I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t
201
9
,
p
p
.
3
2
7
9
~
3
2
8
5
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
9
i
4
.
p
p
3
2
79
-
3285
3279
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Path bas
ed loa
d b
a
la
ncing
f
o
r data
center
net
w
o
rk
s
u
sing
S
DN
V.
Dee
ba
n Cha
k
ra
v
a
rt
hy
,
B
.
A
m
utha
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
SR
M
I
n
s
tit
u
te
o
f
Scien
ce
a
n
d
T
ec
h
n
o
lo
g
y
,
I
n
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
1
6
,
2
0
1
8
R
ev
i
s
ed
Ma
r
2
7
,
2
0
1
9
A
cc
ep
ted
A
p
r
9
,
2
0
1
9
Du
e
to
t
h
e
in
c
re
a
se
in
th
e
n
u
m
b
e
r
o
f
u
se
rs
o
n
t
h
e
in
tern
e
t
a
n
d
t
h
e
n
u
m
b
e
r
o
f
a
p
p
li
c
a
ti
o
n
s
th
a
t
is
a
v
a
il
a
b
le
in
th
e
c
lo
u
d
m
a
k
e
s
Da
ta
Ce
n
ter
N
e
tw
o
rk
in
g
(DCN
)
h
a
s
th
e
b
a
c
k
b
o
n
e
f
o
r
c
o
m
p
u
ti
n
g
.
T
h
e
se
d
a
ta
c
e
n
tre
re
q
u
ires
h
ig
h
o
p
e
ra
ti
o
n
a
l
c
o
st
a
n
d
a
lso
e
x
p
e
rien
c
e
th
e
li
n
k
f
a
il
u
re
s
a
n
d
c
o
n
g
e
stio
n
s
o
f
ten
.
He
n
c
e
th
e
so
lu
t
io
n
is
t
o
u
se
S
o
f
tw
a
r
e
De
f
in
e
d
Ne
tw
o
rk
in
g
(S
DN
)
b
a
se
d
lo
a
d
b
a
lan
c
e
r
w
h
ich
i
m
p
ro
v
e
s
th
e
e
fficie
n
c
y
o
f
th
e
n
e
t
w
o
rk
b
y
d
istri
b
u
ti
n
g
th
e
traff
ic
a
c
ro
ss
m
u
lt
ip
le
p
a
th
s
to
o
p
ti
m
ize
th
e
e
ff
icie
n
c
y
o
f
th
e
n
e
tw
o
rk
.
T
ra
d
it
io
n
a
l
l
o
a
d
b
a
lan
c
e
rs are
v
e
ry
e
x
p
e
n
siv
e
a
n
d
in
f
lex
ib
le.
T
h
e
se
S
DN
lo
a
d
b
a
lan
c
e
rs
d
o
n
o
t
re
q
u
ire
c
o
stly
h
a
rd
w
a
re
a
n
d
c
a
n
b
e
p
ro
g
ra
m
m
e
d
,
w
h
ich
it
m
a
k
e
s
it
e
a
sie
r
to
im
p
lem
e
n
t
u
se
r
-
d
e
f
in
e
d
a
lg
o
rit
h
m
s
a
n
d
l
o
a
d
b
a
lan
c
in
g
stra
teg
ies
.
In
th
is
p
a
p
e
r,
w
e
h
a
v
e
p
ro
p
o
se
d
a
n
e
ff
icie
n
t
lo
a
d
b
a
lan
c
in
g
tec
h
n
iq
u
e
b
y
c
o
n
sid
e
rin
g
d
if
fe
re
n
t
p
a
ra
m
e
ter
s
to
m
a
in
tain
th
e
lo
a
d
e
ff
ici
e
n
tl
y
u
sin
g
Op
e
n
F
lo
w
S
w
it
c
h
e
s c
o
n
n
e
c
ted
to
ON
OS
c
o
n
tr
o
ll
e
r.
K
ey
w
o
r
d
s
:
L
o
ad
b
alan
cin
g
Op
en
Flo
w
SDN
So
f
t
w
ar
e
-
d
ef
i
n
ed
n
et
w
o
r
k
Co
p
y
rig
h
t
©
2
0
1
9
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
V.
Dee
b
an
C
h
a
k
r
a
v
ar
th
y
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
i
n
ee
r
in
g
,
SR
M
I
n
s
tit
u
te
o
f
Sc
ien
ce
a
n
d
T
ec
h
n
o
lo
g
y
,
SR
M
Na
g
ar
,
Kattan
k
u
lat
h
u
r
,
T
am
il Na
d
u
6
0
3
2
0
3
,
I
n
d
ia.
E
m
ail:
v
d
ee
b
an
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
W
ith
in
cr
ea
s
i
n
g
l
y
tr
a
n
s
m
itte
d
d
ata
in
d
ata
ce
n
ter
n
et
w
o
r
k
s
,
th
e
tr
af
f
ic
e
x
ch
a
n
g
ed
b
etw
ee
n
d
ata
ce
n
ter
n
et
w
o
r
k
s
w
i
tch
e
s
g
r
e
w
r
ap
id
l
y
,
w
h
ic
h
co
u
ld
lead
to
co
n
g
e
s
tio
n
a
n
d
lead
to
a
l
o
n
g
d
ela
y
a
n
d
lo
w
p
er
f
o
r
m
a
n
ce
i
n
d
ata
ce
n
ter
n
et
w
o
r
k
s
.
T
h
is
p
r
o
b
le
m
o
f
co
n
g
esti
o
n
n
ee
d
s
to
b
e
ad
d
r
ess
ed
u
r
g
e
n
tl
y
u
s
i
n
g
a
n
ef
f
icien
t p
ath
o
p
ti
m
iza
tio
n
al
g
o
r
ith
m
to
i
m
p
r
o
v
e
t
h
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
d
ata
ce
n
ter
n
et
w
o
r
k
s
.
L
o
ad
b
alan
cin
g
i
s
w
id
el
y
i
m
p
l
e
m
en
ted
to
d
is
tr
ib
u
te
lo
ad
o
n
d
if
f
er
e
n
t r
eso
u
r
ce
s
,
i
t is
u
s
ed
t
o
o
p
tim
i
s
e
th
e
n
et
w
o
r
k
a
n
d
p
r
ev
en
t
co
n
g
esti
o
n
.
L
o
ad
b
alan
ci
n
g
i
s
o
n
e
o
f
th
e
cr
u
cial
i
s
s
u
es,
w
h
ic
h
d
i
v
id
es
t
h
e
w
o
r
k
lo
ad
d
y
n
a
m
icall
y
a
m
o
n
g
t
h
e
s
er
v
e
r
s
b
y
i
m
p
r
o
v
in
g
t
h
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
s
y
s
te
m
[
1
]
.
T
h
er
e
ar
e
t
w
o
t
y
p
e
s
o
f
ne
t
w
o
r
k
lo
ad
b
alan
cin
g
tech
n
iq
u
es
n
a
m
el
y
s
er
v
er
an
d
lin
k
.
T
h
e
s
er
v
er
lo
ad
b
alan
ce
r
s
en
d
s
th
e
lo
ad
to
th
e
o
th
er
s
er
v
er
s
w
h
ic
h
ar
e
j
u
s
t
r
ep
lica
o
f
m
ai
n
s
er
v
er
.
T
h
is
ap
p
r
o
ac
h
is
v
er
y
co
s
tl
y
a
n
d
it
h
as
s
ca
lab
il
it
y
is
s
u
e
s
to
o
.
A
n
o
t
h
er
ap
p
r
o
ac
h
is
li
n
k
lo
ad
b
alan
ci
n
g
i
n
th
i
s
w
e
j
o
in
m
u
ltip
le
p
h
y
s
ical
li
n
k
s
w
it
h
v
ir
tu
a
l
li
n
k
s
f
o
r
tr
an
s
m
itti
n
g
th
e
d
ata.
MP
L
S
T
r
af
f
ic
E
n
g
i
n
ee
r
i
n
g
is
an
a
n
o
th
er
ap
p
r
o
ac
h
o
f
lin
k
lo
ad
b
alan
cin
g
,
b
u
t
it
is
v
er
y
co
m
p
le
x
b
ec
a
u
s
e
e
v
er
y
h
o
s
t a
n
d
s
w
i
tch
in
th
e
t
u
n
n
el
i
s
co
n
f
ig
u
r
ed
f
ir
s
t
an
d
e
v
er
y
d
ev
ice
d
ir
ec
ts
t
h
e
w
o
r
k
lo
ad
to
th
e
p
ath
.
MP
L
S
T
E
is
n
o
t
u
s
u
all
y
d
ep
lo
y
ed
b
ec
au
s
e
o
f
it
’
s
co
m
p
lex
it
y
.
A
b
o
v
e
m
en
t
io
n
e
d
m
o
d
els
ar
e
u
s
e
f
u
l
b
u
t th
e
y
ar
e
n
o
t p
r
o
p
er
s
o
lu
tio
n
f
o
r
ev
er
y
p
r
o
b
le
m
.
