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.
8
,
No
.
3
,
J
u
n
e
201
8
,
p
p
.
1741
~
1
7
4
6
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
8
i
3
.
p
p
1
7
4
1
-
1746
1741
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
A P
ro
lifi
c Sc
he
m
e
f
o
r Lo
a
d Ba
la
ncing
Relying
on Ta
sk
Co
m
ple
tion Ti
m
e
V.
Ana
nd
1
,
K.
Anura
dh
a
2
1
S
c
h
o
o
l
o
f
Co
m
p
u
ti
n
g
,
S
A
S
T
R
A
De
e
m
e
d
Un
iv
e
rsit
y
,
T
iru
m
a
laisa
m
u
d
ra
m
,
T
h
a
n
jav
u
r
,
In
d
ia
2
S
c
h
o
o
l
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
En
g
in
e
e
rin
g
,
S
A
S
T
R
A
De
e
m
e
d
Un
iv
e
rsity
,
T
iru
m
a
l
a
isa
m
u
d
ra
m
,
T
h
a
n
jav
u
r
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
28
,
2
0
1
7
R
ev
i
s
ed
Dec
2
8
,
2
0
1
7
A
cc
ep
ted
J
an
4
,
2
0
1
8
In
n
e
tw
o
rk
s
w
it
h
lo
t
o
f
c
o
m
p
u
tatio
n
,
l
o
a
d
b
a
lan
c
in
g
g
a
in
s
i
n
c
re
a
sin
g
sig
n
if
ica
n
c
e
.
T
o
o
ff
e
r
v
a
rio
u
s
re
so
u
rc
e
s,
se
rv
ice
s
a
n
d
a
p
p
li
c
a
ti
o
n
s,
th
e
u
lt
im
a
te
a
i
m
is
to
f
a
c
il
it
a
te
th
e
sh
a
rin
g
o
f
se
rv
ice
s
a
n
d
re
so
u
r
c
e
s
o
n
t
h
e
n
e
tw
o
rk
o
v
e
r
th
e
In
tern
e
t.
A
k
e
y
issu
e
to
b
e
f
o
c
u
se
d
a
n
d
a
d
d
re
ss
e
d
in
n
e
tw
o
rk
s
w
i
th
larg
e
a
m
o
u
n
t
o
f
c
o
m
p
u
tatio
n
is
lo
a
d
b
a
la
n
c
in
g
.
L
o
a
d
is
th
e
n
u
m
b
e
r
o
f
tas
k
s
„t‟
p
e
r
f
o
r
m
e
d
b
y
a
c
o
m
p
u
tatio
n
sy
ste
m
.
T
h
e
lo
a
d
c
a
n
b
e
c
a
teg
o
rize
d
a
s
n
e
t
w
o
rk
lo
a
d
a
n
d
CP
U
lo
a
d
.
F
o
r
a
n
e
ff
icie
n
t
lo
a
d
b
a
lan
c
in
g
stra
teg
y
,
th
e
p
ro
c
e
ss
o
f
a
s
sig
n
in
g
th
e
lo
a
d
b
e
tw
e
e
n
th
e
n
o
d
e
s
sh
o
u
l
d
e
n
h
a
n
c
e
th
e
re
so
u
rc
e
u
ti
li
z
a
ti
o
n
a
n
d
m
in
im
iz
e
th
e
c
o
m
p
u
tatio
n
ti
m
e
.
T
h
is
c
a
n
b
e
a
c
c
o
m
p
li
sh
e
d
b
y
a
u
n
if
o
rm
d
istri
b
u
ti
o
n
o
f
lo
a
d
o
f
to
a
ll
th
e
n
o
d
e
s.
A
L
o
a
d
b
a
lan
c
in
g
m
e
th
o
d
sh
o
u
ld
g
u
a
ra
n
tee
th
a
t,
e
a
c
h
n
o
d
e
in
a
n
e
tw
o
rk
p
e
rf
o
r
m
s
a
lm
o
st
e
q
u
a
l
a
m
o
u
n
t
o
f
w
o
rk
p
e
rti
n
e
n
t
t
o
t
h
e
ir
c
a
p
a
c
it
y
a
n
d
a
v
a
il
a
b
il
it
y
o
f
re
so
u
rc
e
s.
Re
l
y
in
g
o
n
tas
k
su
b
trac
ti
o
n
,
t
h
is
w
o
rk
h
a
s
p
re
se
n
ted
a
p
io
n
e
e
rin
g
a
lg
o
rit
h
m
ter
m
e
d
a
s
E
-
T
S
(E
ff
ic
ien
t
-
T
a
sk
S
u
b
trac
ti
o
n
).
T
h
is
a
lg
o
rit
h
m
h
a
s
se
lec
ted
a
p
p
ro
p
riate
n
o
d
e
s
f
o
r
e
a
c
h
tas
k
.
T
h
e
p
ro
p
o
se
d
a
lg
o
rit
h
m
h
a
s
im
p
ro
v
e
d
th
e
u
ti
li
z
a
ti
o
n
o
f
c
o
m
p
u
ti
n
g
re
so
u
rc
e
s
a
n
d
h
a
s
p
re
se
rv
e
d
th
e
n
e
u
tralit
y
in
a
ss
ig
n
in
g
th
e
l
o
a
d
to
th
e
n
o
d
e
s i
n
th
e
n
e
tw
o
rk
.
K
ey
w
o
r
d
:
C
o
m
p
u
tatio
n
ti
m
e
C
P
U
lo
ad
L
o
ad
b
alan
cin
g
Net
w
o
r
k
lo
ad
T
ask
s
u
b
tr
ac
tio
n
Co
p
y
rig
h
t
©
2
0
1
8
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.
An
an
d
,
Sch
o
o
l o
f
C
o
m
p
u
ti
n
g
,
S
A
ST
R
A
Un
i
v
er
s
i
t
y
,
T
ir
u
m
ala
is
a
m
u
d
r
a
m
,
T
h
an
j
av
u
r
6
1
3
4
0
1
,
T
am
il Na
d
u
,
I
n
d
ia.
E
-
Ma
il:
a
n
an
d
w
i
th
a
h
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
An
i
m
p
o
r
tan
t
i
s
s
u
e
in
n
e
t
wo
r
k
s
is
lo
ad
b
alan
cin
g
.
T
h
e
lo
ad
ca
n
b
e
b
alan
ce
d
d
y
n
a
m
icall
y
o
r
s
taticall
y
.
I
n
s
o
m
e
s
it
u
atio
n
s
,
th
e
lo
ad
o
n
a
p
ar
tic
u
lar
n
o
d
e
ca
n
n
o
t
b
e
f
o
r
eseen
[
1
]
.
T
h
is
r
e
m
ai
n
s
a
k
e
y
is
s
u
e
s
p
ec
if
icall
y
i
n
d
is
tr
ib
u
ted
n
e
t
w
o
r
k
s
.
Lo
a
d
B
a
la
n
ci
n
g
is
a
m
et
h
o
d
o
f
d
is
p
en
s
i
n
g
t
h
e
r
eq
u
est
s
to
m
a
n
y
s
er
v
er
s
o
r
r
eso
u
r
ce
s
.
T
h
is
h
elp
s
i
n
e
n
h
an
ci
n
g
t
h
e
p
er
f
o
r
m
a
n
ce
,
e
f
f
i
cien
t
u
tili
za
tio
n
o
f
r
eso
u
r
ce
s
,
av
o
id
s
o
v
er
lo
ad
in
g
,
r
ed
u
ce
d
r
esp
o
n
s
e
ti
m
e
a
n
d
i
m
p
r
o
v
ed
th
r
o
u
g
h
p
u
t.
I
n
a
d
is
tr
ib
u
ted
n
et
w
o
r
k
,
it
n
o
t
co
s
t
ef
f
ec
ti
v
e
to
h
a
v
e
id
le
s
er
v
er
s
w
h
e
n
t
h
er
e
ar
e
m
u
lt
ip
le
clien
t
r
eq
u
ests
.
T
h
e
p
r
esen
ce
o
f
in
a
ctiv
e
o
r
id
le
s
er
v
er
s
m
a
y
r
es
u
lt
in
in
e
f
f
icie
n
t
u
til
izatio
n
o
f
r
eso
u
r
ce
s
.
T
h
e
s
a
m
e
s
er
v
er
ca
n
n
o
t
b
e
o
v
er
lo
ad
ed
b
y
as
s
i
g
n
i
n
g
all
th
e
clie
n
t
r
eq
u
ests
to
th
e
s
a
m
e
s
er
v
er
.
An
e
f
f
icie
n
t
lo
ad
b
alan
cin
g
al
g
o
r
ith
m
s
h
o
u
ld
d
i
s
tr
ib
u
te
t
h
e
lo
ad
a
m
o
n
g
th
e
s
e
r
v
er
s
in
a
b
alan
ce
d
m
an
n
er
.
A
d
is
tr
ib
u
ted
n
et
w
o
r
k
is
a
co
m
b
i
n
atio
n
o
f
d
if
f
er
e
n
t
t
y
p
es
o
f
n
et
w
o
r
k
s
.
A
n
e
m
er
g
i
n
g
d
is
tr
ib
u
ted
co
m
p
u
ti
n
g
tec
h
n
o
lo
g
y
i
s
C
l
o
u
d
.
C
lo
u
d
co
m
p
u
ti
n
g
en
s
u
r
es
elasticit
y
an
d
s
ca
lab
ilit
y
f
o
r
o
p
er
atin
g
an
d
m
ai
n
tai
n
in
g
i
n
f
r
astru
c
tu
r
e,
p
la
tf
o
r
m
a
n
d
s
o
f
t
w
ar
e.
