I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
,
p
p
.
1360
~
1
3
6
8
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
1
6
i
3
.
pp
1
3
6
0
-
1
3
6
8
1360
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Desig
ning
self
-
he
a
ling
data
ba
se fabrics
f
o
r
real
-
tim
e pay
ment
ra
ils
Ra
g
hu
G
o
lla
pu
di
F
i
serv
I
n
c
,
U
n
i
t
e
d
S
t
a
t
e
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
2
2
,
2
0
2
6
R
ev
is
ed
Ma
r
2
7
,
2
0
26
Acc
ep
ted
Ap
r
2
6
,
2
0
2
6
Re
a
l
-
ti
m
e
p
a
y
m
e
n
t
p
latf
o
rm
s
o
p
e
ra
ti
n
g
a
t
sc
a
le
fa
c
e
a
n
u
n
fo
r
g
iv
i
n
g
o
p
e
ra
ti
o
n
a
l
re
a
li
t
y
:
e
v
e
n
b
rief
o
u
tag
e
s
tran
sla
te
d
irec
tl
y
i
n
to
fa
il
e
d
tran
sa
c
ti
o
n
s,
re
g
u
lato
r
y
e
x
p
o
s
u
r
e
,
a
n
d
e
ro
d
e
d
c
u
st
o
m
e
r
tru
st.
Da
tab
a
se
re
p
li
c
a
ti
o
n
a
n
d
fa
il
o
v
e
r
a
u
t
o
m
a
ti
o
n
h
a
v
e
m
a
tu
re
d
c
o
n
sid
e
ra
b
l
y
o
v
e
r
th
e
p
a
st
two
d
e
c
a
d
e
s,
y
e
t
a
tro
u
b
li
n
g
b
l
in
d
sp
o
t
re
m
a
in
s.
Re
c
o
v
e
ry
fra
m
e
wo
rk
s
b
u
il
t
fo
r
g
e
n
e
ra
l
-
p
u
r
p
o
se
d
istri
b
u
ted
sy
ste
m
s
we
re
n
e
v
e
r
d
e
sig
n
e
d
with
se
tt
lem
e
n
t
fin
a
li
ty
i
n
m
in
d
,
a
n
d
th
a
t
d
e
sig
n
o
m
issio
n
lea
v
e
s
p
a
y
m
e
n
t
o
p
e
ra
to
rs
e
x
p
o
se
d
to
sp
li
t
-
b
ra
in
sc
e
n
a
rio
s
t
h
a
t
g
e
n
e
ric
h
ig
h
-
a
v
a
il
a
b
il
i
ty
t
o
o
l
in
g
c
a
n
n
o
t
re
li
a
b
l
y
p
re
v
e
n
t.
Th
is
p
a
p
e
r
a
d
d
re
ss
e
s
th
a
t
o
m
issio
n
h
e
a
d
-
o
n
th
r
o
u
g
h
a
se
lf
-
h
e
a
li
n
g
d
a
tab
a
se
fa
b
ric
p
u
r
p
o
se
-
b
u
il
t
fo
r
p
a
y
m
e
n
t
ra
il
e
n
v
i
ro
n
m
e
n
ts.
T
h
e
p
r
o
p
o
se
d
a
u
to
n
o
m
o
u
s
re
sili
e
n
c
e
fa
b
ric
a
r
c
h
it
e
c
tu
re
(ARFA)
o
p
e
ra
tes
a
c
ro
ss
th
re
e
c
o
o
rd
i
n
a
ted
lay
e
rs:
a
c
o
n
t
in
u
o
u
s
m
o
n
i
to
ri
n
g
lay
e
r
th
a
t
h
a
rv
e
sts
tele
m
e
try
fro
m
c
o
m
p
u
te,
sto
ra
g
e
,
a
n
d
n
e
tw
o
rk
su
b
sy
ste
m
s;
a
d
e
c
isio
n
lay
e
r
th
a
t
fu
se
s
ru
le
-
b
a
se
d
h
e
u
risti
c
s
with
a
n
e
n
s
e
m
b
le
o
f
iso
lati
o
n
fo
re
sts,
re
c
u
rr
e
n
t
n
e
u
ra
l
n
e
two
rk
s,
a
n
d
g
ra
d
ien
t
b
o
o
sti
n
g
c
las
sifiers
to
se
p
a
ra
te
g
e
n
u
in
e
fa
u
lt
c
o
n
d
i
ti
o
n
s
fro
m
tran
sie
n
t
n
o
ise
;
a
n
d
a
d
e
term
in
isti
c
a
c
ti
o
n
l
a
y
e
r
th
a
t
e
x
e
c
u
tes
re
c
o
v
e
ry
p
r
o
c
e
d
u
re
s
a
n
c
h
o
re
d
t
o
e
x
p
li
c
it
se
tt
lem
e
n
t
fin
a
li
t
y
c
o
n
stra
in
ts.
In
fa
u
l
t
i
n
jec
ti
o
n
tri
a
ls
c
o
v
e
rin
g
n
o
d
e
c
ra
sh
e
s
,
n
e
tw
o
rk
p
a
rti
ti
o
n
s,
re
p
li
c
a
ti
o
n
lag
,
a
n
d
p
e
rfo
rm
a
n
c
e
d
e
g
ra
d
a
ti
o
n
,
th
e
a
rc
h
it
e
c
tu
re
c
u
t
a
v
e
ra
g
e
re
c
o
v
e
ry
ti
m
e
s
b
y
8
8
%
a
g
a
in
st
m
a
n
u
a
l
b
a
se
li
n
e
s,
re
sto
ri
n
g
se
rv
ice
in
ro
u
g
h
ly
8
se
c
o
n
d
s
ra
th
e
r
t
h
a
n
th
e
1
8
0
se
c
o
n
d
s
th
a
t
h
u
m
a
n
-
d
ri
v
e
n
re
m
e
d
iatio
n
ty
p
ica
ll
y
re
q
u
ires
.
F
a
lse
p
o
sit
iv
e
ra
tes
h
e
ld
b
e
lo
w
2
%
a
c
ro
ss
a
ll
fa
il
u
re
c
a
teg
o
ries
,
a
n
d
t
h
e
sy
ste
m
a
c
h
iev
e
d
a
9
8
%
re
c
o
v
e
ry
su
c
c
e
ss
ra
te.
Tak
e
n
t
o
g
e
t
h
e
r,
th
e
se
re
su
l
ts
m
a
k
e
a
p
ra
c
ti
c
a
l
c
a
se
t
h
a
t
a
u
to
n
o
m
o
u
s
re
sili
e
n
c
e
a
n
d
re
g
u
lat
o
ry
c
o
m
p
li
a
n
c
e
re
in
fo
rc
e
ra
th
e
r
t
h
a
n
c
o
n
f
li
c
t
with
e
a
c
h
o
th
e
r
wh
e
n
t
h
e
re
g
u
lat
o
ry
c
o
n
stra
in
ts are
d
e
sig
n
e
d
i
n
fr
o
m
th
e
sta
rt.
K
ey
w
o
r
d
s
:
Au
to
m
ated
r
ec
o
v
er
y
Au
to
n
o
m
o
u
s
f
au
lt m
a
n
ag
em
e
n
t
Data
b
ase
r
esil
ien
ce
Fau
lt to
ler
an
ce
Self
-
h
ea
lin
g
s
y
s
tem
s
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
R
ag
h
u
Go
llap
u
d
i
Sen
io
r
Or
ac
le
DB
A
an
d
Data
b
ase
R
eliab
ilit
y
E
n
g
in
ee
r
,
Fis
er
v
I
n
c
6
0
0
N
Vel
R
.
Ph
illi
p
s
Av
e,
Milwa
u
k
ee
,
W
I
5
3
2
0
3
,
U
n
ited
State
E
m
ail: r
ea
ch
r
ag
h
u
2
5
1
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Pay
m
en
t
in
f
r
astru
ct
u
r
e
s
its
at
th
e
in
ter
s
ec
tio
n
o
f
two
d
em
an
d
in
g
an
d
s
o
m
etim
es
co
n
f
lictin
g
r
eq
u
ir
em
e
n
ts
:
th
e
o
p
er
atio
n
al
n
ee
d
f
o
r
c
o
n
tin
u
o
u
s
av
ailab
ilit
y
an
d
th
e
r
eg
u
lato
r
y
m
an
d
ate
f
o
r
s
ettlem
en
t
f
in
ality
.
Sy
s
tem
s
p
r
o
ce
s
s
in
g
m
illi
o
n
s
o
f
tr
an
s
ac
tio
n
s
p
er
h
o
u
r
ca
n
n
o
t
to
ler
ate
r
ec
o
v
er
y
p
r
o
ce
d
u
r
es
th
at
s
tr
etch
in
to
m
in
u
tes,
y
et
th
e
f
in
an
cial
co
n
s
eq
u
en
ce
s
o
f
d
ata
in
co
n
s
is
ten
cy
d
u
r
in
g
r
ec
o
v
er
y
ar
e
s
ev
er
e
en
o
u
g
h
th
at
ca
r
ef
u
l,
v
alid
ated
r
em
ed
iatio
n
is
n
o
t
o
p
tio
n
al
[
1
]
.