A
p
ath
co
n
n
ec
ti
n
g
t
h
e
s
o
u
r
c
e
an
d
t
h
e
d
esti
n
atio
n
n
o
d
e
i
n
clu
d
e
s
n
u
m
er
u
o
s
s
w
itc
h
es
a
n
d
li
n
k
s
to
co
n
n
ec
t
th
o
s
e
n
o
d
es.
T
h
e
tr
ad
itio
n
al
p
at
h
lo
ad
b
ala
n
cin
g
p
r
o
to
co
l
tr
an
s
f
er
s
d
ata
b
y
p
at
h
f
ir
s
t,
s
o
m
eti
m
es
th
is
cr
ea
te
co
n
g
esti
o
n
o
n
s
i
n
g
le
n
o
d
e.
P
ath
lo
ad
b
alan
cin
g
ca
n
b
e
g
r
ea
t
o
v
er
co
m
e
o
f
th
i
s
p
r
o
b
le
m
i
n
w
h
ich
w
e
ca
n
u
s
e
SD
N
an
d
d
is
tr
ib
u
te
t
h
e
w
o
r
k
lo
ad
b
y
u
s
i
n
g
d
if
f
er
en
t p
ar
am
eter
s
.
SDN
[
2
]
is
an
ap
p
r
o
ac
h
to
p
r
o
v
id
e
n
et
w
o
r
k
m
a
n
ag
e
m
e
n
t
a
n
d
to
in
cr
ea
s
e
t
h
e
e
f
f
ic
ien
c
y
o
f
n
et
w
o
r
k
th
r
o
u
g
h
p
r
o
g
r
a
m
s
.
S
DN
h
as
l
o
g
icall
y
ce
n
tr
alize
d
t
h
e
i
n
tell
ig
en
ce
in
th
e
n
et
w
o
r
k
[
3
]
b
y
d
is
co
n
n
ec
tin
g
t
h
e
fu
r
t
h
er
p
r
o
ce
s
s
o
f
p
ac
k
ets,
t
h
a
t
in
telli
g
en
ce
is
t
h
e
al
g
o
r
ith
m
s
u
s
ed
i
n
t
h
e
co
n
tr
o
ller
w
h
ich
c
o
n
tr
o
ls
t
h
e
r
o
u
t
in
g
p
r
o
ce
s
s
o
f
p
ac
k
ets.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t
201
9
:
3
2
7
9
-
3
2
8
5
3280
Op
en
Flo
w
[
4
]
is
th
e
m
a
in
te
ch
n
o
lo
g
y
t
h
at
is
u
s
ed
in
SD
N,
it
is
u
s
ed
to
p
r
o
v
id
e
co
m
m
u
n
icat
io
n
b
et
w
ee
n
SDN
co
n
tr
o
ller
an
d
an
d
o
th
er
f
o
r
w
ar
d
in
g
la
y
er
,
it
m
ain
tain
s
th
e
f
lo
w
o
f
n
et
w
o
r
k
tr
a
f
f
ic.
SDN
co
n
tr
o
ller
o
p
tim
i
s
e
th
e
f
lo
w
o
f
n
et
w
o
r
k
tr
af
f
ic
b
y
m
a
k
i
n
g
d
ec
is
io
n
s
o
n
th
e
p
ath
li
n
k
s
.
Di
f
f
er
e
n
t
SDN
co
n
tr
o
ller
s
s
u
c
h
a
s
ON
OS,
P
O
X
[
5
]
,
R
YU,
NOX
[
6
]
,
Flo
o
d
l
ig
h
t
[
7
]
etc.
W
e
ar
e
u
s
i
n
g
O
N
OS
[
8
]
co
n
tr
o
ller
i
n
o
u
r
ex
p
er
i
m
e
n
t.
Min
i
n
et
[
9
]
is
u
s
ed
f
o
r
cr
ea
tin
g
v
ir
t
u
al
to
p
o
lo
g
y
n
et
w
o
r
k
a
n
d
tr
an
s
f
er
p
ac
k
ets
t
o
d
if
f
er
en
t
s
w
itc
h
es,o
n
o
s
co
n
tr
o
ller
is
u
s
ed
to
c
o
n
tr
o
l
th
e
n
et
w
o
r
k
an
d
to
s
h
o
w
th
e
g
u
i
o
f
th
e
to
p
o
lo
g
y
an
d
h
o
w
t
h
e
p
ac
k
ets tr
a
n
s
f
er
s
f
r
o
m
o
n
e
h
o
s
t to
an
o
t
h
er
.
I
n
o
r
d
er
to
s
o
lv
e
t
h
e
co
n
g
es
tio
n
p
r
o
b
lem
o
n
a
d
ata
ce
n
ter
n
et
w
o
r
k
,
n
et
w
o
r
k
ad
m
in
i
s
tr
ato
r
s
ca
n
t
h
er
ef
o
r
e
in
s
tall e
f
f
icie
n
t p
ath
o
p
ti
m
izatio
n
alg
o
r
it
h
m
s
w
it
h
O
p
en
Flo
w
p
r
o
to
co
l to
co
n
tr
o
l
p
ath
s
o
f
n
e
w
f
lo
w
s
an
d
ch
an
g
e
p
ath
s
o
f
s
tr
ea
m
s
d
u
r
i
n
g
th
eir
tr
a
n
s
m
is
s
io
n
s
.
SDN
-
b
ased
p
ath
o
p
tim
izatio
n
al
g
o
r
ith
m
s
ar
e
ab
le
to
r
o
u
te
f
lo
w
s
f
r
o
m
a
d
ata
ce
n
ter
n
et
w
o
r
k
ac
co
r
d
in
g
to
th
e
lin
k
lo
ad
,
b
ec
au
s
e
th
e
ce
n
tr
al
ized
SDN
co
n
tr
o
ller
is
ab
le
to
o
b
tain
lo
ad
s
tatis
tics
f
o
r
ea
ch
d
ata
ce
n
ter
s
w
itc
h
.
SDN
i
s
t
h
e
b
e
s
t
ap
p
r
o
ac
h
f
o
r
p
ath
lo
ad
b
alan
ci
n
g
,
th
i
s
p
ap
er
p
r
esen
ts
a
n
e
f
f
ec
ti
v
e
tec
h
n
iq
u
e
f
o
r
p
at
h
lo
ad
b
alan
cin
g
,
in
th
i
s
e
x
p
er
im
en
t
w
e
u
s
e
s
co
n
tr
o
ller
to
c
o
n
tr
o
l
th
e
n
et
w
o
r
k
b
y
u
s
i
n
g
d
if
f
er
e
n
t
p
ar
a
m
eter
s
.
T
h
en
th
e
co
n
tr
o
ller
tr
an
s
m
i
ts
th
e
d
ec
is
io
n
m
ad
e
b
y
co
n
tr
o
ll
e
r
to
th
e
f
o
r
w
ar
d
s
w
itc
h
es.E
x
p
er
im
e
n
tal
r
es
u
lt
s
s
h
o
w
n
th
at
t
h
i
s
tech
n
iq
u
e
h
as
in
cr
ea
s
ed
th
e
e
f
f
icie
n
c
y
o
f
t
h
e
n
et
w
o
r
k
lo
ad
b
alan
cin
g
.
2.
RE
L
AT
E
D
WO
RK
I
n
SD
N,
t
h
er
e
h
a
s
b
ee
n
m
an
y
s
t
u
d
ies
d
o
n
e
i
n
p
ath
lo
ad
b
al
an
cin
g
.
Fo
r
ex
a
m
p
le,
P
lu
g
-
n
-
s
er
v
e
[
1
0
]
s
y
s
te
m
,
it
m
i
n
i
m
izes
t
h
e
r
esp
o
n
s
e
ti
m
e
ta
k
en
b
y
t
h
e
clie
n
t
to
s
er
v
er
.
W
h
en
a
r
eq
u
est
co
m
e
s
,
th
e
d
ec
is
io
n
s
ar
e
m
ad
e
b
y
co
n
tr
o
ller
a
n
d
cr
ea
te
f
o
r
w
ar
d
i
n
g
r
u
les
th
a
t
h
a
n
d
le
t
h
e
r
eq
u
est
o
f
a
clie
n
t
f
r
o
m
a
s
er
v
er
.
P
lu
g
-
n
-
s
er
v
e
s
y
s
te
m
h
as
t
w
o
m
ai
n
ad
v
a
n
ta
g
es
f
lex
ib
il
it
y
a
n
d
r
ed
u
ce
d
r
esp
o
n
s
e
ti
m
e
b
u
t
it
h
as
th
e
b
i
g
g
est
d
is
ad
v
an
tag
e
o
f
s
ca
lab
ilit
y
,
it
h
a
s
s
ca
lab
ilit
y
is
s
u
es.