W
ith
clo
u
d
co
m
p
u
ti
n
g
s
er
v
ice
s
,
clie
n
t c
an
s
ca
le
u
p
o
r
s
ca
le
d
o
w
n
t
h
e
u
tili
za
t
io
n
o
f
r
eso
u
r
ce
s
r
ely
in
g
o
n
th
eir
d
e
m
a
n
d
.
I
n
clo
u
d
co
m
p
u
ti
n
g
,
r
eso
u
r
ce
s
ar
e
s
h
ar
ed
b
etw
ee
n
d
if
f
er
e
n
t
s
y
s
te
m
s
.
T
h
is
n
ee
d
s
a
m
eth
o
d
f
o
r
allo
ca
ti
n
g
t
h
e
task
s
to
s
y
s
te
m
s
.
T
h
e
tas
k
s
i
n
clu
d
e
h
a
n
d
li
n
g
o
f
r
eq
u
ests
,
m
a
k
e
t
h
e
d
ata
av
ail
ab
le
to
ev
er
y
s
y
s
te
m
a
n
d
p
r
o
ce
s
s
th
e
r
esp
o
n
s
e
s
en
t
b
y
th
e
d
if
f
er
e
n
t
p
r
o
ce
s
s
o
r
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
3
,
J
u
n
e
201
8
:
1
7
4
1
–
1746
1742
(
s
y
s
te
m
s
)
.
R
eso
u
r
ce
s
ca
n
b
e
m
a
n
ag
ed
e
f
f
icien
tl
y
w
it
h
a
n
ef
f
ec
t
iv
e
al
g
o
r
ith
m
f
o
r
lo
ad
b
alan
cin
g
.
An
in
e
f
f
ec
t
iv
e
lo
ad
b
alan
ci
n
g
al
g
o
r
ith
m
m
a
y
e
n
d
u
p
i
n
ex
ec
u
tio
n
d
ela
y
.
T
h
e
p
ap
er
is
o
r
g
an
ized
as f
o
ll
o
w
s
:
Sectio
n
2
o
f
f
er
s
an
in
s
i
g
h
t i
n
to
clo
u
d
co
m
p
u
ti
n
g
w
it
h
d
ep
lo
y
m
en
t
an
d
s
er
v
ice
m
o
d
els;
Sectio
n
3
p
r
esen
ts
t
h
e
tec
h
n
iq
u
es
an
d
a
lg
o
r
ith
m
s
f
o
r
lo
ad
b
alan
cin
g
w
it
h
lo
ad
b
alan
ce
r
s
,
s
ch
ed
u
lin
g
al
g
o
r
it
h
m
s
a
n
d
M
ap
R
ed
u
ce
;
Sectio
n
4
p
r
o
p
o
s
es
t
h
e
m
eth
o
d
E
-
T
S
w
i
th
th
e
a
lg
o
r
ith
m
;
Sectio
n
5
p
r
esen
ts
a
n
d
co
m
p
ar
es th
e
r
es
u
lts
o
b
tai
n
ed
; Sec
tio
n
6
co
n
cl
u
d
es th
e
w
o
r
k
.
2.
AN
I
NS
I
G
H
T
I
N
T
O
CL
O
U
D
CO
M
P
UT
I
NG
C
lo
u
d
co
m
p
u
ti
n
g
ca
n
b
e
d
ef
in
ed
as
an
o
n
-
d
e
m
an
d
w
eb
b
ase
d
s
er
v
ice,
i
n
w
h
ich
s
h
ar
ed
r
eso
u
r
ce
s
ar
e
u
s
ed
f
o
r
ex
ec
u
ti
n
g
a
ta
s
k
to
o
b
tain
th
e
o
u
tco
m
e
i
n
least
p
r
o
b
ab
le
tim
e.
T
h
is
ca
n
b
e
d
o
n
e
b
y
s
h
ar
i
n
g
a
d
atase
t
b
et
w
ee
n
all
t
h
e
li
n
k
ed
p
r
o
ce
s
s
o
r
s
.
T
o
en
ab
le
th
e
i
m
p
l
e
m
en
tatio
n
o
f
co
m
p
le
x
ta
s
k
s
w
it
h
lar
g
e
-
s
ca
le
co
m
p
u
tatio
n
,
clo
u
d
co
m
p
u
ti
n
g
ca
n
b
e
u
s
ed
f
o
r
p
r
o
p
er
u
tili
za
tio
n
o
f
co
m
p
u
tin
g
r
eso
u
r
ce
s
o
n
th
e
n
et
w
o
r
k
.
T
o
ac
co
m
p
li
s
h
ef
f
icie
n
c
y
,
n
o
d
es f
o
r
th
e
task
s
s
h
o
u
ld
b
e
ch
o
s
en
b
ased
o
n
th
e
p
r
o
p
er
ties
o
f
th
e
task
[
2
]
.
Usi
n
g
v
ir
tu
al
s
er
v
er
s
,
clo
u
d
co
m
p
u
ti
n
g
o
f
f
er
s
n
et
w
o
r
k
-
b
ased
s
er
v
ices
to
clien
ts
i
n
d
if
f
er
e
n
t
lo
ca
tio
n
s
.
Vir
t
u
al
in
s
ta
n
ce
s
(
i.
e)
s
er
v
er
s
,
s
to
r
ag
e
o
r
o
t
h
er
n
e
t
w
o
r
k
r
eso
u
r
ce
s
ca
n
b
e
cr
ea
te
d
w
i
th
th
e
h
elp
o
f
v
ir
tu
a
lizatio
n
tech
n
o
lo
g
ie
s
.
T
ask
s
w
h
ich
in
v
o
l
v
es b
i
g
d
ataset
s
,
ca
n
b
e
p
ar
titi
o
n
ed
in
to
m
u
lti
p
le
s
m
al
l d
ataset
s
.
T
h
en
,
th
e
s
m
a
ll
d
ataset
s
ca
n
b
e
ass
ig
n
ed
to
n
u
m
er
o
u
s
p
r
o
ce
s
s
o
r
s
s
o
t
h
at
t
h
e
ta
s
k
s
ca
n
b
e
co
m
p
leted
in
least
f
ea
s
ib
le
ti
m
e.
A
clo
u
d
h
as
a
cl
u
s
ter
o
f
n
o
d
es
w
it
h
a
s
p
ec
if
ic
co
n
f
i
g
u
r
a
tio
n
a
n
d
in
f
r
as
tr
u
ct
u
r
e,
w
h
ic
h
p
r
o
v
id
es
s
er
v
ices
r
eq
u
e
s
ted
b
y
clie
n
ts
.
A
p
r
o
ce
s
s
i
n
g
u
n
it
o
r
n
o
d
e
is
i
d
en
tifie
d
as
a
m
a
s
ter
.
A
l
l
th
e
n
o
d
es
ar
e
lin
k
ed
to
th
e
m
aster
.
T
h
e
m
as
ter
allo
ca
tes
th
e
ta
s
k
s
to
th
e
o
t
h
er
n
o
d
es
co
n
n
ec
ted
to
it.
A
n
o
d
e
d
esi
g
n
ated
as
h
ea
d
n
o
d
e
in
t
h
e
clo
u
d
,
ac
ce
p
ts
t
h
e
tas
k
s
.
T
h
en
,
t
h
e
h
ea
d
n
o
d
es d
e
co
m
p
o
s
es t
h
e
ta
s
k
s
i
n
to
s
u
b
ta
s
k
s
a
n
d
allo
ca
tes
it b
ac
k
to
th
e
m
a
s
ter
.
T
h
e
m
a
s
ter
ass
ig
n
s
t
h
e
s
u
b
tas
k
s
to
t
h
e
n
o
d
es
lin
k
ed
to
it.
T
o
co
m
p
lete
t
h
e
task
s
i
n
th
e
least
p
o
s
s
ib
le
ti
m
e,
t
h
e
n
o
d
es s
h
ar
e
th
e
d
atasets
.
2
.
1
.
Deplo
y
m
e
nt
m
o
dels
o
f
clo
ud
A
clo
u
d
d
e
p
lo
y
m
e
n
t
m
o
d
el
e
m
b
o
d
ies
a
p
ar
ticu
lar
t
y
p
e
o
f
clo
u
d
s
et
u
p
,
m
a
in
l
y
d
i
f
f
er
e
n
tiated
b
y
ac
ce
s
s
,
s
ize
a
n
d
o
w
n
er
s
h
ip
.
Mo
d
els
o
f
clo
u
d
co
m
p
u
t
in
g
[
3
]
b
r
o
ad
ly
f
alls
in
to
t
h
r
ee
ca
t
eg
o
r
ies,
v
iz.
:
P
u
b
lic,
P
r
iv
ate
an
d
H
y
b
r
id
C
lo
u
d
.
A
P
u
b
lic
clo
u
d
is
ac
ce
s
s
ib
le
p
u
b
licl
y
a
n
d
m
a
y
b
e
o
w
n
ed
b
y
o
n
e
o
r
m
o
r
e
o
r
g
a
n
izatio
n
s
(
T
h
ir
d
-
p
ar
t
y
s
er
v
ice
p
r
o
v
id
er
s
)
.
I
n
a
p
u
b
lic
clo
u
d
,
th
e
i
n
f
r
astr
u
ctu
r
e
a
n
d
r
eso
u
r
ce
s
ar
e
p
r
o
v
is
io
n
ed
to
co
n
s
u
m
er
s
th
r
o
u
g
h
clo
u
d
d
eliv
er
y
m
o
d
el
s
.