Mo
d
e
r
n
p
a
y
m
en
t
n
etwo
r
k
s
h
av
e
g
r
o
wn
in
co
m
p
lex
ity
co
n
s
id
er
ab
ly
f
aster
th
an
th
e
to
o
lin
g
d
esig
n
e
d
to
p
r
o
tect
t
h
em
,
ex
p
o
s
in
g
f
ailu
r
e
s
ce
n
ar
io
s
th
at
tr
ad
itio
n
al
h
ig
h
-
av
ailab
ilit
y
m
ec
h
an
is
m
s
s
im
p
ly
wer
e
n
o
t
b
u
ilt to
h
a
n
d
le
[
2
]
.
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
-
8
7
0
8
Desig
n
in
g
s
elf
-
h
ea
lin
g
d
a
ta
b
a
s
e
fa
b
r
ics fo
r
r
ea
l
-
time
p
a
yme
n
t ra
ils
(
R
a
g
h
u
G
o
lla
p
u
d
i
)
1361
W
h
en
s
o
m
eth
in
g
f
ails
in
a
co
n
v
en
tio
n
al
d
e
p
lo
y
m
e
n
t,
h
u
m
a
n
o
p
er
ato
r
s
s
tep
in
ass
ess
in
g
wh
at
wen
t
wr
o
n
g
,
d
ec
id
i
n
g
o
n
a
r
esp
o
n
s
e,
an
d
co
o
r
d
in
atin
g
th
e
ac
t
u
al
f
ailo
v
e
r
o
r
r
estar
t.
T
h
is
ap
p
r
o
ac
h
h
o
ld
s
u
p
r
ea
s
o
n
ab
ly
well
wh
en
f
ailu
r
es
ar
e
is
o
lated
an
d
p
r
ed
ictab
le.
I
t
b
r
ea
k
s
d
o
wn
u
n
d
er
p
r
ec
is
ely
th
e
c
o
n
d
itio
n
s
th
at
m
atter
m
o
s
t:
s
im
u
ltan
eo
u
s
d
eg
r
ad
atio
n
ac
r
o
s
s
m
u
ltip
le
co
m
p
o
n
e
n
ts
,
ca
s
ca
d
in
g
f
ailu
r
es
th
at
o
u
tr
u
n
m
an
u
al
r
esp
o
n
s
e
ca
p
ac
ity
,
an
d
in
f
r
astru
ctu
r
e
ev
en
ts
th
at
s
tr
ik
e
at
3
a.
m
.
o
n
a
h
o
lid
ay
wee
k
en
d
[
3
]
,
[
4
]
.
Me
asu
r
ed
ag
ain
s
t
th
e
s
u
b
-
s
ec
o
n
d
tr
an
s
a
ctio
n
p
r
o
ce
s
s
in
g
d
em
an
d
s
o
f
Fed
No
w
an
d
co
m
p
ar
ab
le
p
ay
m
en
t
r
ails
,
in
cid
en
t
r
esp
o
n
s
e
win
d
o
ws
o
f
9
0
to
3
0
0
s
ec
o
n
d
s
ar
e
n
o
t
to
ler
ab
le
ed
g
e
ca
s
es
—
th
ey
r
ep
r
esen
t
a
s
tr
u
ctu
r
al
g
a
p
b
etwe
en
wh
at
th
e
in
f
r
astru
ctu
r
e
ca
n
d
eliv
er
an
d
w
h
at
th
e
b
u
s
in
ess
r
eq
u
ir
es
[
5
]
.
R
u
le
-
b
ased
au
to
m
atio
n
n
ar
r
o
ws
th
is
g
ap
o
n
l
y
p
ar
tially
.
T
h
r
esh
o
ld
m
o
n
ito
r
s
ca
tc
h
clea
r
-
cu
t
f
ailu
r
e
ev
en
ts
r
eliab
ly
en
o
u
g
h
,
b
u
t
t
h
ey
wer
e
n
o
t
d
esig
n
ed
to
d
e
tect
th
e
s
u
b
tler
d
eg
r
ad
atio
n
p
atter
n
s
—
cr
ee
p
in
g
r
ep
licatio
n
lag
,
a
b
u
f
f
e
r
p
o
o
l
f
illi
n
g
s
lo
wly
,
i
n
ter
m
itten
t
n
etwo
r
k
jitt
er
—
th
at
ten
d
to
p
r
ec
ed
e
ca
tast
r
o
p
h
ic
f
ailu
r
es b
y
m
ea
n
in
g
f
u
l in
ter
v
a
ls
[
6
]
,
[
7
]
.
M
o
r
e
cr
itically
,
s
tatic
r
u
les h
av
e
n
o
n
at
u
r
al
way
t
o
en
co
d
e
s
ettlem
en
t
f
in
ality
co
n
s
tr
ain
ts
.
T
h
ey
ad
d
r
ess
s
y
m
p
to
m
s
with
o
u
t
u
n
d
e
r
s
tan
d
in
g
wh
at
it
m
ea
n
s
f
o
r
a
f
ai
lo
v
er
to
b
e
s
af
e
in
th
e
co
n
tex
t
o
f
an
ac
tiv
e
p
a
y
m
en
t
s
ettlem
en
t
[
8
]
,
[
9
]
.
Ge
o
-
d
is
tr
ib
u
ted
p
ay
m
e
n
t
d
e
p
lo
y
m
e
n
ts
h
av
e
s
h
ar
p
en
ed
th
is
p
r
o
b
lem
co
n
s
id
er
ab
ly
:
en
f
o
r
cin
g
c
o
n
s
is
ten
cy
g
u
a
r
an
tee
s
ac
r
o
s
s
n
o
d
es
s
ep
ar
ated
b
y
r
ea
l
n
etwo
r
k
laten
cy
r
eq
u
ir
es c
o
o
r
d
in
atio
n
lo
g
ic
th
a
t th
r
esh
o
ld
m
o
n
ito
r
in
g
s
im
p
ly
ca
n
n
o
t p
r
o
v
id
e
[
1
0
]
.
T
h
e
ar
ch
itectu
r
e
p
r
esen
ted
in
th
is
p
ap
er
ta
k
es
s
ettlem
en
t
f
in
ality
n
o
t
as
a
p
o
s
t
-
r
ec
o
v
er
y
ch
ec
k
b
u
t
as
a
g
o
v
e
r
n
in
g
co
n
d
itio
n
wo
v
en
in
to
ev
er
y
r
ec
o
v
er
y
d
ec
is
io
n
[
1
1
]
.
I
ts
s
p
ec
if
ic
co
n
tr
i
b
u
tio
n
s
ar
e
th
e
f
o
llo
win
g
:
i)
th
e
au
to
n
o
m
o
u
s
r
esil
ien
ce
f
ab
r
ic
ar
c
h
itectu
r
e
(
AR
FA)
,
a
th
r
ee
-
la
y
er
s
elf
-
h
ea
lin
g
co
n
tr
o
l
f
r
am
ewo
r
k
in
teg
r
atin
g
en
s
em
b
le
m
ac
h
in
e
lear
n
in
g
(
ML
)
b
ased
an
o
m
aly
d
etec
tio
n
,
d
eter
m
in
is
tic
f
ailu
r
e
class
if
icatio
n
,
an
d
c
o
o
r
d
i
n
ated
r
ec
o
v
e
r
y
ex
ec
u
tio
n
;
ii)
a
f
o
r
m
al
d
ec
is
io
n
alg
o
r
ith
m
t
h
at
m
a
p
s
f
ailu
r
e
ty
p
es
to
r
ec
o
v
er
y
ac
tio
n
s
wh
ile
ex
p
licitly
e
n
f
o
r
cin
g
f
i
n
ality
co
n
s
tr
ain
ts
—
a
ca
p
ab
ilit
y
a
b
s
en
t
f
r
o
m
p
u
b
lis
h
ed
f
r
am
ewo
r
k
s
;
iii)
d
ir
ec
t
in
teg
r
ati
o
n
o
f
f
e
d
er
al
r
eg
u
lato
r
y
r
eq
u
ir
e
m
en
ts
in
to
r
ec
o
v
er
y
o
r
ch
estra
tio
n
lo
g
ic,
s
tr
u
ct
u
r
ally
p
r
ev
en
tin
g
s
p
lit
-
b
r
ain
s
ce
n
ar
io
s
f
r
o
m
co
r
r
u
p
tin
g
p
ay
m
en
t
s
et
tlem
en
t
s
tate;
an
d
iv
)
ex
p
er
im
en
tal
r
esu
lts
ac
r
o
s
s
s
im
u
lated
f
au
lt
s
ce
n
ar
i
o
s
d
e
m
o
n
s
tr
atin
g
a
n
8
8
%
im
p
r
o
v
e
m
en
t
in
r
ec
o
v
e
r
y
tim
es
ag
ain
s
t
m
an
u
al
b
aselin
es,
with
f
alse
p
o
s
itiv
e
r
ates
h
eld
b
elo
w
2
%.
