12211 2
21
1 22121
12211 2
21
1 122121
1
1
1
1
, - kkk wuy
k
)k( y
.xxaa
xxaxayR
,xxaa
yxxaxaL
kj)k(
jkj)k(
j
n
j jjjkj
n
jkjkj
n
j jj
)k(j
)k(jjkj
n
j
)k(jkjk
kj)k(
jkj)k(
j
n
j jjjkj
n
jkjkj
k
n
j jj
)k(j
)k(jjkj
n
j
)k(jkj
12211 2
21
1 22121
12211 2
21
1 22121
11
11
11
11
-
nx,,x,xX 21 . --
. k
*)k( X
,xxaa
xxaxaxxaR
,xxaa
xxaxxaxaL
kj)k(
jkj)k(
j
n
j jjjkj
n
jkjkj
n
j jj
)k(j
)k(jjkj
n
j
)k(jkj
n
j jj
*j
*jjkj
kj)k(
jkj)k(
j
n
j jjjkj
n
jkjkj
n
j jj
*j
*jjkj
n
j jj
)k(j
)k(jjkj
n
j
)k(jkj
12211 2
21
1 22121
1 22121
12211 2
21
1 22121
1 22121
11
111
11
111
p,...,,k 21 .
« » , 49 145
, -
,
X , .
. , , -
. , --
. , --
. , --
, .
, -,
--,
– -
. --
RL , -
.
1. . : . . / . – .: , 1989. –
338 . 2. . : . . / . . – .: -
, 1991. – 285 . 3. Zadeh L.A. The concept of linguistic variable and its application to approximate reasoning / L.A. Zadeh //
Information Sciences, 1975. – Vol.4. – pp 199-249. 4. . Fuzzy Technology:
. – .: , – , 2003. – 296 . 5. . , : . . / .
, . , . . – .: – , 2004. – 452 . 6. akagi T. Fuzzy identifications of systems and its application to modeling and control / T. akagi, M.
Cugeno // IEEE Trans. SMC, 1985. – pp 116-132. 7. . -
: . . : 05.13.06; 17.01.02; . 13.03.02 / ; « ». – ., 2001. – 251 .
8. . . : . . / . , . . – .: , 1990. – 286 .
9. . MATLAB fuzzy TECH. / . . – .: – , 2003. – 736 .
10. . . / . -. – .: , 1984. – 206 .
004.942
.,
.
-. ,
, .
Educatory and learning processes were studied in psychological aspect in the article, which gives a possibility to
make a mathematical model. Analysis of methods and models of statistics educational theory was also made, as well as theory of stochastic processes, which are used during processing of results of control and education plan-ning, was studied.
( ),
-, -
.
, ,
-.
, , -
, -, -, ,
.
: , , -
, . -
, -, -
.
,
, , -, . -
– . , . , . , . , . ,
. , . , . , . .
, -
, -.
-
. ,
, . ( ) ( )
( ) ( ). -
, , , -
.
, ,
. – , ,
– , -,
, .
-. -
, . -
. . -
. --
, -
« » , 49 147
, ( . 1). -
. « ».
( ) -, -
. ,
, -
. -, -
, . –
. :
ktkb clog
100 ,
b – - ( )
« »; t – -
; c k – , -
, [1].
. 1.
, , , , . -
-, -
, , , -.
. ,
: cnbeay ,
y – , ( )
; n – ; a – - n ; b c – .
. -,
:
n
n
ecbey ,
y – ; n – ; = a b; c – ( ); b – n .
. :
bcncnay
)()( ,
y – ; n – ; c – ; b – .
. « », -:
)1( bnSR eMH ,
SRH – « », -
, ; – - « »; b – ,
; n – ( -
). -,
. . .
, . -
, -,
. . , . , . , .
, « ».
np , n
1A .
2A . , n 2A , -
np1 . , -
, , [2].
jE n - 1n :
jnjn bpap 1 , ja jb -
. -, jE
148
1A 2A . , :
,)1(;)1(
2222
11111 Ap
App
n
nn
)1,0(, 2121 – , -
)2,1(ip in , nn pp 1 . . -
.
( -) -
. -
. , -
. .
NEEE ,...,, 21 – ,
RAAA ,...,, 21 – , nijp , – , iE
n jA . -
: )1( ,,1, nijnijnij pcpp ,
– )10( c .
. , :
)1)(1(1 nnn ppp , np – ( -
) n ; )10( – .
, ,
[3]. -
np , 0 1n 1 n .
:
bnnpn 1
1 ,
b – .
:
n
n
np)1(
)1(11
,
– . , -
« -» , -
« ». - nx :
.,0
,,1
2
1
A
Axn
1nn xPp –
nnn pxPqA 10,1 . , -
1S , in - 2S -
. np
.,1
;,
2
1
S
Sp
pn
, 1S 2S n :
,112 nSSP – , 10 .
, . -
, , , -
. 2).
.
. 2.
P1
S4 P3
P2
S2
S3
S1
« » , 49 149
-. , . ,
. . , , -
:
. -
, , -
, --
[4]. -
, -:
)](exp[)()(),( itCiBiAitP . -
t,
-. ( ), ( )
( ) [5]
-.