T
h
e
m
o
r
e
s
ca
lab
le
s
o
lu
tio
n
h
a
s
b
ee
n
p
r
o
v
id
ed
b
y
W
an
g
et.
al
[
1
0
]
.
T
h
is
alg
o
r
ith
m
co
m
p
u
tes
th
e
r
u
les
o
f
w
i
ld
ca
r
d
f
o
r
r
ed
u
cin
g
t
h
e
d
atalo
ad
o
n
SDN
co
n
tr
o
ller
.
I
t
h
as
t
w
o
alg
o
r
ith
m
s
“
p
ar
titi
o
n
in
g
al
g
o
r
ith
m
”
an
d
“tr
an
s
itio
n
i
n
g
alg
o
r
i
th
m
”,
p
ar
titi
o
n
in
g
al
g
o
r
ith
m
is
u
s
ed
to
d
eter
m
i
n
e
th
e
w
i
ld
ca
r
d
r
u
les
a
n
d
“
tr
an
s
itio
n
i
n
g
al
g
o
r
ith
m
”
i
s
u
s
ed
t
o
ch
an
g
e
th
e
r
u
les
o
f
w
ild
ca
r
d
w
it
h
r
esp
ec
t
to
ch
an
g
e
o
f
t
h
e
p
o
lic
y
o
f
lo
ad
b
alan
cin
g
.
T
h
ese
alg
o
r
ith
m
s
ar
e
f
o
r
s
in
g
le
n
et
w
o
r
k
w
h
e
n
th
e
n
et
w
o
r
k
g
ets
d
is
tr
ib
u
ted
,
w
e
h
a
v
e
to
u
s
e
d
is
tr
ib
u
ted
SDN
ap
p
r
o
ac
h
w
h
ic
h
h
a
s
g
i
v
e
n
b
y
Ko
en
er
an
d
Kao
[
1
1
]
.
T
h
ey
d
e
v
elo
p
ed
a
lo
ad
b
alan
cin
g
tech
n
iq
u
e
to
h
a
n
d
le
m
u
ltip
l
e
s
er
v
ices b
y
u
s
i
n
g
m
u
ltip
le
wo
r
k
in
g
Op
en
F
lo
w
.
Fo
r
ex
a
m
p
le,
o
n
e
f
lo
w
is
h
an
d
lin
g
in
ter
n
et
b
r
o
w
s
i
n
g
a
n
d
an
o
th
er
f
lo
w
is
h
a
n
d
lin
g
e
-
m
ail
tr
a
n
s
f
er
r
in
g
.
T
h
is
m
et
h
o
d
h
a
s
i
n
cr
ea
s
ed
t
h
e
e
f
f
ic
ien
c
y
o
f
th
e
n
et
w
o
r
k
lo
ad
b
alan
cin
g
b
y
d
i
s
tr
ib
u
ti
n
g
th
e
w
o
r
k
lo
ad
to
m
u
ltip
le
Op
en
Flo
w
s
.
Har
d
ee
p
Up
p
al
an
d
Dan
e
B
r
an
d
o
n
[
1
2
]
h
av
e
d
e
v
elo
p
ed
an
Op
en
F
lo
w
tec
h
n
o
lo
g
y
-
b
ase
d
lo
ad
b
alan
ce
r
ar
ch
itectu
r
e
t
h
at
r
ed
u
ce
s
co
s
ts
a
n
d
g
i
v
es
f
l
ex
ib
ilit
y
.
I
t
co
n
tai
n
s
t
h
r
ee
r
an
d
o
m
lo
ad
b
alan
ce
p
o
licies
to
r
an
d
o
m
l
y
s
elec
t
t
h
e
r
eg
is
ter
ed
s
er
v
er
s
,
a
r
o
u
n
d
r
o
b
in
lo
ad
b
alan
ce
p
o
licy
to
r
o
tate
th
e
r
eg
is
ter
ed
s
er
v
er
s
f
o
r
th
e
p
u
r
p
o
s
es
o
f
s
er
v
in
g
t
h
e
lo
ad
-
b
ased
lo
ad
b
ala
n
ce
alg
o
r
ith
m
,
to
s
elec
t
th
e
s
e
r
v
er
w
it
h
th
e
lo
w
es
t
p
o
s
s
i
b
le
lo
ad
.
T
h
is
m
eth
o
d
ca
n
h
a
v
e
s
ca
lab
ilit
y
is
s
u
e
s
d
u
e
to
co
n
s
ta
n
t
c
h
an
g
e
a
n
d
in
cr
ea
s
ed
s
er
v
ices.
C
h
o
u
et.
al
[
1
3
]
.
d
ev
elo
p
ed
a
g
en
et
ic
alg
o
r
it
h
m
lo
ad
b
alan
ci
n
g
s
y
s
te
m
w
h
ic
h
p
r
e
-
d
ec
id
e
th
e
r
u
les
to
tr
an
s
m
it
th
e
d
ata
to
th
e
s
w
i
tch
e
s
in
ad
v
an
ce
.
I
n
th
e
co
n
te
x
t
o
f
i
n
ter
-
DC
W
A
N,
B
4
an
d
SW
AN
[
1
4
,
1
5
]
ar
e
u
s
in
g
S
DN.
B
y
co
o
r
d
in
atin
g
th
e
s
en
d
i
n
g
r
ate
o
f
s
er
v
ice
s
a
n
d
ce
n
tr
all
y
co
n
f
ig
u
r
i
n
g
t
h
e
n
et
w
o
r
k
d
ata
p
la
n
e,
th
e
s
e
en
ab
l
e
ef
f
ec
tiv
e
in
ter
D
C
w
AN
s
to
b
e
tr
an
s
m
i
tted
u
s
i
n
g
a
m
a
x
-
m
i
n
-
m
et
h
o
d
.
A
lt
h
o
u
g
h
B
4
d
ev
elo
p
s
cu
s
to
m
s
w
itc
h
es
an
d
p
r
o
ce
s
s
es
to
in
te
g
r
ate
cu
r
r
en
t
r
o
u
ti
n
g
p
r
o
to
co
ls
w
i
th
i
n
t
h
e
S
D
N
en
v
ir
o
n
m
e
n
t.
L
o
n
g
et
a
l
.
[
1
6
]
p
r
o
p
o
s
e
a
lo
ad
b
alan
ce
r
o
u
tin
g
alg
o
r
it
h
m
f
o
r
Op
en
Fl
o
w
-
en
ab
led
d
atac
en
ter
n
e
t
w
o
r
k
s
(
L
A
B
E
R
I
O)
to
s
o
lv
e
th
e
s
tatic
r
o
u
ti
n
g
p
at
h
p
r
o
b
lem
.
T
h
e
ce
n
tr
alize
d
co
n
tr
o
ller
w
il
l
b
e
n
o
ti
f
ied
a
n
d
h
e
lp
s
ch
ed
u
le
t
h
e
g
r
ea
test
f
lo
w
in
t
h
e
b
u
s
ie
s
t
h
o
p
w
it
h
th
e
m
ax
i
m
u
m
r
e
m
ai
n
d
er
ca
p
ac
it
y
s
tr
ateg
y
if
t
h
e
li
n
k
lo
ad
d
etec
to
r
ex
ce
ed
s
t
h
e
L
A
B
E
R
I
O
tr
ig
g
er
th
r
es
h
o
ld
.
Ho
w
e
v
er
,
all
t
h
e
s
ch
e
m
e
s
w
e
h
a
v
e
d
escr
ib
ed
ab
o
v
e
r
el
y
o
n
tr
a
f
f
ic
d
e
m
a
n
d
i
n
f
o
r
m
atio
n
,
b
u
t
f
o
r
m
an
y
n
et
w
o
r
k
s
tr
a
f
f
ic
d
e
m
an
d
ca
n
n
o
t
b
e
k
n
o
w
n
i
n
ad
v
an
c
e.
A
ll
t
h
e
ap
p
r
o
ac
h
es
w
e
d
is
c
u
s
s
ed
ea
r
lier
is
r
el
y
o
n
d
e
m
an
d
o
f
tr
af
f
ic
b
u
t
i
t
is
n
o
t
s
u
f
f
icie
n
t
to
g
e
t
in
f
o
r
m
a
t
io
n
ab
o
u
t
n
et
w
o
r
k
in
b
e
f
o
r
e.
I
n
th
i
s
r
esear
ch
w
e
d
o
n
’
t u
s
e
tr
af
f
ic
d
e
m
an
d
i
n
s
te
ad
o
f
th
at
w
e
u
s
e
d
if
f
er
en
t p
ar
a
m
eter
s
to
k
n
o
w
th
e
n
et
w
o
r
k
i
n
ad
v
an
ce
3.