P
r
iv
ate
clo
u
d
is
o
w
n
ed
a
n
d
b
ei
n
g
o
p
er
ated
b
y
a
p
r
iv
a
te
o
r
g
a
n
izatio
n
.
I
n
a
p
r
iv
ate
clo
u
d
,
t
h
e
s
y
s
te
m
s
,
s
to
r
ag
e
an
d
o
th
er
n
e
t
w
o
r
k
r
es
o
u
r
ce
s
ar
e
ac
ce
s
s
ib
le
to
th
e
m
e
m
b
er
s
o
f
th
e
o
r
g
a
n
izatio
n
.
T
h
e
o
w
n
er
(
i.e
)
th
e
o
r
g
an
izatio
n
h
as t
h
e
a
u
th
o
r
it
y
f
o
r
m
ai
n
tai
n
in
g
t
h
e
in
f
r
as
tr
u
ct
u
r
e
o
f
th
e
clo
u
d
.
A
h
y
b
r
id
clo
u
d
is
a
m
i
x
o
f
d
ep
lo
y
m
e
n
t
m
o
d
els.
A
co
n
s
u
m
er
ca
n
m
ai
n
tai
n
a
p
o
r
tio
n
o
f
th
e
c
lo
u
d
in
f
r
astru
ct
u
r
e
(
w
it
h
s
en
s
iti
v
e
d
ata)
in
th
e
o
r
g
an
izatio
n
an
d
co
n
s
u
m
e
s
s
er
v
ice
s
(
les
s
s
en
s
iti
v
e)
f
r
o
m
t
h
e
p
u
b
lic
clo
u
d
s
.
Dep
lo
y
m
e
n
t
m
o
d
el
s
h
a
v
e
m
er
its
an
d
d
e
m
er
it
s
as
w
ell.
T
h
e
m
ai
n
b
en
e
f
it
w
it
h
a
P
u
b
lic
clo
u
d
is
,
th
e
s
er
v
ice
p
r
o
v
id
es
is
r
esp
o
n
s
ib
l
e
f
o
r
o
p
er
atio
n
s
,
m
ai
n
ten
a
n
ce
an
d
s
er
v
ice
s
in
t
h
e
clo
u
d
.
T
h
e
p
r
o
b
lem
w
i
th
a
p
u
b
lic
clo
u
d
is
,
co
n
s
u
m
er
s
d
ep
en
d
s
to
tall
y
o
n
t
h
e
s
er
v
ice
p
r
o
v
id
er
f
o
r
ac
ce
s
s
in
g
th
e
i
n
f
r
a
s
tr
u
ct
u
r
e
an
d
o
th
er
r
eso
u
r
ce
s
.
P
r
iv
ate
clo
u
d
en
ab
les
u
s
er
s
to
g
o
v
er
n
t
h
e
in
f
r
as
tr
u
ctu
r
e
a
n
d
r
eso
u
r
ce
s
,
s
i
n
ce
it
is
o
w
n
ed
b
y
th
e
o
r
g
an
izatio
n
its
el
f
.
He
n
ce
,
th
e
u
s
er
s
ca
n
co
n
s
u
m
e
t
h
e
in
f
r
ast
r
u
ctu
r
e
b
ased
o
n
t
h
eir
v
ar
y
in
g
d
em
a
n
d
.
T
h
e
k
e
y
is
s
u
e
is
,
o
r
g
an
izatio
n
h
as
to
in
v
est
lo
t
o
f
f
u
n
d
s
f
o
r
s
etti
n
g
u
p
a
p
r
iv
ate
clo
u
d
,
o
p
er
atio
n
s
an
d
m
ai
n
te
n
a
n
ce
.
I
n
a
H
y
b
r
id
clo
u
d
,
d
e
p
lo
y
i
n
g
a
n
ar
ch
itectu
r
e
is
a
n
in
tr
icate
ta
s
k
f
o
r
t
w
o
r
ea
s
o
n
s
:
(
a)
b
ec
au
s
e
o
f
th
e
p
r
o
b
ab
le
di
f
f
er
e
n
ce
s
i
n
clo
u
d
en
v
ir
o
n
m
en
ts
,
(
b
)
d
u
t
ies
o
f
r
eso
u
r
ce
m
a
n
ag
e
m
e
n
t
ar
e
s
h
ar
ed
a
m
o
n
g
t
h
e
p
u
b
li
c
clo
u
d
p
r
o
v
id
er
an
d
th
e
p
r
iv
ate
clo
u
d
p
r
o
v
id
er
(
i.e
)
th
e
o
r
g
an
i
za
tio
n
its
el
f
.
2
.
2
.
Serv
ice
m
o
de
ls
in clo
u
d
Ser
v
ice
m
o
d
els
i
n
clo
u
d
v
ar
y
b
ased
o
n
th
e
r
eq
u
ir
e
m
e
n
ts
o
f
cu
s
to
m
er
s
.
Var
io
u
s
s
er
v
ice
m
o
d
els
in
th
e
clo
u
d
ar
e
SaaS
(
So
f
t
w
ar
e
as a
Ser
v
ice)
,
P
aa
S (
P
latf
o
r
m
as a
Ser
v
ice)
an
d
I
aa
S (
I
n
f
r
a
s
tr
u
ct
u
r
e
as a
Ser
v
ice)
.
So
f
t
w
ar
e
a
s
a
s
er
v
ice
i
s
a
m
o
d
el
w
h
ich
p
r
o
v
id
es
s
o
f
t
war
e
s
er
v
ices
to
t
h
e
c
u
s
to
m
er
s
.
I
n
SaaS
,
cu
s
to
m
er
s
ar
e
n
o
t
p
er
m
itted
t
o
alter
th
e
th
e
ap
p
licatio
n
s
.
P
latf
o
r
m
as
a
Ser
v
ice
m
o
d
el
le
ts
th
e
u
s
er
s
to
u
s
e
A
p
p
licatio
n
P
r
o
g
r
a
m
I
n
ter
f
ac
es
(
A
P
I
)
in
v
ar
io
u
s
la
n
g
u
a
g
es.
T
h
is
e
n
ab
les
t
h
e
u
s
er
s
to
d
ev
elo
p
an
d
cu
s
t
o
m
iz
e
th
e
ap
p
licatio
n
s
f
o
r
th
eir
r
eq
u
i
r
e
m
en
t
s
.
I
n
f
r
as
tr
u
ct
u
r
e
as
a
Ser
v
ice
m
o
d
el
allo
w
s
c
u
s
to
m
er
s
to
u
s
e
an
d
cu
s
to
m
ize
th
e
s
to
r
ag
e
an
d
o
th
er
co
m
p
u
ti
n
g
r
eso
u
r
ce
s
p
er
tin
e
n
t
to
th
e
ir
n
ee
d
s
.
T
h
is
m
o
d
el
allo
w
s
t
h
e
c
u
s
to
m
er
s
to
s
ca
le
u
p
o
r
d
o
w
n
th
e
co
m
p
u
ti
n
g
r
eso
u
r
ce
s
b
ased
o
n
th
eir
d
e
m
a
n
d
.
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
A
P
r
o
lifi
c
S
ch
eme
fo
r
Lo
a
d
B
a
la
n
cin
g
R
elyin
g
o
n
Ta
s
k
C
o
mp
letio
n
Time
(
V
.
A
n
a
n
d
)
1743
3.
T
E
CH
N
I
Q
UE
S AN
D
A
L
G
O
RIT
H
M
S F
O
R
L
O
AD
B
AL
ANCIN
G
T
w
o
k
in
d
s
o
f
al
g
o
r
ith
m
s
ar
e
av
ailab
le
f
o
r
lo
ad
b
alan
cin
g
.
T
h
ey
ar
e
s
tatic
an
d
d
y
n
a
m
ic.
T
h
ey
ar
e
d
if
f
er
e
n
tiated
b
y
t
h
e
d
ec
is
io
n
s
th
at
ar
e
tak
e
n
.
So
m
eti
m
es,
d
ec
is
io
n
s
r
el
y
o
n
th
e
p
r
ese
n
t
s
tate
o
f
t
h
e
s
y
s
te
m
(
d
y
n
a
m
ic)
o
r
al
w
a
y
s
t
h
e
s
a
m
e
(
s
tatic)
[
4
]
.
T
h
e
p
r
o
p
o
s
ed
w
o
r
k
h
as
s
tu
d
ied
t
h
e
s
tatic
m
eth
o
d
s
p
r
esen
ted
in
[
5
]
-
[
7
]
an
d
d
y
n
a
m
ic
p
r
esen
te
d
in
[
8
]
-
[
1
4
]
.
I
n
d
is
tr
ib
u
ted
m
et
h
o
d
s
,
th
e
n
o
d
es
co
m
m
u
n
i
ca
te
w
it
h
ea
c
h
o
th
er
th
r
o
u
g
h
a
n
o
n
-
co
o
p
er
ativ
e
o
r
co
-
o
p
er
ativ
e
s
c
h
e
m
e
[
1
5
]
.
T
h
e
d
is
tr
ib
u
ted
s
c
h
e
m
e
p
r
o
p
o
s
ed
in
[
1
5
]
r
ev
ea
ls
t
h
at
ea
ch
n
o
d
e
in
a
s
y
s
te
m
e
x
ec
u
te
th
e
alg
o
r
it
h
m
a
n
d
also
s
h
ar
e
th
e
r
esp
o
n
s
ib
ili
t
y
o
f
lo
ad
b
alan
cin
g
.