--
. , ,
, .
0t -, 0t
. t
, – -. , -
tetP 1)( ,
– [1/c]. /1 [c].
-, -
. :
1exp1)(
1
attQ
a
.
.
:
,!)(1)(
1
0
a
r
tr
erttQ
– . , -
( = 1).
, , . -
: t
a
eattq
)()()(
1
.
, --
, - > 1 )(t
[6]. .
.
ePP 0 ,
0P – -, – ,
. . . -
-
. – , 0i – -
- ( ).
, « »
10 , , - ,
0iJ ,
n
JJ , n –
. , , ,
, -
t , )(tK . - -
. --
t [7]:
nitKJ 0)( .
150 .
, --
, --
: n
jjnt
1,
i- t =0 -. 1 .
. 2
. -.
-, - t1, t2, ..., tn
1i i -.
j , j = 1,2,... , -
, -
: ,,...,,),...,( 22112,1 jjj zzzPzzzF .
t -
tetF 1)( , -
.
( --
) , ,
( 0) ( 1). - itt 1
0 --
. 1. -.
1 . 2.
,...2,1,...
,...
122112
122111
ntt
nni
n
nnni
n
.
-, .
0,0,1)( tetG , .
-:
2.1,1 DTM
, t
, -, ,
.
, -, -
-,
.
. -,
, --
, . -
: ; -
; ; -
.
: --
; -
-.
--
. ,
-
« » , 49 151
,
. --
, -
: 1. -
,
, , ,
,
. --
. 2. ( )
«
». 3. « -»
--
.
1. ., . . – : , 1988.
– 160 . 2. ., . . – .: , 1962. – 484 . 3. ., . . – : , 1988.
– 160 . 4. ., ., . : . . – .:
, 1969. – 486 . 5. . . – .: , 1980. – 542 . 6. . . . 2. -
. – ., 1974. – 152 . 7. . // -
: I . . . – , 1970. – .225-228.
004.942(045)
, .,
.
802.11. -
. - WI-FI.
. Case and distributing of traffic frames are built in the networks of standard of IEEE 802.11. The parameters of
models fully answer parameters works real of network. A management is modeled a traffic in the network stan-dards of WI-FI. The comparative analysis of algorithms of management turns is conducted from the point of view a management and distributing of traffic.
-
, -
, . -
,
. ,
. , -
-,
) .
,
. ---
( WI-FI).
, .
-
.
,
.
, , -
. ,
. ,
, .
-
. ,
. , -
, -.
, -
. .
FIFO - -
-, .
FIFO -. -
-.
– -. -
. -
« » , 49 153
, , --
, -.
-
, -, -
.
-.
-
-
– . .
: ,
, , -,
, . -
.
, -
. ,
-
. ,
, ,
. , -
, . -
, ( ) – .
-
802.11 ( ) -
. -
. , --
, , WI-FI
. - ( -
) G\G\1 -.
-
, , -.
, -
.
3 -
: 1. WI-FI IEEE 802.11a
5 ; 54 ; 1; 2; “FIFO”.
2. WI-FI IEEE 802.11b 2,4 ; 11 ; 4; 2; “ -
”. 3. WI-FI IEEE 802.11g
2,4 ; 54 ; 4; 2. “
”.
-, -
-.
, -, -
: ,/1 f – -, f – -
. WI-FI:
…
154
0000000002,05000000000/1/1 aa f0000000004,02400000000/1/1 bb f 0000000004,02400000000/1/1 gg f
WI-FI -
: ./1 mb b – , m –
. WI-
FI: 0000000185,054000000/1/1 aa mb0,0000000911000000/1/1 mb mb 0000000185,054000000/1/1 gg mb
, b, g, n - 4 , 802.11 a, g, n – -
. 0000000185,0 10,00000009 4. -
-. 802.11g
0,0000000046, 802.11b – 0,0000000227. .
1000000 -
, :
1. : 802.11 – t = 0,000321 ; 802.11b: t ( 1) = 0,001652 ,
t ( 2) = 0,001619 , t ( 3) = 0,001619 , t ( 4) 0,001589 ;
802.11g: t ( 1) = 0,000328 , t ( 2) = 0,000326 , t ( 3) = 0,000329 , t ( 4) = 0,000329 .
. ,
-:
k
ktC
1/1 ( – , t –
, k= ). -:
802.11 – a= 1 / 0,000321 = 3,115 ;
802.11b – Cb = 2,5 ;
802.11g – Cg = 12,18 M /c;
.
2. -:
. 1.
Wi-Fi 802.11a 3,115 300 Wi-Fi 802.11b 2,5 300 Wi-Fi 802.11g 12,18 M /c 300
:
802.11a ( - “FIFO”) : -
-,
. : -
-. -
“FIFO”,
, . -
;
802.11b ( ) :
-,
. --
802.11a ( 4 ). -
2,5 - Bluetooth -
( ). :
. -
,
« » , 49 155
, . -
, --
. 802.11g :
-,
. -,
,
, . :
. , -
-
, , -.
.
:
- WI-FI, -
; WI-FI -
; -
-.
1. . ., . . . – ., : 2008. – 957 . 2. . . GPSS WORLD. – ., : 2004. – 405 .