P
AT
H
B
ASE
D
L
O
AD
B
AL
ANCIN
G
A
L
G
O
R
I
T
H
M
T
o
im
p
r
o
v
e
th
e
cl
u
s
ter
p
er
f
o
r
m
an
ce
,
th
e
lo
ad
b
alan
ci
n
g
al
g
o
r
ith
m
w
h
ic
h
w
i
ll
r
u
n
o
n
lo
a
d
b
alan
ce
r
p
lay
s
a
s
i
g
n
if
ica
n
t
r
o
le.
Dis
t
r
ib
u
ted
S
y
s
te
m
i
s
i
m
p
o
r
tan
t
to
d
is
tr
ib
u
ti
n
g
t
h
e
w
o
r
k
lo
ad
o
n
t
h
e
s
er
v
er
s
[
17
].
L
o
ad
B
alan
ce
r
p
lace
s
a
v
ital
r
o
le
in
f
u
lf
illi
n
g
t
h
e
r
eq
u
est
o
f
th
e
clien
ts
t
h
r
o
u
g
h
s
er
v
er
s
an
d
f
o
r
th
is
w
o
r
k
lo
ad
b
alan
ce
r
r
o
u
tes
r
eq
u
est
s
to
t
h
o
s
e
s
er
v
er
s
,
w
h
ich
h
a
s
t
h
e
ca
p
ab
i
lit
y
o
f
d
o
in
g
it
s
j
o
b
in
a
n
e
f
f
ec
tiv
e
w
a
y
t
h
at
i
s
m
ax
i
m
izatio
n
o
f
s
p
ee
d
,
m
a
x
i
m
u
m
u
t
ilizatio
n
o
f
ca
p
ac
i
t
y
a
n
d
ca
n
f
u
l
f
ill
th
e
clie
n
t’
s
r
eq
u
est
s
[
1
8
]
.
Ou
r
s
o
lu
tio
n
f
o
llo
w
s
th
e
la
y
e
r
ar
ch
itectu
r
e
o
f
SD
N,
in
w
h
i
ch
th
er
e
ar
e
t
w
o
p
la
n
es,
co
n
t
r
o
l
p
lan
e
an
d
d
ata
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
P
a
th
b
a
s
ed
l
o
a
d
b
a
la
n
cin
g
f
o
r
d
a
ta
ce
n
ter n
etw
o
r
ks u
s
in
g
S
DN
(
V
.
Dee
b
a
n
C
h
a
kra
va
r
th
y
)
3281
p
lan
e,
Fig
u
r
e
1
is
s
h
o
w
i
n
g
t
h
e
ar
ch
itect
u
r
e
o
f
SDN.
T
h
e
p
ath
s
w
h
ic
h
tr
a
n
s
f
er
w
o
r
k
lo
ad
f
r
o
m
s
o
u
r
ce
to
d
esti
n
atio
n
s
h
o
u
ld
b
e
o
p
ti
m
al,
th
is
i
s
d
o
n
e
b
y
ON
OS
co
n
tr
o
ller
.
T
h
e
co
n
tr
o
ller
m
a
k
es
d
ec
is
io
n
o
n
t
h
e
b
asi
s
o
n
p
ath
ev
al
u
ato
r
w
h
ich
s
elec
ts
th
e
p
at
h
w
h
ich
a
r
e
o
p
ti
m
al
f
o
r
p
ath
lo
ad
b
alan
cin
g
.
I
t
co
n
tain
s
d
ata
co
llecto
r
m
et
h
o
d
an
d
p
ath
ev
al
u
ato
r
m
e
th
o
d
.
F
i
g
u
r
e 1
.
SDN
L
o
ad
b
alan
cin
g
ar
ch
itectu
r
e
3
.
1
.
Da
t
a
c
o
llect
o
r
m
et
ho
d
T
o
p
o
lo
g
y
li
n
k
co
llecto
r
T
o
p
o
lo
g
y
li
n
k
co
llecto
r
g
ets
th
e
in
f
o
r
m
atio
n
b
et
w
ee
n
th
e
lin
k
s
o
f
th
e
to
p
o
lo
g
y
,
it
i
s
d
o
n
e
b
y
tr
an
s
m
itti
n
g
L
L
DP
d
ata
p
ac
k
ets to
ev
er
y
n
o
d
e.
T
r
af
f
ic
d
ata
co
llecto
r
Sin
ce
w
e
ar
e
u
s
i
n
g
d
if
f
er
e
n
t
p
ar
am
eter
s
f
o
r
m
a
k
i
n
g
d
ec
is
i
o
n
s
f
o
r
p
ath
lo
ad
b
ala
n
cin
g
,
w
e
n
ee
d
to
co
l
lect
th
e
d
ata
ab
o
u
t
th
o
s
e
p
ar
a
m
eter
s
.
W
e
ar
e
u
s
in
g
b
y
te
co
u
n
t,
p
ac
k
et
co
u
n
t
f
o
r
w
ar
d
in
g
r
ate
an
d
d
u
r
atio
n
o
f
tr
an
s
m
it
tin
g
b
et
w
ee
n
s
o
u
r
c
e
n
o
d
e
an
d
d
esti
n
atio
n
n
o
d
e.
P
ath
ev
alu
ato
r
m
eth
o
d
T
h
is
m
et
h
o
d
co
n
tain
s
T
o
p
-
K
p
ath
alg
o
r
it
h
m
an
d
ev
al
u
ato
r
m
eth
o
d
.
K
P
ath
alg
o
r
ith
m
T
h
er
e
ar
e
m
a
n
y
p
at
h
s
i
n
th
e
n
et
w
o
r
k
f
o
r
tr
an
s
m
itti
n
g
p
ac
k
ets
f
r
o
m
s
o
u
r
ce
to
d
esti
n
a
tio
n
.
T
h
o
s
e
p
ath
s
ar
e
u
s
ed
to
p
u
t
i
n
t
h
e
A
l
g
o
r
ith
m
2
,
t
h
at
h
as
b
ee
n
p
r
o
v
id
ed
b
elo
w
.
T
h
is
i
s
d
o
n
e
in
t
w
o
d
if
f
er
en
t
v
ar
ian
t
s
o
n
e
is
s
tar
t
an
d
a
n
o
th
er
is
in
te
r
v
al.
I
n
th
e
s
tar
t
v
ar
ia
n
t,
w
e
u
s
e
t
h
e
Fl
y
o
d
’
s
al
g
o
r
it
h
m
[
1
9
]
to
f
in
d
t
h
e
to
p
-
k
p
at
h
s
.
I
n
t
h
e
in
ter
v
al
v
a
r
ian
t,
w
e
u
s
e
t
h
e
d
ata
co
llec
te
d
b
y
tr
af
f
ic
d
ata
co
llecto
r
to
g
et
o
p
ti
m
al
p
ath
f
o
r
tr
an
s
m
itti
n
g
th
e
d
ata.
A
l
g
o
r
ith
m
1
s
h
o
w
s
th
e
k
p
ath
al
g
o
r
ith
m
.
Alg
o
rit
hm
1
:
K
pa
t
h
co
llect
o
r
Star
t
m
o
d
e:
Getti
n
g
K
p
at
h
s
W
e
ass
u
m
e
th
at
E
b
e
t
h
e
s
o
u
r
ce
an
d
G
b
e
t
h
e
d
esti
n
atio
n
n
o
d
e,
in
b
et
w
ee
n
t
h
ese
n
o
d
es
w
e
h
a
v
e
to
f
i
n
d
k
p
ath
s
.
M[
r
]
is
th
e
ar
r
ay
i
n
w
h
ic
h
th
e
p
ath
s
w
ill b
e
s
to
r
ed
an
d
th
e
f
ir
s
t p
ath
is
M[
0
]
1
.
M
[
0
]
r
ep
r
esen
ts
p
ath
in
n
o
d
es
E
to
G
2
.
Fo
r
n
in
{1
,
2
,
…,
N
-
1
}
3
.
Fo
r
a
in
{0
,
1
,
2
,
…,
len
g
th
(
M[
n
-
1
]
)
-
2
}
4
.
n
o
d
eP
= P
[
n
-
1
]
.
n
o
d
e(
a
)
5
.
Fo
r
b
in
{0
,
1
,
2
,
…,
n
-
1}
6
.
n
o
d
eQ
= P
[
b
]
.
n
o
d
e(
i
)
7
.
I
f
n
o
d
eP
=
n
o
d
eQ
8
.
T
h
e
d
is
tan
ce
b
et
w
ee
n
n
o
d
eP
to
M
[
b
]
.
n
o
d
e(
a+
1
)
=
in
f
in
it
e
9
.
E
n
d
I
F
1
0
.
E
n
d
Fo
r
1
1
.
E
[a
]
r
ep
r
esen
ts
t
h
e
p
ath
in
n
o
d
es
n
o
d
eP
an
d
G
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t
201
9
:
3
2
7
9
-
3
2
8
5
3282
1
2
.
R
[
a]
w
ill b
e
f
o
u
n
d
in
M
[
k
-
1
]
f
r
o
m
E
to
n
o
d
eP
13.