An
alg
o
r
it
h
m
n
a
m
ed
as
L
B
3
M
p
r
esen
ted
in
[
1
6
]
,
h
as
co
m
b
i
n
ed
lo
ad
b
alan
cin
g
m
eth
o
d
s
an
d
co
m
p
let
io
n
ti
m
e.
B
y
m
ain
tai
n
in
g
t
h
e
lo
ad
in
a
b
alan
ce
d
s
tate,
L
B
3
M
h
as
o
f
f
er
ed
ef
f
e
ctiv
e
u
ti
lizatio
n
o
f
r
eso
u
r
ce
s
in
a
clo
u
d
co
m
p
u
ti
n
g
s
et
u
p
.
W
h
en
a
q
u
er
y
i
s
r
ec
ei
v
ed
o
n
a
n
o
d
e,
it
is
s
e
n
t
to
a
ll
t
h
e
n
o
d
es
i
n
t
h
e
clo
u
d
.
T
h
is
k
i
n
d
o
f
d
is
tr
ib
u
tio
n
s
h
o
u
ld
b
e
d
o
n
e
in
s
u
c
h
a
w
a
y
t
h
at
r
ed
u
ce
s
th
e
w
aiti
n
g
ti
m
e
f
o
r
th
e
r
esp
o
n
s
e
f
r
o
m
t
h
e
n
o
d
e
s
.
T
ak
in
g
in
to
ac
co
u
n
t
t
h
e
clo
u
d
en
v
ir
o
n
m
e
n
t,
as
s
es
s
m
e
n
t
o
f
e
x
ec
u
t
io
n
ti
m
e
b
ec
o
m
e
s
n
ec
e
s
s
ar
y
w
it
h
d
ev
ices
o
f
d
if
f
er
en
t
t
y
p
es.
So
m
e
o
f
t
h
e
s
i
g
n
if
ican
t
f
ac
to
r
s
s
h
o
u
ld
b
e
co
n
s
id
er
ed
f
o
r
task
s
c
h
ed
u
li
n
g
o
n
a
n
e
t
w
o
r
k
w
it
h
m
a
n
y
u
s
er
s
an
d
h
eter
o
g
e
n
eo
u
s
d
e
v
ices.
T
h
ese
f
ac
to
r
s
in
cl
u
d
e
h
eter
o
g
en
eit
y
,
n
et
w
o
r
k
u
s
ag
e
a
n
d
tas
k
ex
ec
u
t
io
n
ti
m
e
[
1
7
]
.
As
y
m
m
etr
ic
lo
ad
d
is
tr
ib
u
tio
n
is
a
k
e
y
f
ea
t
u
r
e
in
a
lo
ad
b
alan
cin
g
m
et
h
o
d
.
B
ased
o
n
th
e
co
m
p
u
tatio
n
al
ef
f
icie
n
c
y
,
h
i
g
h
a
m
o
u
n
t
o
f
w
o
r
k
lo
ad
m
a
y
b
e
allo
ca
ted
to
a
n
o
d
e.
B
u
t,
allo
ca
tio
n
o
f
tas
k
s
s
h
o
u
l
d
n
o
t
b
e
d
o
n
e
o
n
l
y
b
ase
d
o
n
th
e
co
m
p
u
tatio
n
a
l
e
f
f
ici
en
c
y
o
f
a
n
o
d
e.
I
n
th
is
co
n
te
x
t,
a
lo
ad
b
ala
n
cin
g
alg
o
r
ith
m
ca
n
b
e
ap
p
lied
f
o
r
a
llo
ca
tin
g
tas
k
s
to
n
o
d
es in
a
h
e
ter
o
g
en
eo
u
s
s
et
u
p
li
k
e
a
clo
u
d
.
A
w
o
r
k
p
r
o
p
o
s
ed
in
[
1
8
]
o
f
f
er
s
a
co
m
p
ete
n
t
lo
c
al
s
ea
r
ch
m
et
h
o
d
f
o
r
s
ch
ed
u
li
n
g
ta
s
k
s
in
h
eter
o
g
e
n
eo
u
s
co
m
p
u
ti
n
g
s
et
u
p
.
T
h
e
au
th
o
r
s
clai
m
t
h
at,
t
h
is
m
et
h
o
d
s
o
b
tain
s
s
c
h
ed
u
le
s
o
f
s
h
o
r
t d
u
r
atio
n
.
An
i
m
p
r
o
v
ed
d
y
n
a
m
ic
lo
ad
b
alan
ce
r
p
r
esen
ted
in
[
1
9
]
h
as
o
b
tain
ed
g
o
o
d
r
esu
l
ts
.
T
h
r
o
u
g
h
t
h
e
r
esu
lt
s
,
th
e
i
m
p
le
m
e
n
tat
io
n
h
as
d
e
m
o
n
s
tr
ated
an
i
m
p
r
o
v
ed
r
eso
u
r
ce
u
tili
za
tio
n
,
p
er
f
o
r
m
a
n
ce
an
d
lo
ad
b
alan
cin
g
.
T
o
p
r
eser
v
e
th
e
p
er
f
o
r
m
an
ce
o
f
co
m
p
u
tin
g
,
a
lo
ad
b
alan
cin
g
al
g
o
r
ith
m
w
i
th
'
P
r
io
r
ity
Acti
v
atio
n
'
ca
n
b
e
u
s
ed
.
T
h
at
is
,
allo
ca
te
th
e
t
ask
s
to
n
o
d
es
w
it
h
les
s
p
r
io
r
ity
w
h
e
n
n
o
d
es
w
it
h
h
i
g
h
p
r
io
r
it
y
ar
e
d
o
w
n
b
e
y
o
n
d
a
ce
r
tain
lev
el.
An
i
m
p
o
r
ta
n
t
asp
ec
t
o
f
lo
ad
b
alan
ci
n
g
i
s
p
er
s
is
te
n
ce
.
I
t
is
ab
o
u
t
h
an
d
li
n
g
in
f
o
r
m
atio
n
ac
r
o
s
s
m
a
n
y
r
eq
u
ests
i
n
a
s
es
s
io
n
.
A
ll
t
h
e
n
o
d
es
ca
n
n
o
t
ac
ce
s
s
th
e
d
ata
if
it
is
s
to
r
ed
o
n
a
p
ar
tic
u
lar
s
er
v
er
.
So
,
th
e
s
u
cc
e
s
s
i
v
e
r
eq
u
est
to
o
th
er
n
o
d
es
co
n
s
u
m
e
s
s
o
m
e
s
p
ac
e
.
T
h
is
b
ec
o
m
e
s
an
is
s
u
e
r
elate
d
to
c
o
m
p
u
tin
g
p
er
f
o
r
m
a
n
ce
a
n
d
h
a
s
to
b
e
ad
d
r
ess
ed
.
T
h
is
is
s
u
e
ca
n
b
e
s
o
l
v
ed
b
y
s
e
n
d
in
g
t
h
e
r
eq
u
est
s
t
o
th
e
s
a
m
e
n
o
d
e
„
n
‟
w
it
h
th
e
in
f
r
as
tr
u
ct
u
r
e
a
n
d
r
es
o
u
r
ce
s
.
B
u
t,
t
h
i
s
b
ec
o
m
e
s
a
ce
n
tr
alize
d
ap
p
r
o
ac
h
.
I
f
t
h
e
p
ar
ti
cu
lar
n
o
d
e
„
n
‟
w
it
h
th
e
d
ata
an
d
o
t
h
er
i
n
f
r
astru
c
tu
r
e
f
ail
s
,
t
h
e
s
es
s
io
n
s
r
u
n
n
i
n
g
o
n
t
h
e
n
o
d
e
w
it
h
t
h
e
p
r
o
ce
s
s
e
s
ar
e
al
s
o
lo
s
t.
T
h
is
ca
n
b
e
r
eso
lv
ed
b
y
h
a
v
in
g
a
b
ac
k
u
p
o
f
t
h
e
n
o
d
e
„
n
‟
.
An
o
th
er
s
o
l
u
tio
n
i
s
,
ap
p
licatio
n
o
f
d
atab
ases
.
T
h
e
d
atab
as
es
ca
n
b
e
u
s
ed
f
o
r
s
to
r
in
g
th
e
r
eso
u
r
ce
s
.
B
y
ta
k
i
n
g
a
b
ac
k
u
p
o
f
d
atab
as
es,
co
n
s
eq
u
e
n
ce
s
o
f
n
o
d
e
f
ail
u
r
es
ca
n
b
e
r
eso
lv
ed
.
B
y
m
a
in
tain
i
n
g
a
b
a
ck
u
p
,
a
b
etter
p
er
f
o
r
m
a
n
ce
ca
n
b
e
atta
in
ed
b
y
d
is
tr
ib
u
ti
n
g
th
e
r
eq
u
e
s
ts
to
n
o
d
es
w
it
h
s
i
m
ilar
d
ata
an
d
in
f
r
astr
u
ctu
r
e.
3
.
1
.
Desig
n o
f
lo
a
d
ba
la
ncing
m
e
t
ho
ds
L
o
ad
b
alan
ci
n
g
ca
n
d
o
n
e
w
it
h
lo
t
o
f
h
ar
d
w
ar
e
a
n
d
s
o
f
t
w
ar
e
av
ailab
le
w
it
h
d
i
f
f
er
en
t
v
e
n
d
o
r
s
.