Fig
u
r
e
1
h
o
w
th
e
th
r
ee
lay
er
s
in
ter
lo
ck
:
th
e
m
o
n
ito
r
in
g
lay
e
r
co
n
tin
u
o
u
s
ly
f
ee
d
s
telem
etr
y
u
p
war
d
,
th
e
d
ec
is
io
n
lay
er
a
p
p
lies
h
y
b
r
id
class
if
icatio
n
lo
g
ic,
an
d
th
e
ac
tio
n
lay
er
ex
ec
u
tes r
ec
o
v
er
y
s
tep
s
with
in
f
in
ality
-
awa
r
e
b
o
u
n
d
ar
i
es.
Fig
u
r
e
1
.
Au
t
o
n
o
m
o
u
s
r
esil
ien
ce
f
ab
r
ic
ar
c
h
itectu
r
e
(
AR
FA)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
3
6
0
-
1
3
6
8
1362
L
o
o
k
in
g
at
th
e
b
r
o
a
d
er
r
esear
ch
lan
d
s
ca
p
e
,
s
elf
-
h
ea
lin
g
s
y
s
tem
s
h
av
e
m
o
v
e
d
well
p
ast
th
eo
r
etica
l
d
is
cu
s
s
io
n
;
p
r
o
d
u
ctio
n
d
ep
l
o
y
m
en
ts
in
m
is
s
io
n
-
cr
itical
e
n
v
ir
o
n
m
en
ts
n
o
w
r
o
u
tin
ely
d
em
o
n
s
tr
ate
ze
r
o
-
d
o
wn
tim
e
r
ec
o
v
e
r
y
d
r
iv
e
n
b
y
AI
-
b
ased
f
r
a
m
ewo
r
k
s
[
1
]
,
[
1
2
]
,
[
1
3
]
.
Dee
p
lear
n
in
g
,
in
p
a
r
ticu
lar
,
h
as
p
r
o
v
en
ef
f
ec
tiv
e
o
n
h
i
g
h
-
f
r
eq
u
e
n
cy
tr
an
s
ac
tio
n
d
ata,
i
d
en
tify
in
g
a
n
o
m
alies
th
at
co
n
v
en
tio
n
al
th
r
e
s
h
o
ld
m
eth
o
d
s
m
is
s
b
ec
au
s
e
it
an
aly
ze
s
p
atter
n
s
o
v
er
tim
e
r
ath
e
r
th
an
is
o
late
d
m
etr
ic
v
alu
es
[
1
4
]
,
[
9
]
,
[
1
5
]
.
Fau
lt
to
ler
an
ce
r
esear
ch
h
as
m
ap
p
ed
t
h
e
f
ailu
r
e
s
p
ac
e
th
o
r
o
u
g
h
ly
,
f
r
o
m
co
m
p
o
n
e
n
t
o
u
tag
es
to
B
y
za
n
tin
e
f
au
lts
an
d
n
etwo
r
k
p
ar
titi
o
n
s
,
p
r
o
v
id
in
g
th
e
th
eo
r
etica
l
s
ca
f
f
o
ld
in
g
t
h
at
p
r
a
ctica
l
ar
ch
itectu
r
es
lik
e
AR
FA
d
e
p
en
d
o
n
[
1
6
]
,
[
1
7
]
,
[
7
]
.
T
h
e
u
n
iq
u
e
ch
allen
g
es o
f
co
n
s
is
ten
cy
an
d
in
teg
r
ity
in
f
i
n
an
cial
d
ata
h
av
e
d
r
awn
o
n
g
o
i
n
g
r
esear
ch
in
ter
est
[
1
8
]
,
[
1
9
]
,
an
d
B
y
za
n
tin
e
f
au
lt
-
to
ler
an
t
co
n
s
en
s
u
s
alg
o
r
ith
m
s
h
av
e
m
atu
r
ed
to
a
p
o
in
t
wh
er
e
th
ey
ca
n
b
e
u
s
ed
in
p
r
o
d
u
ctio
n
[
1
7
]
,
[
7
]
.
Hig
h
-
av
ailab
ilit
y
d
atab
ase
ar
ch
itectu
r
e
h
as
ad
v
a
n
ce
d
c
o
n
s
id
er
ab
ly
alo
n
g
s
id
e
th
ese
d
e
v
elo
p
m
en
ts
,
with
s
y
n
ch
r
o
n
o
u
s
r
e
p
licatio
n
,
lead
er
elec
tio
n
,
an
d
au
t
o
m
ated
f
ailo
v
e
r
n
o
w
s
tan
d
ar
d
f
ea
tu
r
es
in
m
o
s
t
co
m
m
er
cial
s
y
s
tem
s
[
2
0
]
,
[
4
]
,
[
2
1
]
.
Au
to
n
o
m
o
u
s
d
ata
b
ase
p
latf
o
r
m
s
h
av
e
p
u
s
h
ed
f
u
r
th
er
,
in
co
r
p
o
r
atin
g
p
r
ed
ictiv
e
f
ailu
r
e
an
aly
s
is
,
ad
a
p
tiv
e
s
elf
-
tu
n
in
g
,
an
d
au
to
m
at
ed
b
ac
k
u
p
a
n
d
r
ec
o
v
er
y
[
2
2
]
,
[
2
3
]
,
[
5
]
,
[
1
1
]
.
T
h
e
in
teg
r
atio
n
o
f
ML
in
to
d
atab
a
s
e
m
an
ag
em
en
t
h
as
en
ab
led
a
n
o
m
aly
d
etec
tio
n
a
n
d
ad
ap
tiv
e
r
eso
u
r
ce
allo
ca
tio
n
at
a
s
o
p
h
is
ticatio
n
lev
el
th
at
was
im
p
r
ac
tical
o
n
ly
a
f
ew
y
e
ar
s
ag
o
[
6
]
,
[
2
4
]
.
Pu
b
lis
h
ed
r
e
s
u
lts
f
o
r
co
n
s
en
s
u
s
-
b
ased
f
ailo
v
er
m
ec
h
an
is
m
s
r
ep
o
r
t r
ec
o
v
er
y
win
d
o
ws o
f
3
0
to
1
2
0
s
ec
o
n
d
s
[
1
9
]
,
[
2
5
]
,
a
b
a
s
elin
e
th
at
th
is
wo
r
k
im
p
r
o
v
es u
p
o
n
s
u
b
s
tan
tially
th
r
o
u
g
h
d
ete
r
m
in
is
tic
r
ec
o
v
er
y
s
eq
u
en
cin
g
a
n
d
p
ar
allel
v
alid
atio
n
.
W
h
at
th
e
liter
atu
r
e
n
o
tab
ly
lack
s
is
a
f
r
am
ewo
r
k
th
at
tr
ea
ts
p
ay
m
en
t
s
ettlem
en
t
f
in
ality
as
a
h
ar
d
en
g
in
ee
r
in
g
co
n
s
tr
ain
t
r
at
h
er
th
an
an
o
p
er
atio
n
al
g
u
id
elin
e
to
b
e
s
atis
f
ied
af
ter
th
e
f
ac
t
.
E
x
is
tin
g
s
y
s
tem
s
o
p
tim
ize
f
o
r
av
ailab
ilit
y
o
r
co
n
s
is
ten
cy
in
is
o
latio
n
;
th
e
ar
ch
itectu
r
e
p
r
o
p
o
s
ed
h
er
e
tr
ea
ts
th
e
r
eg
u
lated
in
ter
ac
tio
n
b
etwe
en
th
em
as
th
e
p
r
im
ar
y
d
esig
n
ch
allen
g
e.
T
ab
le
1
d
o
c
u
m
en
ts
wh
er
e
cu
r
r
en
t
AI
-
d
r
iv
en
an
o
m
aly
d
etec
tio
n
ap
p
r
o
ac
h
es
s
tan
d
o
n
th
e
f
alse
p
o
s
itiv
e
d
im
en
s
io
n
—
co
n
tex
t
th
at
s
itu
ate
s
th
e
s
y
s
tem
'
s
s
u
b
-
2
% f
alse p
o
s
itiv
e
r
ate
ag
ain
s
t th
e
3
%
-
5
% r
ates r
ep
o
r
ted
in
t
h
e
s
tate
o
f
th
e
ar
t.
T
ab
le
1
.