Q[a
]
= R
[
a
]
+ E
[
a
]
1
4
.
E
n
d
Fo
r
1
5
.
T
h
e
m
i
n
i
m
u
m
p
at
h
i
n
Q[
a
]
is
M[
r
]
1
6
.
A
g
ai
n
s
to
r
e
th
e
i
n
itial
v
al
u
e
o
f
d
is
tan
ce
b
et
w
ee
n
n
o
d
eP
to
n
o
d
eQ
1
7
.
E
n
d
Fo
r
3
.
2
.
I
nte
rv
a
l m
o
de
T
ak
e
k
p
ath
s
f
r
o
m
th
e
ab
o
v
e
alg
o
r
ith
m
an
d
u
s
e
th
e
m
i
n
alg
o
r
ith
m
2
an
d
af
ter
ev
er
y
5
m
i
n
u
te
s
tak
e
an
o
th
er
n
e
w
k
p
ath
s
.
E
v
alu
a
to
r
A
l
g
o
r
ith
m
T
h
e
ev
alu
ato
r
alg
o
r
ith
m
i
s
u
s
ed
to
ev
alu
ate
th
e
k
p
ath
w
h
ic
h
h
as
b
ee
n
f
o
u
n
d
b
y
A
l
g
o
r
ith
m
1
.
T
h
e
p
ar
am
eter
s
o
f
ev
a
lu
at
io
n
ar
e
b
y
te
r
ate,
p
ac
k
e
t c
o
u
n
t,
p
o
r
t f
o
r
w
ar
d
r
ate
an
d
d
u
r
at
io
n
b
e
t
w
ee
n
tr
a
n
s
m
is
s
io
n
o
f
p
ac
k
et
f
r
o
m
o
n
e
h
o
s
t
ter
m
i
n
al
to
an
o
th
er
.
A
l
g
o
r
ith
m
2
s
h
o
w
s
t
h
e
al
g
o
r
ith
m
f
o
r
ev
al
u
ati
o
n
o
f
to
p
-
k
p
at
h
.
Alg
o
rit
hm
2
:
E
v
a
lua
t
io
n
o
f
P
a
t
hs
I
nput:
T
he
o
utput
pa
t
hs
f
o
un
d
in
a
lg
o
rit
hm
1
is
K
O
utput
:
t
he
bes
t
o
pti
m
ized p
a
t
h
is
O
P
T
I
M
AL
_
P
AT
H
1
.
Def
i
n
e
f
u
n
ctio
n
f
ac
to
r
F
(
b
y
te_
co
u
n
t,
p
ac
k
et_
co
u
n
t,
Fl
o
w
_
r
ate,
d
u
r
atio
n
)
a
n
d
d
ef
i
n
e
th
e
r
an
k
f
u
n
c
tio
n
R
[
N]
f
o
r
p
ath
.
2
.
Fo
r
i in
{0
,
1
,
2
,
.
.
.
,
K
-
1
}
3
.
b
yte
_
co
u
n
t[
i
]
=
B
yte
co
u
n
t in
K
[
i
]
4
.
p
a
ck
et_
co
u
n
t[
i
]
= P
a
ck
et
co
u
n
t in
K
[
i
]
5
.
F
lo
w
_
r
a
te[
i
]
= F
lo
w
r
a
te
in
K
[
i
]
6
.
Du
r
a
tio
n
[
i
]
= D
u
r
a
tio
n
in
K
[
i
]
7
.
E
n
d
Fo
r
8
.
Fo
r
ea
ch
j
in
{0
,
1
,
2
,
.
.
.
,
K
-
1
}
9
.
b
yte
_
co
u
n
t[
j
]
=
1
.
0
/ l
o
g
(
b
yte_
co
u
n
t
[
j
]
+ 0
.
1
)
1
0
.
p
a
ck
et_
co
u
n
t[
j
]
= 1
.
0
/ lo
g
(
p
a
ck
et_
co
u
n
t
[
j
]
+
0
.
1
)
11.
F
lo
w
_
r
a
te[
j
]
= 1
.
0
/ (
1
+ e
xp
(
F
lo
w
_
r
a
te[
j
]
/
1
0
0
.
0
)
)
1
2
.
Du
r
a
tio
n
[
j
]
=
1
/Du
r
a
tio
n
[
j
]
1
3
.
E
n
d
Fo
r
14.
Def
in
e
w
ei
g
h
t
m
atr
i
x
W
(
b
y
te_
co
u
n
t_
w
ei
g
h
t
,
p
ac
k
et_
co
u
n
t_
w
ei
g
h
t,
F
lo
w
_
r
ate_
w
ei
g
h
t,
d
u
r
atio
n
_
w
eig
h
t)
1
5
.
Fo
r
k
in
{0
,
1
,
2
,
.
.
.
,
K
-
1
}
1
6
.
t_
b
yte_
co
u
n
t[
k
]
=
b
yte_
co
u
n
t_
w
eig
h
t +
b
yte_
c
o
u
n
t
[
k
]
17.
t_
p
a
ck
et_
co
u
n
t[
k
]
=
p
a
ck
et_
co
u
n
t_
w
eig
h
t+p
a
c
ke
t_
co
u
n
t
[
k
]
1
8
.
t_
F
lo
w
_
r
a
te[k
]
= F
lo
w
_
r
a
te_
w
eig
h
t a
n
d
F
lo
w
_
r
a
te[
k
]
1
9
.
t_
Du
r
a
tio
n
[
k
]
= d
u
r
a
tio
n
_
w
eig
h
t +
Du
r
a
tio
n
[
k
]
20.
va
l
u
e_
s
co
r
e[
k
]
=
t_
Du
r
a
tio
n
[
k
]
+ t_
b
ytes[
k
]
+ t_
p
a
ck
et
_
co
u
n
t
[
k
]
+
t_
F
lo
w
_
r
a
te[
k
]
2
1
.
E
n
d
Fo
r
2
2
.
W
h
ich
ev
er
p
ath
g
e
ts
t
h
e
m
ax
i
m
u
m
v
alu
e_
s
co
r
e
is
th
e
OP
T
I
MA
L
_
P
A
T
H
4.
SI
M
UL
AT
I
O
N
A
ND
ANA
L
YSI
S
W
e
h
av
e
u
s
ed
Mi
n
in
e
t
[
2
0
,
21,
2
2
]
an
d
th
e
ONOS
p
lat
f
o
r
m
[
2
3
,
2
4
]
to
s
im
u
late
th
e
n
et
w
o
r
k
to
p
o
lo
g
y
m
en
tio
n
ed
in
t
h
e
p
r
o
p
o
s
ed
tech
n
iq
u
e.
ONOS
is
a
p
latf
o
r
m
,
w
h
ic
h
ca
p
italizes
o
n
th
e
r
eso
u
r
ce
s
o
f
J
A
V
A
f
o
r
th
e
r
ap
id
co
n
f
i
g
u
r
atio
n
an
d
p
r
o
to
t
y
p
in
g
o
f
n
et
w
o
r
k
co
n
tr
o
ller
s
.
Min
i
n
et
a
s
s
is
ts
u
s
er
s
i
n
q
u
ic
k
f
o
r
m
atio
n
o
f
v
ir
t
u
al
n
et
w
o
r
k
s
,
C
o
n
n
ec
t
co
n
n
ec
to
r
s
,
s
w
itc
h
es
,
an
d
lo
ca
l
n
e
t
w
o
r
k
s
to
a
s
i
n
g
l
e
s
er
v
er
.
P
r
o
v
id
in
g
a
lig
h
t
w
ei
g
h
t
e
v
al
u
atio
n
p
lat
f
o
r
m
f
o
r
co
n
f
i
g
u
r
in
g
Op
en
F
lo
w
ap
p
licatio
n
s
,
m
i
n
i
n
et
is
r
u
n
n
in
g
o
n
Vir
t
u
al
B
o
x
in
w
h
ich
t
h
e
h
o
s
t
is
li
n
u
x
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
Mi
n
i
n
et
is
d
ep
en
d
ed
o
n
th
e
ca
p
ab
ilit
y
an
d
c
o
n
f
i
g
u
r
atio
n
o
f
t
h
e
v
er
s
io
n
o
f
lin
u
x
.
T
h
e
s
p
ec
if
ica
tio
n
o
f
t
h
e
h
o
s
t i
s
lis
ted
d
o
w
n
in
T
ab
le
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
P
a
th
b
a
s
ed
l
o
a
d
b
a
la
n
cin
g
f
o
r
d
a
ta
ce
n
ter n
etw
o
r
ks u
s
in
g
S
DN
(
V
.
Dee
b
a
n
C
h
a
kra
va
r
th
y
)
3283
T
ab
le
1
.
Sp
eci
f
icatio
n
o
f
h
o
s
t
P
a
r
a
me
t
e
t
s
V
a
l
u
e
s
P
h
y
si
c
a
l
M
a
c
h
i
n
e
I
n
t
e
l
C
o
r
e
i
7
-
2
1
0
0
3
.