So
m
e
o
f
t
h
e
m
ar
e
P
o
u
n
d
R
ev
er
s
e
P
r
o
x
y
,
n
g
i
n
x
a
n
d
A
p
ac
h
e
m
o
d
_
p
r
o
x
y
_
b
alan
ce
r
.
T
o
allo
ca
te
lar
g
e
tas
k
s
to
m
an
y
n
o
d
es,
Gea
r
m
a
n
ca
n
b
e
u
s
ed
.
T
h
is
r
esu
lt
s
in
m
i
n
i
m
izi
n
g
t
h
e
co
m
p
let
io
n
ti
m
e
o
f
ta
s
k
s
.
T
o
p
r
o
d
u
ce
a
tr
ee
lik
e
s
tr
u
ct
u
r
e,
m
a
n
y
la
y
er
s
o
f
lo
ad
b
alan
cin
g
ca
n
b
e
d
o
n
e.
T
h
e
h
ig
h
er
lev
el
lo
ad
b
alan
cin
g
m
ec
h
an
i
s
m
d
is
p
e
n
s
es
th
e
tas
k
to
co
m
p
u
ti
n
g
n
o
d
es
in
t
h
e
n
e
x
t
lev
el.
T
h
ese
n
o
d
es
ap
p
l
y
t
h
eir
lo
ad
b
alan
ce
r
s
to
allo
ca
te
th
e
task
s
to
n
o
d
es
in
th
e
n
ex
t
lev
e
l.
T
h
is
m
a
y
en
d
u
p
in
m
an
y
lev
e
l
s
d
ep
en
d
in
g
o
n
th
e
s
ize
o
f
th
e
clo
u
d
(
n
o
d
es a
n
d
in
f
r
astr
u
ctu
r
e)
.
3
.
2
.
Alg
o
rit
h
m
s
f
o
r
s
chedul
ing
Nu
m
er
o
u
s
al
g
o
r
ith
m
s
f
o
r
s
ch
e
d
u
lin
g
t
h
e
tas
k
s
ar
e
av
ailab
le.
B
ased
o
n
f
ac
to
r
s
s
u
c
h
as
co
m
p
u
tatio
n
a
l
ef
f
icien
c
y
an
d
co
m
p
letio
n
ti
m
e,
th
ese
alg
o
r
ith
m
s
a
llo
ca
te
th
e
task
s
to
v
ar
io
u
s
n
o
d
es.
On
e
o
f
th
e
alg
o
r
it
h
m
s
is
R
o
u
n
d
R
o
b
in
Sch
ed
u
l
in
g
L
o
ad
B
alan
cin
g
a
lg
o
r
ith
m
.
I
n
t
h
i
s
alg
o
r
it
h
m
,
th
e
in
co
m
i
n
g
r
eq
u
est
s
ar
e
d
is
p
en
s
ed
i
n
a
s
eq
u
en
ce
to
a
s
et
o
f
s
er
v
er
s
.
T
h
is
alg
o
r
ith
m
co
n
s
id
er
s
all
th
e
s
er
v
er
s
eq
u
all
y
r
eg
ar
d
le
s
s
o
f
t
h
eir
lo
ad
o
r
ca
p
ac
it
y
.
T
h
is
al
g
o
r
ith
m
s
n
e
ed
s
all
t
h
e
s
er
v
er
s
w
it
h
s
i
m
ilar
co
n
f
i
g
u
r
atio
n
s
i
n
ce
a
s
e
r
v
er
w
it
h
a
les
s
er
co
n
f
i
g
u
r
atio
n
(
w
ea
k
s
er
v
er
)
m
a
y
b
e
o
v
er
lo
ad
ed
w
h
ile
r
ec
eiv
i
n
g
m
u
lt
ip
le
r
eq
u
est
s
.
T
h
is
w
ill
a
f
f
ec
t
th
e
p
er
f
o
r
m
a
n
ce
.
Ma
n
y
s
ch
ed
u
lin
g
al
g
o
r
ith
m
s
ar
e
av
ailab
le,
w
h
ich
p
er
f
o
r
m
s
lo
ad
b
alan
cin
g
b
ased
o
n
th
e
f
o
llo
w
in
g
f
ac
to
r
s
: a
ct
iv
e/
in
ac
ti
v
e
s
tat
u
s
o
f
a
n
o
d
e,
n
u
m
b
er
o
f
ac
ti
v
e
l
i
n
k
s
,
s
er
v
er
‟
s
lo
ad
,
s
er
v
er
‟
s
lo
c
atio
n
,
r
esp
o
n
s
e
ti
m
e
an
d
th
e
tr
af
f
ic
h
a
n
d
led
r
ec
en
tl
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
3
,
J
u
n
e
201
8
:
1
7
4
1
–
1746
1744
T
h
e
w
o
r
k
p
r
esen
ted
in
[
2
0
]
h
as
e
x
p
lo
r
ed
v
ar
io
u
s
al
g
o
r
ith
m
s
f
o
r
tas
k
s
ch
ed
u
li
n
g
b
ased
o
n
q
u
alit
y
o
f
s
er
v
ice
(
QOS)
,
w
h
ich
e
n
h
an
c
es
th
e
e
f
f
icien
c
y
b
y
co
n
s
id
er
i
n
g
s
ev
er
al
p
ar
a
m
eter
s
v
iz.
:
ac
ce
p
ted
r
ate,
f
air
n
ess
co
m
p
let
io
n
ti
m
e,
co
s
t,
m
ak
e
s
p
an
an
d
m
i
n
i
m
u
m
co
m
p
le
tio
n
ti
m
e.
An
alg
o
r
it
h
m
ter
m
ed
as
E
x
ten
d
ed
No
n
-
p
r
ee
m
p
ti
v
e
P
P
-
a
w
a
r
e
s
ch
ed
u
l
in
g
(
E
NP
P
)
w
as
p
r
o
p
o
s
ed
in
[
2
1
]
.
R
el
y
i
n
g
o
n
n
o
n
-
e
x
cl
u
s
iv
e
s
ch
e
d
u
li
n
g
,
th
e
alg
o
r
ith
m
h
as
s
c
h
ed
u
led
th
e
task
s
b
y
s
to
p
p
in
g
t
h
e
tas
k
s
w
it
h
t
h
e
h
ig
h
es
t
p
en
alt
y
in
t
h
e
m
in
i
m
u
m
ti
m
e.
T
h
r
o
u
g
h
t
h
e
r
e
s
u
l
ts
,
t
h
e
w
o
r
k
h
as
ac
h
ie
v
ed
a
r
ed
u
ctio
n
i
n
ta
s
k
p
r
o
ce
s
s
i
n
g
ti
m
e
a
n
d
a
m
o
u
n
t
o
f
ta
s
k
s
s
to
p
p
ed
.
Fo
r
o
p
tim
al
s
c
h
ed
u
li
n
g
o
f
tas
k
s
w
i
t
h
m
i
n
i
m
u
m
ta
s
k
e
x
ec
u
ti
o
n
d
elay
i
n
a
d
y
n
a
m
ic
clo
u
d
co
m
p
u
ti
n
g
s
etu
p
,
an
alg
o
r
ith
m
n
a
m
ed
as
OT
B
-
C
S
O
(
Or
th
o
g
o
n
al
T
ag
u
c
h
i
B
ased
-
ca
t
s
w
ar
m
o
p
ti
m
izat
io
n
alg
o
r
ith
m
)
w
as
p
r
o
p
o
s
ed
in
[
2
2
]
.
Ma
p
R
ed
u
ce
is
a
f
r
a
m
e
w
o
r
k
,
w
h
ic
h
p
r
o
ce
s
s
es
b
ig
d
ata
s
ets
t
h
r
o
u
g
h
a
n
u
m
b
er
o
f
n
o
d
es
in
a
clu
s
ter
i
n
p
ar
allel.
E
ac
h
n
o
d
e
in
a
clu
s
te
r
h
as its
o
w
n
s
to
r
ag
e.
Ma
p
an
d
R
ed
u
ce
is
b
ein
g
d
o
n
e
b
y
t
h
is
f
r
a
m
e
w
o
r
k
.
First
i
s
"
Ma
p
",
w
h
ich
co
n
v
er
t
s
a
tas
k
i
n
to
s
u
b
-
tas
k
s
.
T
h
en
t
h
e
s
u
b
-
tas
k
s
ar
e
d
is
p
e
n
s
ed
to
th
e
n
o
d
es
id
en
ti
f
ied
as
“sla
v
e”
i
n
t
h
e
clu
s
ter
.
So
m
eti
m
e,
a
tr
ee
s
tr
u
ctu
r
e
w
it
h
m
u
l
tip
le
lev
el
s
ca
n
b
e
f
o
r
m
ed
b
y
in
cl
u
d
in
g
m
o
r
e
s
u
b
-
tas
k
s
.
T
h
i
s
d
ep
en
d
s
o
n
th
e
s
ize
o
f
th
e
ta
s
k
ta
k
en
i
n
to
ac
co
u
n
t.
P
r
o
ce
s
s
in
g
t
h
e
s
u
b
-
ta
s
k
s
is
d
o
n
e
b
y
“slav
e”
n
o
d
es
an
d
t
h
e
r
esu
lt
s
ar
e
r
etu
r
n
ed
to
th
e
“m
aster
”
n
o
d
e.
Nex
t
i
s
“
R
ed
u
ce
”,
th
e
“m
aster
”
n
o
d
e
co
m
b
i
n
es t
h
e
r
es
u
lts
r
et
u
r
n
ed
b
y
t
h
e
“
s
la
v
e”
n
o
d
es.