AI
-
d
r
iv
e
n
an
o
m
aly
d
etec
tio
n
tech
n
iq
u
es f
o
r
p
a
y
m
e
n
t sy
s
tem
s
D
e
t
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
A
p
p
l
i
c
a
t
i
o
n
a
r
e
a
K
e
y
c
a
p
a
b
i
l
i
t
y
D
e
e
p
l
e
a
r
n
i
n
g
n
e
u
r
a
l
n
e
t
w
o
r
k
s
M
u
l
t
i
-
d
i
me
n
si
o
n
a
l
t
e
l
e
me
t
r
y
a
n
a
l
y
si
s
I
d
e
n
t
i
f
i
e
s
c
o
mp
o
s
i
t
e
f
a
i
l
u
r
e
s
i
g
n
a
t
u
r
e
s
a
c
r
o
ss c
o
r
r
e
l
a
t
e
d
met
r
i
c
s
Te
mp
o
r
a
l
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
Tr
a
n
s
a
c
t
i
o
n
l
a
t
e
n
c
y
mo
n
i
t
o
r
i
n
g
D
e
t
e
c
t
s
g
r
a
d
u
a
l
d
e
g
r
a
d
a
t
i
o
n
b
e
f
o
r
e
v
i
si
b
l
e
s
t
a
l
l
s
o
c
c
u
r
A
d
a
p
t
i
v
e
b
a
se
l
i
n
e
m
o
d
e
l
l
i
n
g
W
o
r
k
l
o
a
d
f
l
u
c
t
u
a
t
i
o
n
m
a
n
a
g
e
me
n
t
D
i
st
i
n
g
u
i
s
h
e
s
h
a
r
m
l
e
ss
v
a
r
i
a
t
i
o
n
s fr
o
m g
e
n
u
i
n
e
p
r
e
c
u
r
so
r
s
S
t
a
t
i
st
i
c
a
l
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
H
i
g
h
-
f
r
e
q
u
e
n
c
y
t
r
a
n
sac
t
i
o
n
s
t
r
e
a
ms
P
r
o
c
e
ss
e
s
t
h
o
u
sa
n
d
s
o
f
d
a
t
a
p
o
i
n
t
s s
i
mu
l
t
a
n
e
o
u
sl
y
B
e
h
a
v
i
o
r
a
l
a
n
o
ma
l
y
c
l
a
ssi
f
i
c
a
t
i
o
n
S
y
st
e
m
h
e
a
l
t
h
a
ssess
me
n
t
S
e
p
a
r
a
t
e
s
n
o
i
se
f
r
o
m
t
r
u
e
i
n
d
i
c
a
t
o
r
s
i
g
n
a
l
s
2.
M
E
T
H
O
D
AR
FA
i
s
o
r
g
an
ized
ar
o
u
n
d
a
s
tr
aig
h
tf
o
r
war
d
p
r
in
cip
le:
o
p
e
r
ato
r
s
m
u
s
t
b
e
ab
le
to
k
n
o
w,
i
n
ad
v
an
ce
,
ex
ac
tly
wh
at
th
e
s
y
s
tem
will
d
o
wh
en
a
s
p
ec
if
ic
f
ailu
r
e
o
cc
u
r
s
.
T
h
at
p
r
e
d
ictab
ilit
y
is
n
o
t
in
ten
s
io
n
with
th
e
u
s
e
o
f
m
ac
h
in
e
lear
n
in
g
;
it
c
o
n
s
tr
ain
s
wh
er
e
ML
is
ap
p
lie
d
.
T
h
e
ar
ch
itectu
r
e
u
s
es
d
ete
r
m
in
is
tic
lo
g
ic
f
o
r
d
ec
is
io
n
s
wh
er
e
co
r
r
ec
t
n
ess
is
n
o
n
-
n
eg
o
tiab
le
a
n
d
r
eser
v
es
ML
f
o
r
th
e
p
atter
n
r
ec
o
g
n
itio
n
task
s
wh
er
e
r
u
le
-
b
ased
ap
p
r
o
ac
h
es
g
en
u
in
ely
f
all
s
h
o
r
t.
T
h
r
ee
la
y
er
s
ex
p
r
ess
th
is
d
iv
is
io
n
o
f
r
esp
o
n
s
ib
ilit
y
:
m
o
n
ito
r
in
g
p
r
o
v
id
es
s
itu
atio
n
al
awa
r
en
es
s
,
d
ec
is
io
n
class
if
ies
an
d
d
ec
i
d
es,
an
d
ac
tio
n
ex
ec
u
tes
with
f
in
ality
awa
r
en
ess
b
u
ilt in
.
2
.
1
.
M
o
nito
ring
la
y
er
Go
o
d
r
ec
o
v
er
y
s
tar
ts
with
g
o
o
d
d
ata.
B
ef
o
r
e
an
y
class
if
icatio
n
o
r
d
ec
is
io
n
lo
g
ic
ca
n
f
u
n
ctio
n
,
th
e
m
o
n
ito
r
in
g
lay
er
m
u
s
t
m
ain
tain
an
ac
cu
r
ate,
cu
r
r
e
n
t
p
ictu
r
e
o
f
clu
s
ter
h
ea
lth
an
d
f
o
r
p
a
y
m
en
t
wo
r
k
lo
ad
s
,
th
at
m
ea
n
s
tr
ac
k
in
g
th
e
m
etr
ics
t
h
at
ac
tu
ally
m
atter
:
tr
an
s
ac
tio
n
laten
cy
,
co
m
m
it
r
ates,
r
ep
licat
io
n
lag
,
b
u
f
f
er
p
o
o
l
f
ill
lev
els,
an
d
lo
ck
wait
tim
es.
C
o
llectin
g
ev
er
y
th
in
g
c
o
n
tin
u
o
u
s
ly
at
m
ax
im
u
m
g
r
a
n
u
lar
ity
wo
u
ld
its
elf
d
eg
r
ad
e
th
e
s
y
s
tem
b
ein
g
m
o
n
ito
r
ed
,
s
o
th
e
la
y
er
u
s
es
ad
ap
tiv
e
s
am
p
lin
g
,
tig
h
ten
in
g
co
llectio
n
in
ter
v
als
wh
en
em
er
g
i
n
g
s
tr
ess
p
atter
n
s
ap
p
ea
r
a
n
d
r
ela
x
in
g
th
e
m
u
n
d
er
s
tab
le
co
n
d
itio
n
s
[
2
6
]
.
Me
tr
ics
ar
r
iv
in
g
f
r
o
m
d
is
tr
ib
u
t
ed
n
o
d
es
ar
e
co
n
s
o
lid
ated
b
y
ag
g
r
eg
atio
n
m
o
d
u
les
in
to
u
n
if
ied
clu
s
ter
h
ea
lth
r
ep
r
esen
tatio
n
s
.
T
h
ese
r
ep
r
esen
tatio
n
s
p
er
s
is
ted
in
ti
m
e
-
s
er
ies
d
atab
ases
,
wh
ich
s
er
v
e
two
p
u
r
p
o
s
es:
th
ey
estab
lis
h
th
e
b
eh
av
io
r
al
b
aselin
es
ag
ain
s
t
wh
ich
an
o
m
alies
ar
e
m
ea
s
u
r
ed
,
an
d
th
ey
p
r
eser
v
e
h
is
to
r
ical
tr
en
d
s
th
at
ca
n
r
ev
ea
l
s
lo
w
-
m
o
v
in
g
d
eg
r
ad
atio
n
p
atter
n
s
b
ef
o
r
e
th
ey
b
ec
o
m
e
ac
u
t
e.
T
h
e
m
o
n
ito
r
in
g
in
f
r
astru
ctu
r
e
d
elib
er
ately
k
ee
p
s
th
r
ee
s
ep
ar
ate
m
etr
ic
p
ip
el
in
es
p
er
f
o
r
m
a
n
ce
,
co
n
s
is
ten
cy
,
an
d
a
v
ailab
ilit
y
,
r
ath
er
th
an
co
llap
s
in
g
th
em
in
to
a
s
in
g
le
h
ea
lth
s
co
r
e.
A
p
er
f
o
r
m
an
ce
an
o
m
aly
a
n
d
a
c
o
n
s
is
ten
cy
an
o
m
aly
m
ay
r
eq
u
ir
e
en
tire
ly
d
if
f
e
r
en
t
r
esp
o
n
s
es
an
d
lo
s
in
g
th
at
d
i
s
tin
ctio
n
ea
r
ly
in
th
e
p
ip
elin
e
m
ak
es
b
en
ef
icial
d
o
wn
s
tr
ea
m
d
ec
is
io
n
s
h
ar
d
er
.
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
-
8
7
0
8
Desig
n
in
g
s
elf
-
h
ea
lin
g
d
a
ta
b
a
s
e
fa
b
r
ics fo
r
r
ea
l
-
time
p
a
yme
n
t ra
ils
(
R
a
g
h
u
G
o
lla
p
u
d
i
)
1363
2
.
2
.
Dec
is
io
n
l
a
y
er
No
t
all
f
ailu
r
es
r
eq
u
ir
e
t
h
e
s
am
e
k
in
d
o
f
r
ea
s
o
n
in
g
.