1
G
H
z
p
r
o
c
e
sso
r
8
G
B
me
mo
r
y
O
p
e
r
a
t
i
n
g
S
y
st
e
m
x
8
6
_
6
4
-
w
i
t
h
-
U
b
u
n
u
1
4
.
1
0
V
i
r
t
u
a
l
M
a
c
h
i
n
e
V
i
r
t
u
a
l
B
o
x
I
n
o
r
d
er
to
cr
ea
te
u
s
ef
u
l
r
es
u
lt
s
,
w
e
h
a
v
e
tr
ied
to
test
th
e
tec
h
n
iq
u
e
an
d
t
h
e
n
et
w
o
r
k
f
o
r
a
p
r
o
v
en
an
d
r
o
b
u
s
t
o
v
er
all
ef
f
ic
ien
c
y
.
Fo
r
n
et
w
o
r
k
v
er
i
f
icatio
n
ca
p
ab
ilit
y
an
d
ab
ilit
y
,
t
h
e
s
i
m
u
lat
io
n
en
v
ir
o
n
m
en
t
h
a
s
b
ee
n
d
esig
n
ed
w
it
h
6
Op
en
Fl
o
w
s
w
itc
h
es
to
p
o
lo
g
y
a
n
d
8
h
o
s
t
u
s
er
s
.
As
s
h
o
w
n
in
t
h
e
Fig
u
r
e
2
,
th
e
s
w
itc
h
in
d
icate
s
a
n
Op
en
F
lo
w
s
w
itc
h
;
it's
s
w
i
tch
I
D.
T
h
e
h
o
s
t
i
s
a
co
n
n
ec
ted
h
o
s
t
s
i
m
u
lta
n
eo
u
s
l
y
an
d
its
h
o
s
t
I
D
i
s
.
I
n
th
e
m
ea
n
ti
m
e,
th
e
s
o
lu
t
i
o
n
p
r
o
p
o
s
ed
to
b
alan
ce
th
e
lo
ad
is
also
an
al
y
ze
d
to
s
o
lv
e
th
e
w
a
y
s
o
f
u
n
d
er
s
ta
n
d
in
g
t
h
e
d
if
f
er
en
ce
s
.
Fig
u
r
e
2
.
T
o
p
o
lo
g
y
o
f
s
i
m
u
lat
io
n
4
.
1
.
E
f
f
ec
t
i
v
eness
T
o
d
ed
u
ce
th
e
v
alu
e
o
f
th
e
e
v
alu
atio
n
m
o
d
u
le
t
h
e
ef
f
ic
ien
c
y
test
h
as
b
ee
n
ca
r
r
ied
o
u
t
u
n
d
e
r
d
y
n
a
m
ic
ch
an
g
es.
T
h
e
Ho
s
t1
an
d
Ho
s
t
8
ar
e
s
elec
ted
to
b
e
th
e
en
tit
y
u
n
d
er
o
b
s
er
v
atio
n
.
Firstl
y
,
we
iter
ativ
el
y
f
ig
u
r
e
o
u
t
th
e
s
elec
tio
n
o
f
th
r
ee
iter
at
iv
e
p
ath
s
.
Scr
ee
n
in
g
th
e
d
y
n
a
m
ic
s
tat
u
s
o
f
t
h
e
n
et
w
o
r
k
,
b
ased
o
n
r
eq
u
ir
em
e
n
ts
,
th
e
th
r
ee
p
ath
s
ar
e
th
e
n
ev
al
u
ated
u
s
i
n
g
t
h
e
f
u
zz
y
e
v
al
u
atio
n
m
o
d
el.
I
n
itial
l
y
,
t
h
e
v
al
u
e
t
o
ev
alu
ate
a
p
ath
is
f
o
u
n
d
.
T
o
k
n
o
w
t
h
e
w
o
r
k
lo
ad
in
S
w
i
tch
6
a
n
d
S
w
itc
h
4
a
p
ac
k
et
w
ill
b
e
s
en
t
b
y
H
os
t7
.
T
o
g
et
t
h
e
r
eso
u
r
ce
s
o
f
b
an
d
w
id
t
h
in
b
et
w
ee
n
S
w
i
tch
1
an
d
S
w
itc
h
4
,
H
o
s
t6
s
en
d
s
a
p
ac
k
et
to
H
o
s
t2
.
W
h
en
th
e
li
n
k
s
tat
u
s
ch
a
n
g
es
th
e
ev
alu
a
tio
n
v
al
u
e
o
f
t
h
r
ee
p
at
h
s
ch
a
n
g
es
to
o
.
Fo
r
f
ir
s
t
p
at
h
,
i
ts
o
n
e
li
n
k
an
d
2
s
w
i
tch
e
s
ar
e
i
m
p
ac
ted
a
n
d
t
h
e
y
h
av
e
n
o
o
b
s
tacle
o
n
th
e,
s
o
its
v
alu
e
o
f
e
v
al
u
atio
n
d
ec
r
ea
s
ed
m
ax
i
m
u
m
.
Fo
r
th
ir
d
p
ath
,
S
witch
4
is
i
m
p
licated
an
d
ca
u
s
ed
o
b
s
tr
u
ctio
n
,
s
o
t
h
e
v
al
u
e
o
f
e
v
al
u
atio
n
d
ec
r
ea
s
ed
b
u
t
a
le
s
s
t
h
a
n
t
h
e
o
th
er
s
.
T
h
e
r
esu
l
t
s
h
av
e
b
ee
n
s
h
o
w
n
i
n
F
ig
u
r
e
3
.
L
ater
o
n
,
w
e
w
ill
g
et
th
e
o
p
ti
m
al
p
ath
to
tr
an
s
m
it
t
h
e
lo
ad
f
o
r
ac
q
u
ir
in
g
t
h
e
lo
ad
b
alan
c
in
g
in
t
h
e
n
et
w
o
r
k
.
De
m
o
n
s
tr
ati
n
g
e
f
f
ec
ti
v
e
n
ess
,
t
h
e
p
r
ese
n
t
ed
tech
n
iq
u
e
also
p
r
o
m
i
s
es
t
o
d
y
n
a
m
icall
y
alter
p
ath
s
w
it
h
t
h
e
ad
eq
u
ate
tr
af
f
ic
s
tatu
s
th
e
li
n
k
s
a
n
d
n
o
d
es.
Fig
u
r
e
3
.
T
h
e
ev
alu
ati
o
n
o
f
p
ath
s
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0
5
10
15
20
25
30
35
P
a
th
1[2,1,
6]
P
a
th
2[2,4,
6]
P
a
th
3[2
,1,
4,6]
T
i
me
(
s
)
E
v
alu
a
ti
o
n
F
i
g
3:T
h
e
e
v
alu
a
ti
o
n
o
f
p
a
th
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t
201
9
:
3
2
7
9
-
3
2
8
5
3284
4
.
2
.
Ana
ly
s
is
o
f
ef
f
iciency
T
o
u
n
d
er
s
tan
d
t
h
e
p
r
o
p
o
s
ed
ef
f
icie
n
c
y
le
v
el,
t
h
e
k
-
p
at
h
s
alg
o
r
ith
m
is
co
n
s
id
er
ed
an
d
f
u
r
th
er
an
al
y
s
i
s
is
d
o
n
e.
Ho
s
t2
an
d
Ho
s
t9
ar
e
s
elec
ted
as
th
e
s
u
b
j
ec
t
o
f
o
b
s
er
v
atio
n
i
n
t
h
e
9
o
p
en
Flo
w
o
p
en
n
et
w
o
r
k
to
p
o
lo
g
y
an
d
1
0
h
o
s
t
ter
m
i
n
al
s
,
th
at
i
s
r
ep
r
esen
ted
in
Fi
g
u
r
e
4
.
A
s
e
x
p
er
i
m
e
n
tal
r
es
u
lt
t
h
e
R
T
T
h
as
b
ee
n
test
ed
,
t
h
is
h
as
s
h
o
w
n
i
n
Fi
g
u
r
e
5
.
I
t
s
h
o
w
s
th
a
t
R
T
T
f
lu
ctu
a
tes
v
io
len
tl
y
,
w
h
ile
t
h
e
p
ath
b
ec
o
m
e
s
m
o
r
e
f
illed
w
i
th
th
e
p
ath
alg
o
r
ith
m
.
Ho
w
e
v
er
,
R
T
T
d
o
es
n
o
t
f
lu
ctu
ate
w
it
h
o
u
r
s
o
lu
tio
n
,
b
ec
au
s
e
th
e
tr
af
f
ic
is
tr
an
s
m
i
tted
to
th
e
o
p
tim
a
l
r
o
u
te
i
n
t
i
m
e.