Ma
p
p
in
g
ca
n
b
e
d
o
n
e
in
d
ep
en
d
en
tl
y
a
n
d
also
in
p
ar
allel.
4.
T
H
E
P
RO
P
O
SE
D
M
E
T
H
O
D
C
lo
u
d
i
s
a
h
eter
o
g
en
eo
u
s
e
n
v
ir
o
n
m
e
n
t,
w
h
er
e
all
t
h
e
n
o
d
es
ca
n
n
o
t
co
m
p
lete
a
s
et
o
f
ta
s
k
s
i
n
t
h
e
s
a
m
e
d
u
r
atio
n
.
T
h
at
is
,
co
m
p
letio
n
ti
m
e
o
f
ta
s
k
s
d
ep
en
d
s
o
n
t
h
e
ca
p
ab
ilit
y
o
f
ea
c
h
n
o
d
e.
T
h
e
p
r
o
p
o
s
e
d
m
et
h
o
d
E
-
T
S
(
E
f
f
icie
n
t
-
T
ask
Su
b
tr
ac
tio
n
)
as
s
i
g
n
s
t
h
e
tas
k
s
to
n
o
d
es
b
ased
o
n
th
e
ta
s
k
co
m
p
letio
n
t
i
m
e.
T
h
i
s
m
et
h
o
d
ef
f
icie
n
tl
y
al
lo
ca
tes
t
h
e
tas
k
s
to
n
o
d
es
b
y
co
m
p
u
tin
g
th
e
s
u
b
tr
ac
tio
n
o
f
co
m
p
letio
n
ti
m
e
o
f
tas
k
s
.
T
h
e
alg
o
r
ith
m
is
p
r
ese
n
ted
in
S
ec
ti
o
n
4
.
1
.
A
l
g
o
r
ith
m
Step
1
:
A
ll
t
h
e
n
o
d
es
w
it
h
t
h
eir
co
m
p
letio
n
ti
m
e
f
o
r
ea
ch
o
f
th
e
t
ask
s
is
ta
k
en
.
T
h
e
n
,
t
h
e
m
i
n
i
m
u
m
v
al
u
e
s
ar
e
s
u
b
tr
ac
ted
f
r
o
m
t
h
e
m
a
x
i
m
u
m
v
al
u
es.
T
h
en
,
t
h
e
m
a
x
i
m
u
m
ta
s
k
co
m
p
letio
n
ti
m
e
o
f
ea
ch
n
o
d
e
i
s
ch
o
s
en
.
Step
2
:
C
h
o
o
s
e
t
h
e
n
o
d
e
w
it
h
t
h
e
m
a
x
i
m
u
m
s
u
b
tr
ac
tio
n
v
alu
e,
i
f
t
h
e
co
m
p
letio
n
ti
m
e
is
s
a
m
e
f
o
r
m
u
ltip
le
task
s
.
Step
3
:
A
n
o
d
e
w
it
h
m
i
n
i
m
u
m
tas
k
co
m
p
letio
n
ti
m
e
i
s
allo
ca
ted
a
task
.
Step
4
:
I
f
th
e
co
m
p
letio
n
ti
m
e
i
s
s
a
m
e
f
o
r
m
o
r
e
t
h
an
o
n
e
n
o
d
e,
ad
d
th
e
co
m
p
letio
n
ti
m
e
o
f
a
s
p
ec
i
f
ic
tas
k
f
o
r
all
th
e
n
o
d
es.
C
h
o
o
s
e
t
h
e
n
o
d
e
w
it
h
t
h
e
m
a
x
i
m
u
m
v
al
u
e.
Step
5
:
T
h
e
task
is
allo
ca
ted
to
th
e
ch
o
s
en
n
o
d
e
f
o
r
p
r
o
ce
s
s
in
g
.
Step
6
:
C
h
o
o
s
e
t
h
e
n
e
x
t h
ig
h
es
t ta
s
k
co
m
p
let
io
n
ti
m
e.
I
ter
ate
th
e
s
te
p
s
2
to
4
till
all
th
e
task
s
ar
e
co
m
p
leted
.
5.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
ON
T
h
e
p
r
o
p
o
s
ed
m
e
th
o
d
E
-
T
S
(
E
f
f
icien
t
-
T
ask
S
u
b
tr
ac
tio
n
)
is
d
e
m
o
n
s
tr
ated
w
it
h
4
n
o
d
es
a
n
d
4
tas
k
s
.
E
ac
h
e
n
tr
y
in
th
e
tab
le
s
p
ec
if
i
es
t
h
e
ti
m
e
tak
e
n
f
o
r
co
m
p
leti
n
g
t
h
e
tas
k
o
n
t
h
at
n
o
d
e.
A
s
t
ep
-
w
is
e
ill
u
s
tr
atio
n
o
f
th
e
al
g
o
r
ith
m
is
clea
r
l
y
p
r
es
en
ted
in
tab
u
lar
f
o
r
m
.
T
ab
le
1
r
ep
r
esen
ts
th
e
co
m
p
le
tio
n
ti
m
e
o
f
4
ta
s
k
s
f
o
r
4
n
o
d
es.
T
ab
le
2
p
r
esen
ts
t
h
e
co
m
p
u
tatio
n
o
f
d
if
f
er
e
n
ce
in
th
e
co
m
p
letio
n
t
i
m
e
o
f
tas
k
s
f
o
r
all
n
o
d
es
tak
en
in
to
ac
co
u
n
t.
T
ab
le
3
s
h
o
w
s
a
n
o
d
e
„
N
3
‟
w
it
h
m
ax
i
m
u
m
s
u
b
tr
ac
tio
n
v
al
u
e
i
s
ass
i
g
n
ed
th
e
ta
s
k
„
T
1
‟
.
T
ab
le
4
s
h
o
w
s
th
e
a
s
s
i
g
n
m
en
t
o
f
ta
s
k
„
T
2
‟
to
n
o
d
e
„
N
2
‟
w
it
h
t
h
e
n
e
x
t
h
ig
h
es
t
tas
k
co
m
p
letio
n
ti
m
e.
Si
m
ilar
l
y
,
T
ab
le
5
d
ep
icts
th
e
a
s
s
i
g
n
m
e
n
t
o
f
tas
k
s
„
T
4
‟
to
n
o
d
e
„N
1
‟
an
d
„
T
3
‟
to
n
o
d
e
„
N
4
‟.
T
ab
le
1
.
T
ask
co
m
p
letio
n
ti
m
e
f
o
r
4
task
s
w
it
h
4
n
o
d
es
N
o
d
e
s
T
a
sk
s
N
1
N
2
N
3
N
4
T
1
12
11
13
9
T
2
23
15
24
12
T
3
29
24
31
11
T
4
23
16
30
16
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
A
P
r
o
lifi
c
S
ch
eme
fo
r
Lo
a
d
B
a
la
n
cin
g
R
elyin
g
o
n
Ta
s
k
C
o
mp
letio
n
Time
(
V
.
A
n
a
n
d
)
1745
T
ab
le
2
.
Dif
f
er
en
ce
i
n
th
e
co
m
p
letio
n
ti
m
e
o
f
tas
k
s
f
o
r
all
n
o
d
es
N
o
d
e
s
T
a
sk
s
N
1
N
2
N
3
N
4
T
1
12
11
13
9
T
2
23
15
24
12
T
3
29
24
31
11
T
4
23
16
30
16
R
e
su
l
t
o
f
S
u
b
t
r
a
c
t
i
n
g
t
h
e
max
i
m
u
m
a
n
d
mi
n
i
mu
m
c
o
mp
l
e
t
i
o
n
t
i
me
17
13
18
7
T
ab
le
3
.
C
h
o
o
s
e
th
e
n
o
d
e
w
ith
m
ax
i
m
u
m
s
u
b
tr
ac
tio
n
v
a
lu
e
N
o
d
e
s
T
a
sk
s
N
1
N
2
N
3
N
4
T
1
12
11
13
9
T
2
23
15
24
12
T
3
29
24
31
11
T
4
23
16
30
16
R
e
su
l
t
o
f
S
u
b
t
r
a
c
t
i
n
g
t
h
e
max
i
m
u
m
a
n
d
mi
n
i
mu
m
c
o
mp
l
e
t
i
o
n
t
i
me
17
13
18
7
T
ab
le
4
.
C
h
o
o
s
es th
e
n
o
d
e
w
it
h
th
e
n
ex
t h
i
g
h
est ta
s
k
co
m
p
le
tio
n
ti
m
e
N
o
d
e
s
T
a
sk
s
N
1
N
2
N
3
N
4
T
1
12
11
13
9
T
2
23
15
24
12
T
3
29
24
31
11
T
4
23
16
30
16
R
e
su
l
t
o
f
S
u
b
t
r
a
c
t
i
n
g
t
h
e
max
i
m
u
m
a
n
d
mi
n
i
mu
m
c
o
mp
l
e
t
i
o
n
t
i
me
17
13
18
7
T
ab
le
5
.
T
ask
s
„
T
4
‟
an
d
„
T
3
‟
a
s
s
i
g
n
ed
to
n
o
d
es
'
N
1
‟
an
d
„
N
4
‟
N
o
d
e
s
T
a
sk
s
N
1
N
2
N
3
N
4
T
1
12
11
13
9
T
2
23
15
24
12
T
3
29
24
31
11
T
4
23
16
30
16
R
e
su
l
t
o
f
S
u
b
t
r
a
c
t
i
n
g
t
h
e
max
i
m
u
m
a
n
d
mi
n
i
mu
m
c
o
mp
l
e
t
i
o
n
t
i
me
17
13
18
7
W
ith
r
eg
ar
d
to
co
m
p
letio
n
ti
m
e,
th
e
r
es
u
lts
o
f
t
h
e
p
r
o
p
o
s
ed
m
eth
o
d
(
E
-
T
S)
ar
e
ev
alu
ated
w
i
th
L
B
MM
[
2
]
an
d
MM
[
2
3
]
.