No
d
e
cr
ash
es,
p
r
im
ar
y
-
r
e
p
lica
p
r
o
m
o
tio
n
s
eq
u
en
ce
s
,
an
d
well
-
ch
ar
ac
te
r
ized
r
ep
licatio
n
f
ailu
r
es
f
o
l
lo
w
k
n
o
wn
s
ig
n
atu
r
es
ac
cu
m
u
lated
f
r
o
m
r
ea
l
p
r
o
d
u
ctio
n
in
cid
en
ts
a
n
d
f
o
r
t
h
ese,
d
eter
m
i
n
is
tic
r
u
le
e
n
g
in
e
s
ar
e
th
e
r
ig
h
t
to
o
l.
T
h
ey
p
r
o
c
ess
in
co
m
in
g
h
ea
lth
d
ata
th
r
o
u
g
h
lo
g
ic
tr
ee
s
th
at
en
co
d
e
o
p
er
ati
o
n
al
e
x
p
er
ien
c
e
d
ir
ec
tly
,
f
ir
in
g
q
u
ick
ly
an
d
p
r
ed
ictab
ly
with
o
u
t
an
y
o
f
t
h
e
v
ar
ian
ce
th
at
a
p
r
o
b
ab
ilis
tic
m
o
d
el
in
tr
o
d
u
ce
s
.
T
h
is
m
atter
s
in
p
ay
m
en
t
en
v
i
r
o
n
m
en
ts
:
wh
en
a
n
o
d
e
f
ails
d
u
r
in
g
p
ea
k
s
ettlem
en
t
h
o
u
r
s
,
o
p
er
ato
r
s
an
d
r
e
g
u
l
ato
r
s
n
ee
d
co
n
f
id
en
ce
th
at
th
e
s
y
s
tem
's
r
esp
o
n
s
e
was n
o
t sh
ap
ed
b
y
a
m
o
d
el
wh
o
s
e
in
ter
n
al
s
tate
is
d
if
f
icu
lt to
in
s
p
ec
t.
Ma
ch
in
e
lear
n
in
g
h
a
n
d
les
wh
at
r
u
les
ca
n
n
o
t
g
r
ad
u
al,
m
u
lt
i
-
d
im
en
s
io
n
al
d
eg
r
a
d
atio
n
p
at
ter
n
s
th
at
d
ev
elo
p
b
el
o
w
th
r
esh
o
ld
b
o
u
n
d
ar
ies
an
d
f
all
o
u
ts
id
e
an
y
k
n
o
wn
f
ailu
r
e
s
ig
n
atu
r
e.
T
h
e
en
s
em
b
le
co
m
b
in
es
is
o
latio
n
f
o
r
ests
f
o
r
s
tatis
tical
o
u
tlier
d
etec
tio
n
,
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
s
f
o
r
tem
p
o
r
al
p
atter
n
r
ec
o
g
n
itio
n
ac
r
o
s
s
m
etr
ic
s
eq
u
en
ce
s
,
an
d
g
r
ad
ien
t
b
o
o
s
tin
g
class
if
ier
s
f
o
r
r
o
b
u
s
t
m
u
lti
-
f
ea
tu
r
e
d
is
cr
im
in
atio
n
[
1
5
]
.
T
h
ese
m
o
d
els
ar
e
u
p
d
ated
as
n
ew
f
a
ilu
r
e
s
ig
n
atu
r
es
ap
p
ea
r
i
n
p
r
o
d
u
ctio
n
d
ata,
b
u
t
th
e
u
p
d
ate
p
r
o
ce
s
s
is
m
an
ag
ed
ca
r
ef
u
lly
:
h
y
p
er
p
a
r
am
eter
ch
an
g
es
g
o
th
r
o
u
g
h
v
alid
atio
n
b
ef
o
r
e
d
ep
lo
y
m
en
t,
an
d
th
e
u
p
d
ate
p
ip
elin
e
is
d
esig
n
ed
to
p
r
ev
e
n
t
a
d
estab
ilized
m
o
d
el
f
r
o
m
in
tr
o
d
u
ci
n
g
th
e
f
alse
p
o
s
itiv
es
i
t
ex
is
t
s
t
o
s
u
p
p
r
ess
.
E
v
er
y
d
etec
ted
an
o
m
aly
is
class
if
ied
b
y
s
ev
er
ity
,
s
co
p
e,
a
n
d
r
e
q
u
ir
ed
i
n
ter
v
en
tio
n
ty
p
e,
g
iv
i
n
g
th
e
ac
tio
n
la
y
er
p
r
ec
is
e,
ac
tio
n
ab
le
in
p
u
ts
.
2
.
3
.
Act
i
o
n
l
a
y
er
R
ec
o
v
er
y
e
x
ec
u
tio
n
is
h
a
n
d
led
b
y
f
o
u
r
s
p
ec
ialized
in
te
r
v
en
tio
n
m
o
d
u
les.
C
o
m
p
o
n
e
n
t
r
estar
t
m
o
d
u
les
ad
d
r
ess
s
er
v
ice
-
lev
el
f
ailu
r
es
th
r
o
u
g
h
tar
g
ete
d
p
r
o
ce
s
s
cy
clin
g
,
p
r
eser
v
in
g
in
-
m
e
m
o
r
y
s
tate
wh
er
ev
er
f
ea
s
ib
le
to
r
ed
u
ce
r
e
s
to
r
atio
n
tim
e
an
d
a
v
o
id
u
n
n
e
ce
s
s
ar
y
d
ata
lo
s
s
.
T
r
af
f
ic
r
er
o
u
tin
g
m
o
d
u
les
r
e
d
ir
ec
t
clien
t
co
n
n
ec
tio
n
s
awa
y
f
r
o
m
d
e
g
r
ad
e
d
co
m
p
o
n
e
n
ts
d
u
r
in
g
f
ailu
r
e
e
v
en
ts
.
T
h
e
r
o
u
tin
g
d
e
cisi
o
n
s
ac
co
u
n
t
f
o
r
m
o
r
e
th
an
s
im
p
le
h
ea
lth
s
tatu
s
:
cu
r
r
en
t
ca
p
ac
ity
u
tili
za
tio
n
an
d
g
eo
g
r
ap
h
ic
d
is
tr
ib
u
tio
n
a
r
e
b
o
th
f
ac
t
o
r
ed
in
,
b
ec
au
s
e
n
aiv
ely
r
o
u
tin
g
all
tr
af
f
ic
t
o
th
e
n
ea
r
est
h
ea
lth
y
n
o
d
e
ca
n
o
v
e
r
lo
ad
it a
n
d
tu
r
n
a
lo
ca
lized
f
ailu
r
e
in
t
o
a
clu
s
te
r
-
wid
e
p
er
f
o
r
m
a
n
ce
p
r
o
b
lem
.
R
ep
lica
r
esy
n
ch
r
o
n
izatio
n
m
o
d
u
les
h
an
d
le
th
e
d
ata
co
n
s
is
ten
cy
p
r
o
b
lem
s
th
at
ar
is
e
wh
e
n
r
ep
licas
f
all
b
eh
in
d
p
r
im
ar
y
lo
g
p
r
o
ce
s
s
in
g
.
Ov
er
tim
e,
an
u
n
a
d
d
r
ess
ed
lag
th
r
ea
ten
s
th
e
r
ec
o
v
e
r
y
p
o
in
t
o
b
jectiv
es
th
at
s
ettlem
en
t
f
in
ality
g
u
ar
an
tees
d
ep
en
d
o
n
.
T
h
ese
m
o
d
u
les
in
itiate
tar
g
eted
r
eb
u
ild
s
f
r
o
m
v
er
if
ied
s
o
u
r
ce
s
,
tr
ac
k
in
g
lo
g
s
eq
u
en
ce
n
u
m
b
e
r
s
th
r
o
u
g
h
o
u
t
t
h
e
tr
a
n
s
f
er
to
p
r
eser
v
e
tr
an
s
ac
tio
n
al
o
r
d
er
i
n
g
.
N
o
d
e
is
o
latio
n
m
o
d
u
les
q
u
ar
an
tin
e
co
m
p
o
n
e
n
ts
s
h
o
win
g
er
r
atic
b
eh
av
i
o
r
b
ef
o
r
e
t
h
eir
f
ailu
r
es
ca
n
p
r
o
p
a
g
ate.
I
s
o
latio
n
is
n
o
t
ap
p
lied
r
ef
lex
iv
ely
—
th
e
m
o
d
u
le
f
ir
s
t
v
e
r
if
ies
th
at
r
em
o
v
i
n
g
a
n
o
d
e
will
n
o
t
d
r
o
p
th
e
clu
s
ter
b
elo
w
q
u
o
r
u
m
,
b
ec
au
s
e
a
Qu
o
r
u
m
L
o
s
s
s
ce
n
ar
io
d
em
a
n
d
s
ca
r
ef
u
l
h
an
d
lin
g
:
in
co
r
r
ec
t
in
ter
v
e
n
tio
n
in
th
is
co
n
d
itio
n
ca
n
p
r
o
d
u
ce
d
ata
co
r
r
u
p
tio
n
r
ath
er
th
an
p
r
ev
e
n
t
it.