T
h
e
r
es
u
lts
en
s
u
r
e
t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
o
f
lo
ad
-
b
alan
ci
n
g
to
b
e
o
p
ti
m
ized
en
o
u
g
h
t
o
g
u
ar
a
n
tee
s
a
n
it
y
in
p
ac
k
et
tr
an
s
m
i
s
s
io
n
an
d
also
d
y
n
a
m
ic
m
a
n
ag
e
m
e
n
t o
f
t
h
e
p
ath
s
.
Fig
u
r
e
4
.
T
o
p
o
lo
g
y
f
o
r
c
o
m
p
a
r
is
o
n
F
i
g
u
r
e 5
.
R
T
T
o
f
th
e p
r
o
p
o
s
ed
s
o
lu
tio
n
s
5.
CO
NCLU
SI
O
N
W
ith
a
n
ai
m
to
i
n
n
o
v
ati
v
el
y
p
r
o
p
o
s
e
an
SDN
-
b
ased
p
ath
lo
ad
-
b
ala
n
cin
g
tec
h
n
iq
u
e,
th
is
p
ap
er
p
r
esen
ts
h
o
w
it
i
m
p
r
o
v
es
t
h
e
cu
r
r
en
t
r
ates
o
f
e
f
f
icie
n
c
y
,
r
eliab
ilit
y
an
d
u
t
ilizatio
n
.
T
h
e
f
u
zz
y
s
y
n
t
h
etic
ev
alu
a
tio
n
m
o
d
el
h
elp
s
i
n
t
h
e
d
y
n
a
m
ic
an
d
o
p
ti
m
ized
s
el
ec
t
io
n
o
f
a
p
ath
,
as
a
r
e
f
lex
to
th
e
v
ar
iatio
n
in
th
e
n
et
w
o
r
k
tr
af
f
ic.
T
h
e
i
m
p
l
e
m
en
tatio
n
o
f
t
h
e
s
o
lu
t
io
n
i
s
s
u
p
p
o
r
ted
u
s
i
n
g
t
h
e
ONO
S
p
l
atf
o
r
m
,
w
h
er
ea
s
th
e
v
er
if
ica
tio
n
a
n
d
p
r
o
o
f
o
f
t
h
e
r
esu
lt
s
o
b
tain
ed
i
s
b
ac
k
ed
b
y
Min
i
n
et.
T
h
e
g
r
ap
h
ical
r
ep
r
es
en
tatio
n
s
co
n
c
lu
d
e
th
at
t
h
i
s
tec
h
n
iq
u
e
ca
n
d
y
n
a
m
icall
y
n
o
t
o
n
l
y
a
v
o
id
f
a
ilu
r
es
an
d
f
au
lts
in
th
e
p
at
h
b
u
t
al
s
o
ad
j
u
s
t
t
h
e
tr
an
s
m
itti
n
g
p
ath
.
T
h
e
r
es
u
lt
s
ee
m
s
m
o
r
e
ac
cu
r
ate
w
i
th
t
h
e
o
v
e
r
all
p
er
f
o
r
m
an
ce
o
f
th
e
n
et
w
o
r
k
b
ein
g
i
n
cr
ea
s
ed
.
0
1
2
3
4
5
6
7
8
9
0
10
20
30
40
50
60
70
80
E
v
alu
a
ti
o
n
Mo
d
e
l
S
h
o
rt
e
s
t
P
a
th
F
i
g
5:
R
T
T
o
f
th
e
p
r
o
p
o
s
e
d
S
o
l
u
ti
o
n
s
T
i
me
(s)
L
a
t
e
n
c
y(
s
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8708
P
a
th
b
a
s
ed
l
o
a
d
b
a
la
n
cin
g
f
o
r
d
a
ta
ce
n
ter n
etw
o
r
ks u
s
in
g
S
DN
(
V
.
Dee
b
a
n
C
h
a
kra
va
r
th
y
)
3285
RE
F
E
R
E
NC
E
S
[1
]
T
.
S
a
sid
h
a
r,
V
.
Ha
v
ish
a
,
S
.
Ko
u
sh
ik
,
M
.
De
e
p
,
V
K.
Re
d
d
y
,
“
Lo
a
d
Ba
lan
c
i
n
g
T
e
c
h
n
iq
u
e
s
f
o
r
E
ff
icie
n
t
T
ra
ff
ic
M
a
n
a
g
e
m
e
n
t
in
Clo
u
d
E
n
v
iro
n
m
e
n
t,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
6
,
n
o
.
3
,
p
p
.
9
6
3
-
9
7
3
,
2
0
1
6
.
[2
]
“
S
o
f
twa
re
-
De
f
in
e
d
Ne
t
w
o
rk
in
g
:
th
e
n
e
w
n
o
rm
f
o
r
n
e
tw
o
rk
s,”
W
h
it
e
P
a
p
e
r,
Op
e
n
Ne
tw
o
rk
in
g
F
o
u
n
d
a
ti
o
n
,
2
0
1
2
.
[3
]
Ha
lep
li
d
is,
E
.
,
De
n
a
z
is,
S
.
,
K
o
u
f
o
p
a
v
lo
u
,
O
.
,
Ha
lp
e
r
n
,
J.
a
n
d
S
a
li
m
,
J.H.,
“
S
o
f
tw
a
re
-
d
e
f
in
e
d
n
e
tw
o
rk
in
g
:
Ex
p
e
rime
n
ti
n
g
w
it
h
th
e
c
o
n
tro
l
to
f
o
r
w
a
rd
in
g
p
lan
e
in
terf
a
c
e
,
”
In
S
o
f
twa
re
De
fi
n
e
d
Ne
two
rk
in
g
(
EW
S
DN),
2
0
1
2
Eu
ro
p
e
a
n
W
o
rk
sh
o
p
o
n
(p
p
.
9
1
-
9
6
).
IEE
E,
Oc
to
b
e
r,
2
0
1
2
.
[4
]
M
c
Ke
o
w
n
,
N.,
A
n
d
e
rso
n
,
T
.
,
Ba
lak
rish
n
a
n
,
H.,
P
a
ru
lk
a
r,
G
.
,
P
e
ter
so
n
,
L
.
,
Re
x
f
o
rd
,
J.
,
S
h
e
n
k
e
r,
S
.
a
n
d
T
u
rn
e
r,
J.,
“
Op
e
n
F
lo
w
:
e
n
a
b
li
n
g
in
n
o
v
a
ti
o
n
in
c
a
m
p
u
s
n
e
tw
o
rk
s
”
.
ACM
S
IGC
OM
M
Co
mp
u
ter
Co
mm
u
n
ica
ti
o
n
Rev
iew
,
3
8
(2
)
,
p
p
.
6
9
-
7
4
.
2
0
0
8
.
[5
]
P
O
X
,
h
tt
p
s://
o
p
e
n
f
lo
w
.
sta
n
f
o
rd
.
e
d
u
/d
is
p
lay
/ON
L
/P
OX
+
W
ik
i
[6
]
G
u
d
e
,
N.,
Ko
p
o
n
e
n
,
T
.
,
P
e
tt
it
,
J.,
P
f
a
ff
,
B.
,
Ca
sa
d
o
,
M
.
,
M
c
Ke
o
w
n
,
N.
a
n
d
S
h
e
n
k
e
r,
S
.
,
“
NO
X
:
t
o
w
a
rd
s
a
n
o
p
e
ra
ti
n
g
sy
ste
m
f
o
r
n
e
tw
o
rk
s
,
”
ACM
S
IGCO
M
M
Co
mp
u
ter
Co
m
mu
n
ica
t
io
n
Rev
iew
,
3
8
(3
),
p
p
.
1
0
5
-
1
1
0
.
2
0
0
8
.
[7
]
F
lo
o
d
li
g
h
t
,
h
tt
p
:
//
ww
w
.
p
ro
jec
tf
lo
o
d
li
g
h
t
.
o
rg
/f
lo
o
d
li
g
h
t/
[8
]
ONOS
,
h
tt
p
s://
w
ik
i.
o
n
o
s
p
ro
jec
t.
o
rg
/d
isp
lay
/ONO
S
/ONO
S
+
f
ro
m
+
S
c
ra
tch
[9
]
M
in
i
n
e
t,
h
tt
p
:
//
m
in
in
e
t.
o
rg
/
[1
0
]
Ko
e
rn
e
r,
M
.
a
n
d
Ka
o
,
O.
,
“
M
u
lt
i
p
le
se
rv
ice
lo
a
d
-
b
a
lan
c
in
g
w
it
h
Op
e
n
F
lo
w
,
”
In
Hi
g
h
Per
fo
rm
a
n
c
e
S
wit
c
h
i
n
g
a
n
d
Ro
u
ti
n
g
(
HPS
R),
2
0
1
2
IE
EE
1
3
t
h
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
,
(p
p
.
2
1
0
-
2
1
4
).
IEE
E
,
Ju
n
e
,
2
0
1
2
.
[1
1
]
W
a
n
g
,
R.