L
B
MM
is
L
o
ad
B
alan
ce
Mi
n
-
Min
m
eth
o
d
an
d
MM
is
M
in
_
m
in
m
et
h
o
d
.
T
h
e
r
esu
lt
s
o
b
tain
ed
f
o
r
co
m
p
ar
is
o
n
ar
e
s
h
o
w
n
g
r
ap
h
icall
y
i
n
Fig
u
r
e
1
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
al
lo
ca
t
es tas
k
s
to
all
th
e
n
o
d
es
w
h
er
e
as t
h
e
L
B
MM
[
2
]
an
d
MM
[
2
3
]
ap
p
r
o
ac
h
es h
av
e
n
o
t a
llo
ca
ted
th
e
tas
k
s
to
s
o
m
e
o
f
t
h
e
n
o
d
es.
Fro
m
th
e
r
es
u
lts
,
it
is
e
v
id
en
t
th
at
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
E
-
T
S
(
E
f
f
icie
n
t
-
T
ask
S
u
b
tr
ac
tio
n
)
h
as
ac
co
m
p
lis
h
ed
an
i
m
p
r
o
v
ed
p
er
f
o
r
m
a
n
ce
w
h
e
n
co
m
p
ar
ed
w
it
h
t
h
e
o
th
er
m
e
t
h
o
d
s
(
L
B
MM
,
MM
)
tak
e
n
in
t
o
ac
co
u
n
t.
Fig
u
r
e
1
.
C
o
m
p
ar
is
o
n
o
f
E
-
T
S (
P
r
o
p
o
s
ed
Me
th
o
d
)
w
it
h
L
B
MM
[
2
]
an
d
MM
[2
3]
Me
th
o
d
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
3
,
J
u
n
e
201
8
:
1
7
4
1
–
1746
1746
6.
CO
NCLU
SI
O
N
C
lo
u
d
co
m
p
u
ti
n
g
is
in
t
h
e
li
m
eli
g
h
t
a
n
d
lo
ad
b
alan
ci
n
g
i
s
a
k
e
y
ar
ea
to
b
e
ad
d
r
ess
ed
in
c
lo
u
d
co
m
p
u
ti
n
g
.
So
,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
E
-
T
S
(
E
f
f
icie
n
t
-
T
ask
Su
b
tr
ac
tio
n
)
i
s
d
ev
elo
p
ed
f
o
r
en
h
an
ci
n
g
t
h
e
r
eso
u
r
ce
u
tili
za
tio
n
a
n
d
p
er
f
o
r
m
a
n
ce
b
y
m
ai
n
tai
n
i
n
g
a
b
alan
ce
i
n
allo
ca
ti
n
g
th
e
lo
ad
to
t
h
e
n
o
d
es.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
E
-
T
S
(
E
f
f
icien
t
-
T
ask
Su
b
tr
ac
tio
n
)
as
s
ig
n
s
th
e
ta
s
k
s
to
d
if
f
er
en
t
n
o
d
es
b
ased
o
n
th
e
co
m
p
let
io
n
ti
m
e.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
clai
m
s
its
ad
v
an
ta
g
e
b
y
f
i
n
d
i
n
g
a
n
ap
p
r
o
p
r
iate
n
o
d
e
f
o
r
ea
ch
o
f
t
h
e
task
s
to
b
e
co
m
p
leted
.
T
h
is
w
o
r
k
h
as
ac
co
m
p
lis
h
ed
t
h
e
r
ed
u
ctio
n
i
n
t
h
e
to
tal
ti
m
e
ta
k
e
n
f
o
r
co
m
p
let
io
n
o
f
task
s
(
i.e
)
m
a
k
esp
a
n
.
R
el
y
in
g
o
n
th
e
co
m
p
letio
n
ti
m
e,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
h
as
allo
ca
ted
t
h
e
tas
k
s
to
n
o
d
es.
T
h
is
m
et
h
o
d
ca
n
b
e
en
h
an
ce
d
b
y
allo
ca
ti
n
g
t
h
e
tas
k
s
b
ased
o
n
lo
ca
tio
n
o
f
th
e
s
er
v
er
,
s
er
v
er
‟
s
ca
p
ac
it
y
,
an
d
th
e
a
m
o
u
n
t o
f
tr
af
f
ic
h
an
d
led
b
y
ea
c
h
n
o
d
e.
RE
F
E
R
E
NC
E
S
[1
]
Y.F
.
Hu
,
R.
J.
B
lak
e
a
n
d
D.R.
Em
e
r
so
n
,
“
A
n
o
p
ti
m
a
l
m
ig
ra
t
io
n
a
lg
o
r
it
h
m
f
o
r
d
y
n
a
m
ic
lo
a
d
b
a
lan
c
in
g
”
,
Co
n
c
u
rr
e
n
c
y
-
P
ra
c
ti
c
e
a
n
d
Exp
e
rie
n
c
e
,
v
o
l.
1
0
,
n
o
.
6
,
p
p
.
4
6
7
-
4
8
3
,
1
9
9
8
.
[2
]
S
.
W
a
n
g
,
K.
Ya
n
,
W
.
L
iao
,
a
n
d
S
.
W
a
n
g
,
“
T
o
w
a
rd
s
a
lo
a
d
b
a
lan
c
i
n
g
in
a
th
re
e
-
lev
e
l
c
lo
u
d
c
o
m
p
u
ti
n
g
n
e
tw
o
rk
”
,
in
Co
mp
u
ter
S
c
ien
c
e
a
n
d
I
n
f
o
rm
a
t
io
n
T
e
c
h
n
o
lo
g
y
(
ICCS
IT
),
2
0
1
0
3
rd
I
EE
E
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
,
2
0
1
0
,
v
o
l.
1
,
p
p
.
1
0
8
-
1
1
3
.
[3
]
C.
C.
Ra
o
,
M
.
L
e
e
lara
n
i,
a
n
d
Y.
R.
Ku
m
a
r,
“
Clo
u
d
:
C
o
m
p
u
ti
n
g
S
e
rv
ice
s
a
n
d
De
p
lo
y
m
e
n
t
M
o
d
e
ls”
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
Of
E
n
g
i
n
e
e
rin
g
An
d
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
2
,
n
o
.
1
2
,
p
p
.
3
3
8
9
-
3
3
9
0
,
2
0
1
3
.
[4
]
V
.
A
n
a
n
d
,
N.R.
Ra
a
jan
,
a
n
d
K.
A
n
u
ra
d
h
a
,
“
S
ig
n
if
ica
n
t
f
a
c
to
rs
in
t
h
e
d
e
sig
n
o
f
a
n
e
ff
icie
n
t
d
y
n
a
m
ic
lo
a
d
b
a
lan
c
in
g
a
lg
o
rit
h
m
:
A
n
e
x
p
lo
ra
ti
o
n
”
,
AR
P
N
J
o
u
rn
a
l
o
f
E
n
g
i
n
e
e
rin
g
a
n
d
A
p
p
li
e
d
S
c
ien
c
e
s
,
v
o
l.
1
2
,
n
o
.
3
,
p
p
.
8
4
1
-
8
4
8
,
2
0
1
7
.
[5
]
X
.
T
a
n
g
a
n
d
S
.
T
.
Ch
a
n
so
n
,
“
Op
ti
m
izin
g
sta
ti
c
jo
b
sc
h
e
d
u
l
in
g
in
a
n
e
tw
o
rk
o
f
h
e
tero
g
e
n
e
o
u
s
c
o
m
p
u
ters
”
,
in
Pro
c
e
e
d
in
g
s
o
f
t
h
e
In
tl
.
C
o
n
f
.
o
n
Pa
ra
ll
e
l
Pro
c
e
ss
in
g
,
2
0
0
0
,
p
p
.
3
7
3
-
3
8
2
.
[6
]
A
.
N.
T
a
n
ta
w
i
a
n
d
D.
T
o
w
sle
y
,
“
Op
ti
m
a
l
sta
ti
c
lo
a
d
b
a
lan
c
in
g
i
n
d
istri
b
u
ted
c
o
m
p
u
ter
s
y
ste
m
s
”
,
J
o
u
rn
a
l
o
f
t
h
e
ACM
(
J
AC
M
)
,
v
o
l.
3
2
,
n
o
.
2
,
p
p
.
4
4
5
-
4
6
5
,
1
9
8
5
.
[7
]
O.
F
ra
n
e
k
,
“
A
si
m
p
le
m
e
th
o
d
f
o
r
sta
ti
c
lo
a
d
b
a
la
n
c
in
g
o
f
p
a
ra
ll
e
l
F
DT
D
c
o
d
e
s”
,
in
2
0
1
6
I
n
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
E
lec
tro
ma
g
n
e
ti
c
s in
Ad
v
a
n
c
e
d
Ap
p
li
c
a
ti
o
n
s (
ICEA
A)
,
2
0
1
6
,
p
p
.
5
8
7
-
5
9
0
.
[8
]
J.
A
.
S
tan
k
o
v
ic,
“
A
n
a
d
a
p
ti
v
e
b
id
d
in
g
a
lg
o
rit
h
m
f
o
r
p
ro
c
e
ss
e
s,
c
lu
ste
rs
a
n
d
d
istri
b
u
te
d
g
ro
u
p
s”
,
in
Pro
c
.
o
f
4
th
I
n
t.