T
a
b
le
2
p
lace
s
th
ese
m
ec
h
an
is
m
s
ag
ai
n
s
t
p
u
b
lis
h
ed
ze
r
o
-
d
o
wn
tim
e
r
ec
o
v
er
y
ap
p
r
o
ac
h
e
s
f
o
r
f
in
an
cial
s
y
s
tem
s
.
T
ab
le
2
.
Z
er
o
-
d
o
wn
tim
e
r
ec
o
v
er
y
m
ec
h
a
n
is
m
s
f
o
r
f
in
a
n
cial
s
y
s
tem
s
[
1
2
]
,
[
1
6
]
R
e
c
o
v
e
r
y
me
c
h
a
n
i
sm
I
mp
l
e
me
n
t
a
t
i
o
n
s
t
r
a
t
e
g
y
R
e
si
l
i
e
n
c
e
b
e
n
e
f
i
t
C
o
m
p
o
n
e
n
t
i
so
l
a
t
i
o
n
D
e
g
r
a
d
e
d
r
e
s
o
u
r
c
e
s
se
p
a
r
a
t
e
d
w
i
t
h
o
u
t
serv
i
c
e
d
i
sr
u
p
t
i
o
n
P
r
e
v
e
n
t
s
c
a
sca
d
i
n
g
f
a
i
l
u
r
e
s
a
c
r
o
ss
i
n
f
r
a
s
t
r
u
c
t
u
r
e
D
y
n
a
mi
c
t
r
a
f
f
i
c
r
o
u
t
i
n
g
W
o
r
k
l
o
a
d
sh
i
f
t
e
d
t
o
h
e
a
l
t
h
y
c
o
m
p
o
n
e
n
t
s
a
u
t
o
m
a
t
i
c
a
l
l
y
M
a
i
n
t
a
i
n
s
c
o
n
t
i
n
u
o
u
s
t
r
a
n
sa
c
t
i
o
n
p
r
o
c
e
ss
i
n
g
P
r
e
d
i
c
t
i
v
e
p
r
e
v
e
n
t
i
v
e
a
c
t
i
o
n
s
F
a
i
l
u
r
e
s
a
d
d
r
e
sse
d
d
u
r
i
n
g
l
o
w
-
l
o
a
d
w
i
n
d
o
w
s
El
i
m
i
n
a
t
e
s s
u
d
d
e
n
c
a
t
a
s
t
r
o
p
h
i
c
e
v
e
n
t
s
G
e
o
g
r
a
p
h
i
c
r
e
d
u
n
d
a
n
c
y
d
i
st
r
i
b
u
t
i
o
n
Lo
a
d
b
a
l
a
n
c
e
d
a
c
r
o
ss
mu
l
t
i
p
l
e
l
o
c
a
t
i
o
n
s
En
a
b
l
e
s
d
i
s
a
st
e
r
r
e
c
o
v
e
r
y
c
a
p
a
b
i
l
i
t
i
e
s
I
n
c
r
e
men
t
a
l
h
e
a
l
t
h
o
p
t
i
mi
z
a
t
i
o
n
G
r
a
d
u
a
l
d
e
g
r
a
d
a
t
i
o
n
c
o
r
r
e
c
t
e
d
c
o
n
t
i
n
u
o
u
s
l
y
S
u
st
a
i
n
s
p
e
a
k
p
e
r
f
o
r
ma
n
c
e
l
e
v
e
l
s
2
.
4
.
Self
-
hea
lin
g
lo
g
ic
2
.
4
.
1
.
F
a
ilu
re
c
la
s
s
if
ica
t
io
n
T
h
e
s
y
s
tem
d
is
tin
g
u
is
h
es
f
i
v
e
f
ailu
r
e
ca
teg
o
r
ies,
ea
c
h
p
r
esen
tin
g
a
ch
ar
ac
ter
is
tic
s
y
m
p
t
o
m
p
r
o
f
ile
an
d
ca
llin
g
f
o
r
a
d
if
f
er
e
n
t
r
ec
o
v
er
y
ap
p
r
o
ac
h
.
No
d
e
C
r
ash
Failu
r
es
p
r
o
d
u
ce
an
im
m
ed
iat
e,
to
tal
d
is
co
n
tin
u
ity
ac
r
o
s
s
all
m
o
n
ito
r
in
g
d
im
en
s
io
n
s
;
th
e
s
ig
n
atu
r
e
is
u
n
am
b
ig
u
o
u
s
,
an
d
th
e
r
eq
u
ir
e
d
r
esp
o
n
s
e
is
im
m
ed
iate.
Per
f
o
r
m
an
ce
De
g
r
ad
atio
n
is
co
n
s
id
er
ab
ly
s
u
b
tler
:
r
is
in
g
lat
en
cies
an
d
d
ec
lin
in
g
th
r
o
u
g
h
p
u
t
th
at
m
ay
r
e
f
lect
r
eso
u
r
ce
ex
h
a
u
s
tio
n
,
lo
ck
co
n
ten
tio
n
,
o
r
ea
r
l
y
h
ar
d
war
e
s
tr
ess
,
w
ith
o
u
t
an
y
s
in
g
le
m
etr
ic
b
r
ea
ch
in
g
an
o
b
v
io
u
s
th
r
esh
o
ld
.
R
e
p
l
ic
a
t
i
o
n
L
a
g
ac
c
u
m
u
l
a
t
es
w
h
e
n
r
e
p
l
i
c
as
f
a
ll
p
r
o
g
r
e
s
s
i
v
e
ly
f
u
r
t
h
e
r
b
e
h
i
n
d
p
r
i
m
a
r
y
l
o
g
p
r
o
c
e
s
s
i
n
g
.
L
e
f
t
u
n
a
d
d
r
e
s
s
e
d
,
a
g
r
o
w
i
n
g
l
a
g
e
v
e
n
t
u
a
l
l
y
t
h
r
e
a
t
e
n
s
t
h
e
r
ec
o
v
e
r
y
p
o
i
n
t
o
b
j
e
c
t
i
v
e
s
t
h
a
t
u
n
d
e
r
p
i
n
s
e
t
t
l
e
m
e
n
t
f
i
n
a
l
i
t
y
g
u
a
r
a
n
t
e
es
,
w
h
i
c
h
i
s
w
h
y
t
h
e
c
l
a
s
s
i
f
ic
a
t
i
o
n
l
o
g
ic
d
i
s
t
i
n
g
u
is
h
e
s
l
a
g
s
e
v
e
r
i
t
y
l
e
v
e
l
s
a
n
d
a
p
p
l
i
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
3
6
0
-
1
3
6
8
1364
p
r
o
p
o
r
t
i
o
n
a
t
e
r
e
s
p
o
n
s
es
r
at
h
e
r
t
h
a
n
w
a
it
i
n
g
f
o
r
a
t
h
r
es
h
o
l
d
b
r
e
a
c
h
.
Q
u
o
r
u
m
l
o
s
s
a
r
i
s
es
w
h
en
t
o
o
m
a
n
y
r
e
p
l
i
c
as
g
o
o
f
f
l
i
n
e
t
o
s
u
s
t
ai
n
c
o
n
s
e
n
s
u
s
,
c
r
e
a
ti
n
g
a
s
i
t
u
at
i
o
n
w
h
e
r
e
th
e
w
r
o
n
g
i
n
t
e
r
v
e
n
t
i
o
n
c
a
n
c
au
s
e
d
a
t
a
c
o
r
r
u
p
t
i
o
n
i
n
s
t
e
a
d
o
f
p
r
e
v
e
n
t
i
n
g
i
t
.
C
o
n
s
en
s
u
s
c
o
n
f
l
i
ct
s
,
f
ai
l
u
r
e
s
o
f
d
is
t
r
ib
u
t
e
d
a
g
r
e
e
m
e
n
t
o
n
t
r
a
n
s
a
c
ti
o
n
o
r
d
e
r
i
n
g
o
r
c
l
u
s
t
e
r
l
e
a
d
e
r
s
h
i
p
,
a
r
e
t
h
e
m
o
s
t
d
a
n
g
e
r
o
u
s
f
o
r
p
a
y
m
e
n
t
w
o
r
k
l
o
a
d
s
b
e
c
a
u
s
e
t
h
e
y
a
r
e
p
r
e
c
i
s
e
l
y
th
e
c
o
n
d
i
t
i
o
n
s
u
n
d
e
r
w
h
i
c
h
s
p
li
t
-
b
r
a
i
n
s
c
e
n
a
r
i
o
s
d
e
v
e
l
o
p
i
f
r
e
c
o
v
e
r
y
i
s
n
o
t
g
u
i
d
e
d
b
y
e
x
p
l
i
c
i
t
f
i
n
al
i
t
y
c
o
n
s
t
r
a
i
n
ts
[
1
7
]
.
2
.
4
.
2
.
Dec
is
io
n
lo
g
ic
T
h
e
d
ec
is
io
n
alg
o
r
ith
m
b
elo
w
en
co
d
es
th
e
m
ap
p
in
g
f
r
o
m
f
a
ilu
r
e
class
if
icatio
n
s
to
r
ec
o
v
er
y
ac
tio
n
s
.