,
Bu
t
n
a
riu
,
D.
a
n
d
R
e
x
f
o
rd
,
J.,
Op
e
n
F
lo
w
-
Ba
se
d
S
e
rv
e
r
L
o
a
d
Ba
lan
c
i
n
g
G
o
n
e
W
il
d
.
Ho
t
-
ICE,
1
1
,
p
p
.
1
2
-
12
,
2
0
1
1
.
[1
2
]
Up
p
a
l
H,
Bra
n
d
o
n
D,
“
Op
e
n
F
lo
w
b
a
se
d
lo
a
d
b
a
lan
c
in
g
,
”
In
Pro
c
e
e
d
in
g
s
o
f
CS
E
5
6
1
:
n
e
tw
o
rk
in
g
.
p
ro
jec
t
re
p
o
rt
.
Un
ive
rs
it
y
o
f
W
a
sh
i
n
g
t
o
n
,
S
p
rin
g
,
2
0
1
0
.
[1
3
]
Ch
o
u
,
L
.
D.,
Ya
n
g
,
Y.T
.
,
Ho
n
g
,
Y.M
.
,
Hu
,
J.K.
a
n
d
Je
a
n
,
B.
,
“
A
g
e
n
e
tic
-
b
a
se
d
lo
a
d
b
a
lan
c
in
g
a
lg
o
rit
h
m
in
o
p
e
n
f
lo
w
n
e
tw
o
rk
,
”
In
A
d
v
a
n
c
e
d
T
e
c
h
n
o
lo
g
ies
,
Em
b
e
d
d
e
d
a
n
d
M
u
lt
im
e
d
ia
f
o
r
Hu
m
a
n
-
c
e
n
tri
c
Co
m
p
u
ti
n
g
,
S
p
rin
g
e
r,
D
o
rd
re
c
h
t,
(p
p
.
4
1
1
-
4
1
7
)
,
2
0
1
4
.
[1
4
]
S
.
Ja
in
,
e
t
a
l,
“
B4
:
e
x
p
e
rien
c
e
w
it
h
a
g
lo
b
a
ll
y
-
d
e
p
lo
y
e
d
so
f
twa
re
d
e
f
in
e
d
WA
N
,
”
S
I
G
COMM
,
2
0
1
3
.
[1
5
]
Ch
i
-
Ya
o
Ho
n
g
,
S
rik
a
n
th
Ka
n
d
u
la,Ratu
l
M
a
h
a
jan
,
M
i
n
g
Zh
a
n
g
,
Vijay
G
il
l
,
M
o
h
a
n
Na
n
d
u
ri,
R
o
g
e
r
W
a
tt
e
n
h
o
f
e
r,
“
Ac
h
iev
in
g
Hig
h
Util
iza
ti
o
n
w
it
h
S
o
f
tw
a
r
e
-
Driv
e
n
WAN
,
”
S
IG
CO
M
M
,
2
0
1
3
.
[1
6
]
Hu
i
L
o
n
g
,
Ya
o
S
h
e
n
,
M
in
y
i
Gu
o
,
F
e
il
o
n
g
T
a
n
g
,
“
LA
BERIO:
D
y
n
a
m
ic
lo
a
d
-
b
a
lan
c
e
d
r
o
u
ti
n
g
in
Op
e
n
F
lo
w
-
e
n
a
b
led
n
e
tw
o
rk
s
,”
In
Pr
o
c
e
e
d
in
g
o
f
t
h
e
2
7
th
in
ter
n
a
ti
o
n
a
l
c
o
n
fer
e
n
c
e
o
n
a
d
v
a
n
c
e
d
i
n
f
o
rm
a
ti
o
n
n
e
two
rk
in
g
a
n
d
a
p
p
li
c
a
ti
o
n
s
,
p
p
2
9
0
-
1
9
7
,
2
5
-
28
,
M
a
rc
h
2
0
1
3
.
[1
7
]
S
.
P
o
tl
u
ri
a
n
d
K.
S
u
b
b
a
Ra
o
,
“
Qu
a
li
ty
o
f
S
e
r
v
ice
b
a
se
d
T
a
s
k
S
c
h
e
d
u
l
in
g
A
l
g
o
rit
h
m
s
in
Clo
u
d
Co
m
p
u
ti
n
g
”
,
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
),
v
o
l.
7
,
n
o
.
2
,
p
.
1
0
8
8
,
2
0
1
7
.
[1
8
]
Ka
d
iy
a
la
Ra
m
a
n
a
a
n
d
M
.
P
o
n
n
a
v
a
ik
k
o
,
"
AW
S
Q:
A
n
A
p
p
ro
x
i
m
a
ted
W
e
b
S
e
rv
e
r
Qu
e
u
in
g
A
lg
o
rit
h
m
f
o
r
He
tero
g
e
n
e
o
u
s
W
e
b
S
e
rv
e
r
Clu
ste
r
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
),
V
o
l.
9
,
No
.
3
,
p
p
.
2
0
8
3
-
2
0
9
3
,
2
0
1
9
.
[1
9
]
F
lo
y
d
,
R.
W
.
,
“
A
lg
o
rit
h
m
9
7
:
sh
o
r
tes
t
p
a
th
.
C
o
m
m
u
n
ica
ti
o
n
s
o
f
th
e
A
CM
,
”
5
(6
),
p
.
3
4
5
,
1
9
6
2
.
[2
0
]
Ja
in
,
S
.
,
Ku
m
a
r,
A
.
,
M
a
n
d
a
l,
S
.
,
On
g
,
J.,
P
o
u
ti
e
v
sk
i,
L
.
,
S
i
n
g
h
,
A
.
,
V
e
n
k
a
ta,
S
.
,
W
a
n
d
e
re
r,
J.,
Zh
o
u
,
J.,
Zh
u
,
M
.
a
n
d
Zo
ll
a
,
J.,
A
u
g
u
st.
“
B4
:
Ex
p
e
ri
e
n
c
e
w
it
h
a
g
lo
b
a
ll
y
-
d
e
p
lo
y
e
d
so
f
t
w
a
re
d
e
f
in
e
d
WA
N
,
”
In
ACM
S
IGCO
M
M
Co
mp
u
ter
C
o
mm
u
n
ica
ti
o
n
Rev
ie
w
,
Vo
l.
4
3
,
No
.
4
,
p
p
.
3
-
14
,
A
CM
,
2
0
1
3
.
[2
1
]
Ho
n
g
,
C.
Y.,
Ka
n
d
u
la,
S
.
,
M
a
h
a
jan
,
R.
,
Zh
a
n
g
,
M
.
,
G
il
l,
V
.
,
Na
n
d
u
ri,
M
.
a
n
d
W
a
tt
e
n
h
o
f
e
r,
R.
,
A
u
g
u
st.
“
Ac
h
iev
in
g
h
ig
h
u
ti
li
z
a
ti
o
n
w
it
h
so
f
t
w
a
re
-
d
riv
e
n
WAN
,
”
In
ACM
S
IGCO
M
M
Co
mp
u
ter
Co
mm
u
n
ica
t
io
n
Rev
iew
,
V
o
l.
4
3
,
No
.
4
,
p
p
.
1
5
-
26
,
A
CM
,
2
0
1
3
.
[2
2
]
L
o
n
g
,
H.,
S
h
e
n
,
Y.,
G
u
o
,
M
.
a
n
d
T
a
n
g
,
F
.
,
M
a
rc
h
.
“
LA
BERIO:
D
y
n
a
m
ic
lo
a
d
-
b
a
lan
c
e
d
ro
u
t
in
g
in
Op
e
n
F
lo
w
-
e
n
a
b
led
n
e
tw
o
rk
s
,
”
In
Ad
v
a
n
c
e
d
In
fo
rm
a
t
io
n
Ne
two
rk
i
n
g
a
n
d
Ap
p
li
c
a
ti
o
n
s
(
AINA
),
2
0
1
3
IEE
E
2
7
t
h
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
,
IEE
E
,
(
p
p
.
2
9
0
-
2
9
7
)
.
2
0
1
3
.
[2
3
]
Be
ll
m
a
n
,
R.
E.
a
n
d
Zad
e
h
,
L
.
A
.
,
“
De
c
isio
n
-
m
a
k
in
g
in
a
f
u
z
z
y
e
n
v
iro
n
m
e
n
t,
”
M
a
n
a
g
e
me
n
t
sc
ien
c
e
,
1
7
(4
)
,
p
p
.
B
-
1
4
1
.
1
9
7
0
.
[2
4
]
S
h
ier,
D.R.
,
“
On
a
lg
o
rit
h
m
s f
o
r
f
in
d
i
n
g
th
e
k
sh
o
rtes
t
p
a
t
h
s in
a
n
e
t
w
o
rk
,
”
Ne
two
rk
s
,
9
(3
),
p
p
.
1
9
5
-
2
1
4
.
1
9
7
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.