Co
n
f.
o
n
Distrib
u
ted
C
o
mp
u
ti
n
g
S
y
ste
ms
,
1
9
8
4
,
p
p
.
4
9
-
59.
[9
]
A
.
Ka
ri
m
i,
F
.
Zara
f
sh
a
n
,
A
.
Ja
n
tan
,
A
.
R.
Ra
m
li
,
a
n
d
M
.
S
a
ri
p
a
n
,
“
A
n
e
w
f
u
z
z
y
a
p
p
ro
a
c
h
f
o
r
d
y
n
a
m
ic
lo
a
d
b
a
lan
c
in
g
a
lg
o
rit
h
m
”
,
a
rXiv
p
re
p
rin
t
a
rX
iv:0
9
1
0
.
0
3
1
7
,
2
0
0
9
.
[1
0
]
D.J.
Ev
a
n
s
a
n
d
W
.
U.N.
Bu
tt
,
“
D
y
n
a
m
ic
lo
a
d
b
a
lan
c
in
g
u
sin
g
tas
k
-
tran
s
f
e
r
p
ro
b
a
b
il
it
ies
”
,
Pa
r
a
l
lel
Co
mp
u
t
in
g
,
v
o
l.
1
9
,
n
o
.
8
,
p
p
.
8
9
7
-
9
1
6
,
1
9
9
3
.
[1
1
]
D.L
.
Ea
g
e
r,
E.
D.
L
a
z
o
w
s
k
a
,
a
n
d
J.
Zah
o
rjan
,
“
A
d
a
p
ti
v
e
lo
a
d
sh
a
rin
g
in
h
o
m
o
g
e
n
e
o
u
s
d
istri
b
u
ted
sy
ste
m
s
”
,
IEE
E
tra
n
sa
c
ti
o
n
s
o
n
so
ft
w
a
re
e
n
g
i
n
e
e
rin
g
,
n
o
.
5
,
p
p
.
6
6
2
-
6
7
5
,
1
9
8
6
.
[1
2
]
B.
A
.
Blak
e
,
“
A
ss
i
g
n
m
e
n
t
o
f
in
d
e
p
e
n
d
e
n
t
tas
k
s
to
m
in
i
m
ize
c
o
m
p
letio
n
ti
m
e
”
,
S
o
ft
wa
re
:
Pra
c
ti
c
e
a
n
d
Exp
e
rie
n
c
e
,
v
o
l.
2
2
,
n
o
.
9
,
p
p
.
7
2
3
-
7
3
4
,
1
9
9
2
.
[1
3
]
A
.
B
a
ra
k
a
n
d
A
.
S
h
il
o
h
,
“
A
d
i
strib
u
te
d
lo
a
d
-
b
a
lan
c
in
g
p
o
li
c
y
f
o
r
a
m
u
lt
ico
m
p
u
ter”
,
S
o
ft
wa
re
:
Pra
c
ti
c
e
a
n
d
Exp
e
rie
n
c
e
,
v
o
l.
1
5
,
n
o
.
9
,
p
p
.
9
0
1
-
9
1
3
,
1
9
8
5
.
[1
4
]
J.
A
.
S
tan
k
o
v
ic,
“
S
i
m
u
latio
n
s
o
f
th
re
e
a
d
a
p
ti
v
e
,
d
e
c
e
n
tralize
d
c
o
n
tro
ll
e
d
,
j
o
b
sc
h
e
d
u
li
n
g
a
lg
o
rit
h
m
s”
,
Co
mp
u
ter
Ne
two
rk
s
(
1
9
7
6
)
,
v
o
l.
8
,
n
o
.
3
,
p
p
.
1
9
9
-
2
1
7
,
1
9
8
4
.
[1
5
]
D.
G
ro
su
a
n
d
A
.
T
.
Ch
ro
n
o
p
o
u
l
o
s,
“
No
n
c
o
o
p
e
ra
ti
v
e
lo
a
d
b
a
lan
c
in
g
in
d
istri
b
u
te
d
sy
st
e
m
s
”
,
J
o
u
rn
a
l
o
f
p
a
ra
ll
e
l
a
n
d
d
istrib
u
ted
c
o
m
p
u
t
in
g
,
v
o
l.
6
5
,
n
o
.
9
,
p
p
.
1
0
2
2
-
1
0
3
4
,
2
0
0
5
.
[1
6
]
C.
Hu
n
g
,
H.
W
a
n
g
,
a
n
d
Y.
Hu
,
“
Eff
icie
n
t
lo
a
d
b
a
lan
c
i
n
g
a
lg
o
rit
h
m
f
o
r
c
lo
u
d
c
o
m
p
u
ti
n
g
n
e
tw
o
rk
”
,
in
I
n
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
In
f
o
rm
a
ti
o
n
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
lo
g
y
(
IS
T
2
0
1
2
),
A
p
ril
,
2
0
1
2
,
p
p
.
2
8
-
30.
[1
7
]
R.
F
.
F
re
u
n
d
e
t
a
l.
,
“
S
c
h
e
d
u
l
in
g
r
e
so
u
rc
e
s
in
m
u
lt
i
-
u
se
r,
h
e
tero
g
e
n
e
o
u
s,
c
o
m
p
u
ti
n
g
e
n
v
iro
n
m
e
n
ts
with
S
m
a
rtNe
t”
,
in
He
ter
o
g
e
n
e
o
u
s C
o
mp
u
ti
n
g
W
o
rk
sh
o
p
,
1
9
9
8
.
(
HCW
9
8
)
Pro
c
e
e
d
i
n
g
s.
1
9
9
8
S
e
v
e
n
th
,
1
9
9
8
,
p
p
.
1
8
4
-
1
9
9
.
[1
8
]
G
.
Rit
c
h
ie
a
n
d
J.
Lev
in
e
,
“
A
f
a
st,
e
ffe
c
ti
v
e
lo
c
a
l
se
a
r
c
h
f
o
r
sc
h
e
d
u
li
n
g
in
d
e
p
e
n
d
e
n
t
j
o
b
s
i
n
h
e
tero
g
e
n
e
o
u
s
c
o
m
p
u
ti
n
g
e
n
v
iro
n
m
e
n
ts”
,
Pro
c
e
e
d
in
g
s
o
f
t
h
e
2
2
n
d
W
o
rk
sh
o
p
o
f
th
e
UK
Pl
a
n
n
i
n
g
a
n
d
S
c
h
e
d
u
l
in
g
S
p
e
c
ia
l
In
ter
e
st
Gr
o
u
p
(
Pl
a
n
S
IG)
,
2
0
0
3
.
[1
9
]
S
.
A
c
h
a
r
y
a
a
n
d
D.
A
.
D‟Mello
,
“
En
h
a
n
c
e
d
d
y
n
a
m
ic
lo
a
d
b
a
lan
c
in
g
a
lg
o
rit
h
m
f
o
r
re
so
u
rc
e
p
ro
v
isio
n
i
n
g
in
c
lo
u
d
”
,
in
2
0
1
6
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
In
v
e
n
ti
v
e
Co
mp
u
t
a
ti
o
n
T
e
c
h
n
o
lo
g
ies
(
ICICT
)
,
2
0
1
6
,
v
o
l.
2
,
p
p
.
1
-
5.
[2
0
]
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
lg
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
.
[2
1
]
F
.
Ho
se
i
n
i
,
M
.
G
.
A
r
a
n
i,
a
n
d
A
.
T
a
g
h
iza
d
e
h
,
“
EN
P
P
:
Ex
ten
d
e
d
n
o
n
-
p
re
e
m
p
ti
v
e
P
P
-
a
w
a
r
e
sc
h
e
d
u
li
n
g
f
o
r
re
a
l
-
ti
m
e
c
lo
u
d
se
rv
ice
s”
,
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
i
n
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
6
,
n
o
.
5
,
p
p
.
2
2
9
1
-
2
2
9
9
,
2
0
1
6
.
[2
2
]
D.
G
a
b
i,
A
.
S
.
Is
m
a
il
,
A
.
Zain
a
l,
a
n
d
Z
.
Zak
a
ria,
“
S
o
lv
in
g
tas
k
sc
h
e
d
u
li
n
g
p
r
o
b
lem
in
c
lo
u
d
c
o
m
p
u
ti
n
g
e
n
v
iro
n
m
e
n
t
u
sin
g
o
r
th
o
g
o
n
a
l
tag
u
c
h
i
-
c
a
t
a
lg
o
rit
h
m
”
,
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
i
n
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
7
,
n
o
.
3
,
p
p
.
1
4
8
9
-
1
4
9
7
,
2
0
1
7
.
[2
3
]
T
.
D.
Bra
u
n
e
t
a
l.
,
“
A
c
o
m
p
a
ris
o
n
o
f
e
lev
e
n
sta
ti
c
h
e
u
risti
c
s
fo
r
m
a
p
p
in
g
a
c
las
s
o
f
in
d
e
p
e
n
d
e
n
t
tas
k
s
o
n
t
o
h
e
tero
g
e
n
e
o
u
s
d
istri
b
u
te
d
c
o
m
p
u
ti
n
g
sy
ste
m
s
”
,
J
o
u
rn
a
l
o
f
Pa
r
a
ll
e
l
a
n
d
Distrib
u
ted
c
o
mp
u
ti
n
g
,
v
o
l.
6
1
,
n
o
.
6
,
p
p
.
8
1
0
-
8
3
7
,
2
0
0
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.