T
h
e
d
esig
n
is
d
elib
er
ately
d
et
er
m
in
is
tic:
id
en
tical
in
p
u
ts
alwa
y
s
p
r
o
d
u
ce
id
e
n
tical
o
u
tp
u
ts
.
T
h
is
p
r
o
p
er
ty
is
wh
at
allo
ws
th
e
s
y
s
tem
to
m
ak
e
r
ec
o
v
e
r
y
d
ec
is
io
n
s
th
at
ar
e
d
ef
en
s
ib
le
u
n
d
er
r
eg
u
lato
r
y
r
ev
iew
with
o
u
t
r
eq
u
ir
in
g
h
u
m
an
s
ig
n
-
o
f
f
o
n
e
ac
h
o
n
e.
Alg
o
r
ith
m
1
.
Au
to
n
o
m
o
u
s
r
ec
o
v
er
y
d
ec
is
io
n
lo
g
ic
Input: failure_type, severity_level, system_state, quorum_count
Output: recovery_action, validation_procedure, rollback_strategy
PROCEDURE DetermineRecoveryAction(F, S, Σ, Q)
INITIALIZE action ← NULL, validation ← NULL, rollback ← NULL
IF F = NODE_CRASH THEN
target_replica ← SelectHealthiestReplica(Σ)
IF VerifyReplicaConsistency(target_replica) = TRUE THEN
action ← IMMEDIATE_FAILOVER(target_replica)
validation ← VERIFY_TRANSACTION_LOG_INTEGRITY
rollback ← DEMOTE_REPLICA
ELSE
ESCALATE_TO_MANUAL_INTERVENTION("Inconsistent replica state")
END IF
ELSE IF F = PERFORMANCE_DEGRADATION THEN
IF S < THRESHOLD_MINOR THEN
action ← COMPONENT_RESTART_WITH_STATE_PRESERVATION
validation ← MONITOR_PERFORMANCE_RECOVERY(300)
rollback ← TRAFFIC_REROUTING
ELSE IF S ≥ THRESHOLD_MAJOR THEN
action ← TRAFFIC_REROUTING_TO_HEALTHY_NODES
validation ← VERIFY_CONNECTION_HEALTH
rollback ← NODE_ISOLATION
END IF
ELSE IF F = REPLICATION_LAG THEN
lag_duration ←
CalculateLagDuration(Σ)
IF lag_duration < THRESHOLD_MINOR THEN
action ← MONITOR_CONTINUATION
validation ← TRACK_LAG_REDUCTION_RATE
ELSE IF lag_duration ≥ THRESHOLD_MAJOR THEN
action ← REPLICA_RESYNCHRONIZATION(Σ)
validation ← VERIFY_LAG_REDUCTION
rollback ← REBUILD_FROM_VERIFIED_BACKUP
END IF
ELSE IF F = QUORUM_LOSS THEN
IF Q < MINIMUM_QUORUM_SIZE THEN
action ← NODE_ISOLATION + EMERGENCY_PROMOTION
validation ← VERIFY_QUORUM_RESTORATION
IF VerifyQuorum(Σ) = FALSE THEN
REQUIRE_MANUAL_INTERVENTION("Quorum restoration failed")
END IF
rollback ← MANUAL_INTERVENTION_REQUIRED
END IF
ELSE IF F = CONSENSUS_CONFLICT THEN
action ← COORDINATOR_REELECTION + STATE_RECONCILIATION
validation ← VERIFY_CONSENSUS_ACHIEVEMENT(Σ)
rollback ← SPLIT_BRAIN_RESOLUTION_PROTOCOL
IF DetectSplitBrain(Σ) = TRUE THEN
EnforceSettlementFinalityConstraints()
END IF
END IF
RETURN (action, validation, rollback)
END PROCEDURE
2
.
4
.
3
.
Rec
o
v
er
y
s
equencing
Seq
u
en
cin
g
with
in
r
ec
o
v
er
y
o
p
er
atio
n
s
m
atter
s
as
m
u
ch
as
s
elec
tin
g
th
e
r
ig
h
t
o
p
er
atio
n
.
A
f
ailo
v
er
th
at
p
r
o
m
o
tes
a
r
ep
lica
b
e
f
o
r
e
v
er
if
y
in
g
its
co
n
s
is
ten
cy
ca
n
ca
u
s
e
d
ata
lo
s
s
th
at
n
o
s
u
b
s
eq
u
en
t
ac
tio
n
will
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
-
8
7
0
8
Desig
n
in
g
s
elf
-
h
ea
lin
g
d
a
ta
b
a
s
e
fa
b
r
ics fo
r
r
ea
l
-
time
p
a
yme
n
t ra
ils
(
R
a
g
h
u
G
o
lla
p
u
d
i
)
1365
f
u
lly
r
ev
er
s
e.
Fo
r
n
o
d
e
f
ailo
v
er
,
th
e
r
eq
u
ir
ed
o
r
d
er
in
g
is
:
is
o
late
th
e
p
r
im
ar
y
to
h
alt
ad
d
itio
n
al
wr
ites
,
v
er
if
y
th
at
th
e
p
r
o
m
o
tio
n
ca
n
d
id
ate
h
o
ld
s
ev
er
y
co
m
m
itted
tr
an
s
ac
t
io
n
,
ex
ec
u
te
th
e
p
r
o
m
o
tio
n
wi
th
clu
s
ter
m
etad
ata
u
p
d
ates
to
r
ed
ir
ec
t
clien
ts
,
th
en
in
itiate
r
ec
o
v
er
y
p
r
o
ce
d
u
r
es
f
o
r
th
e
f
ailed
n
o
d
e.
C
o
m
p
o
n
e
n
t
r
estar
t
s
eq
u
en
ce
s
d
r
ain
ex
is
tin
g
co
n
n
ec
tio
n
s
f
ir
s
t,
allo
win
g
in
-
f
lig
h
t
tr
an
s
ac
tio
n
s
to
co
m
p
lete
—
th
en
p
r
eser
v
e
in
-
m
em
o
r
y
s
tate
b
ef
o
r
e
r
estar
tin
g
,
an
d
f
in
ally
r
esto
r
e
clien
t c
o
n
n
ec
tiv
ity
.
R
esy
n
ch
r
o
n
izatio
n
s
eq
u
e
n
ce
s
u
s
e
lo
g
s
eq
u
en
ce
n
u
m
b
e
r
c
o
m
p
ar
is
o
n
to
id
e
n
tify
th
e
m
o
s
t
cu
r
r
en
t
co
n
s
is
ten
t
r
ep
lica,
s
tr
ea
m
m
i
s
s
in
g
tr
an
s
ac
tio
n
s
in
o
r
d
e
r
,
v
alid
ate
co
n
s
is
ten
cy
af
te
r
tr
a
n
s
f
er
,
an
d
o
n
ly
t
h
en
r
etu
r
n
th
e
r
esto
r
e
d
r
ep
lica
to
ac
tiv
e
s
er
v
ice.
T
ab
le
3
m
ap
s
ea
ch
f
ailu
r
e
ty
p
e
to
its
co
r
r
es
p
o
n
d
in
g
r
ec
o
v
er
y
s
eq
u
en
ce
an
d
r
atio
n
ale.
T
ab
le
3
.
Failu
r
e
ty
p
e
to
r
e
co
v
e
r
y
ac
tio
n
m
a
p
p
in
g
F
a
i
l
u
r
e
t
y
p
e
P
r
i
mary
a
c
t
i
o
n
V
a
l
i
d
a
t
i
o
n
me
t
h
o
d
R
o
l
l
b
a
c
k
st
r
a
t
e
g
y
N
o
d
e
c
r
a
sh
I
mm
e
d
i
a
t
e
f
a
i
l
o
v
e
r
R
e
p
l
i
c
a
c
o
n
s
i
st
e
n
c
y
c
h
e
c
k
D
e
mo
t
e
i
f
i
n
c
o
n
si
s
t
e
n
t
P
e
r
f
o
r
ma
n
c
e
d
e
g
r
a
d
a
t
i
o
n
C
o
m
p
o
n
e
n
t
r
e
s
t
a
r
t
/
t
r
a
f
f
i
c
r
e
r
o
u
t
i
n
g
P
e
r
f
o
r
ma
n
c
e
r
e
c
o
v
e
r
y
m
o
n
i
t
o
r
i
n
g
Esc
a
l
a
t
e
i
f
n
o
i
m
p
r
o
v
e
m
e
n
t
R
e
p
l
i
c
a
t
i
o
n
l
a
g
M
o
n
i
t
o
r
/
r
e
s
y
n
c
h
r
o
n
i
z
a
t
i
o
n
La
g
r
e
d
u
c
t
i
o
n
v
e
r
i
f
i
c
a
t
i
o
n
R
e
b
u
i
l
d
f
r
o
m
b
a
c
k
u
p
Q
u
o
r
u
m
l
o
ss
N
o
d
e
i
s
o
l
a
t
i
o
n
+
e
mer
g
e
n
c
y
p
r
o
mo
t
i
o
n
Q
u
o
r
u
m res
t
o
r
a
t
i
o
n
v
e
r
i
f
i
c
a
t
i
o
n
M
a
n
u
a
l
i
n
t
e
r
v
e
n
t
i
o
n
C
o
n
se
n
s
u
s c
o
n
f
l
i
c
t
C
o
o
r
d
i
n
a
t
o
r
r
e
e
l
e
c
t
i
o
n
C
o
n
se
n
s
u
s a
c
h
i
e
v
e
me
n
t
c
h
e
c
k
S
p
l
i
t
-
b
r
a
i
n
r
e
s
o
l
u
t
i
o
n
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
T
estbed
co
nfig
ura
t
io
n a
nd
f
a
ilu
re
inje
ct
io
n
T
h
e
ev
alu
atio
n
u
s
ed
a
s
im
u
la
ted
f
iv
e
-
n
o
d
e
d
ata
b
ase
clu
s
ter
ar
r
an
g
e
d
in
p
r
im
ar
y
-
r
ep
lica
to
p
o
lo
g
y
.
Har
d
war
e
p
r
o
v
is
io
n
in
g
p
er
n
o
d
e,
8
GB
m
em
o
r
y
,
f
o
u
r
C
PU
co
r
es,
an
d
1
0
0
GB
s
to
r
ag
e
wer
e
ch
o
s
en
to
r
ef
lect
m
id
-
tier
p
r
o
d
u
ctio
n
d
atab
ase
s
er
v
er
s
ty
p
ical
o
f
f
in
an
cial
d
ep
lo
y
m
en
ts
,
r
ath
er
t
h
an
o
p
ti
m
izin
g
f
o
r
b
est
-
ca
s
e
n
u
m
b
er
s
.
L
ate
n
cy
was
m
o
d
e
led
at
two
lev
els:
2
m
s
with
in
av
ailab
ilit
y
zo
n
es
an
d
5
0
m
s
ac
r
o
s
s
th
em
,
co
n
s
is
ten
t w
ith
th
e
g
eo
g
r
a
p
h
ic
d
is
tr
ib
u
tio
n
th
at
m
u
lti
-
r
eg
io
n
p
ay
m
en
t in
f
r
astru
ctu
r
e
ac
t
u
ally
o
p
er
ates u
n
d
er
.
Sy
n
th
etic
wo
r
k
l
o
ad
s
wer
e
in
jecte
d
at
r
ates
b
etwe
en
1
,
0
0
0
an
d
1
0
,
0
0
0
t
r
an
s
ac
tio
n
s
p
er
s
ec
o
n
d
,
s
p
an
n
in
g
th
e
r
an
g
e
f
r
o
m
r
o
u
t
in
e
o
f
f
-
p
ea
k
tr
af
f
ic
to
p
ea
k
s
ettlem
en
t
win
d
o
ws.
Fau
lt
in
je
ctio
n
co
v
er
ed
th
r
ee
s
ce
n
ar
io
s
:
ab
r
u
p
t p
r
o
ce
s
s
ter
m
in
atio
n
f
o
r
n
o
d
e
cr
ash
s
im
u
lati
o
n
,
n
etwo
r
k
is
o
latio
n
t
o
f
o
r
ce
p
ar
titi
o
n
h
a
n
d
lin
g
,
an
d
d
elib
e
r
ately
in
tr
o
d
u
ce
d
r
e
p
licatio
n
d
elay
s
r
an
g
in
g
f
r
o
m
1
0
0
m
s
to
m
u
lti
-
s
ec
o
n
d
b
ac
k
l
o
g
s
.
T
h
e
r
ep
licatio
n
lag
s
ce
n
ar
io
s
wer
e
test
ed
at
m
u
ltip
le
s
ev
er
ity
lev
els,
s
p
ec
i
f
ically
to
ev
alu
ate
wh
eth
er
th
e
class
if
icatio
n
lo
g
ic
co
r
r
ec
tly
d
is
tin
g
u
is
h
e
d
lag
c
o
n
d
itio
n
s
th
at
r
e
q
u
ir
ed
im
m
ed
iate
in
ter
v
en
tio
n
f
r
o
m
th
o
s
e
m
ild
en
o
u
g
h
to
m
o
n
ito
r
an
d
allo
w
to
r
ec
o
v
er
o
n
th
eir
o
wn
,
a
d
is
tin
ctio
n
th
at
m
atter
s
s
ig
n
if
ican
tly
f
o
r
av
o
id
in
g
u
n
n
ec
ess
ar
y
r
ec
o
v
er
y
ac
tio
n
s
.
3
.
2
.
Rec
o
v
er
y
perf
o
rm
a
nce
re
s
ults
R
ec
o
v
er
y
tim
e
was
m
ea
s
u
r
ed
en
d
-
to
-
e
n
d
-
f
r
o
m
th
e
m
o
m
e
n
t
a
f
ailu
r
e
was
d
etec
ted
th
r
o
u
g
h
v
alid
ated
s
er
v
ice
r
esto
r
atio
n
.
Ma
n
u
al
p
r
o
ce
d
u
r
es
lo
s
e
tim
e
at
ev
e
r
y
h
a
n
d
o
f
f
:
aler
t
ac
k
n
o
wled
g
m
en
t,
d
iag
n
o
s
is
,
d
ec
is
io
n
,
ex
ec
u
tio
n
,
v
er
if
icati
o
n
.
T
h
ese
d
elay
s
ac
cu
m
u
late,
an
d
in
n
o
d
e
cr
ash
s
ce
n
ar
io
s
th
ey
r
ea
ch
clo
s
e
to
th
r
ee
m
in
u
tes.
Au
to
n
o
m
o
u
s
r
ec
o
v
er
y
r
em
o
v
es
th
e
h
an
d
o
f
f
s
en
tire
ly
,
s
u
b
s
titu
tin
g
p
ar
alle
l
v
alid
atio
n
f
o
r
th
e
s
eq
u
en
tial h
u
m
an
d
ec
is
io
n
ch
a
in
.
No
d
e
cr
ash
r
ec
o
v
e
r
y
p
r
o
d
u
ce
d
th
e
s
tar
k
est
co
n
tr
ast:
8
s
ec
o
n
d
s
f
o
r
au
to
n
o
m
o
u
s
r
ec
o
v
e
r
y
v
er
s
u
s
1
8
0
s
ec
o
n
d
s
u
n
d
er
m
an
u
al
p
r
o
ce
d
u
r
es,
r
o
u
g
h
ly
a
9
6
%
r
e
d
u
ct
io
n
o
n
t
h
at
s
p
ec
if
ic
s
ce
n
ar
io
.
Netwo
r
k
p
ar
titi
o
n
r
ec
o
v
er
y
av
er
ag
e
d
8
8
%
im
p
r
o
v
em
en
t
ac
r
o
s
s
test
ed
co
n
f
ig
u
r
atio
n
s
.
R
ep
licatio
n
lag
s
ce
n
ar
i
o
s
,
wh
er
e
r
ec
o
v
er
y
co
m
p
lex
ity
v
a
r
ies with
lag
s
ev
er
ity
,
s
h
o
wed
7
5
.
6
% im
p
r
o
v
e
m
en
t.
T
h
e
r
ates o
f
f
alse p
o
s
itiv
es r
em
ain
ed
b
elo
w
2
%
ac
r
o
s
s
ev
er
y
f
ailu
r
e
ca
teg
o
r
y
,
with
p
er
f
o
r
m
a
n
ce
d
e
g
r
ad
atio
n
d
etec
tio
n
ac
h
iev
in
g
9
9
.
1
%
ac
cu
r
ac
y
.
R
ec
o
v
er
y
s
u
cc
ess
r
ates
ex
ce
ed
ed
9
8
%
o
v
er
all.
T
h
e
s
m
all
f
r
ac
tio
n
th
at
d
id
n
o
t
co
m
p
lete
au
to
n
o
m
o
u
s
ly
f
o
llo
wed
th
e
escalatio
n
p
ath
s
b
u
ilt
in
to
th
e
alg
o
r
ith
m
,
ed
g
e
-
ca
s
e
co
n
d
itio
n
s
th
e
s
y
s
tem
wa
s
d
esig
n
ed
to
h
an
d
o
f
f
r
ath
e
r
th
an
r
eso
lv
e
alo
n
e.
T
h
e
tim
elin
e
d
iag
r
am
tr
an
s
l
ates
th
e
p
er
f
o
r
m
an
ce
n
u
m
b
e
r
s
in
to
s
o
m
eth
in
g
m
o
r
e
in
t
u
itiv
e
.
T
h
e
s
ig
n
if
ican
t
ef
f
icien
cy
g
ain
s
o
f
th
e
au
to
n
o
m
o
u
s
s
y
s
tem
ar
e
s
u
m
m
ar
ized
in
T
ab
le
4
,
wh
ich
h
ig
h
lig
h
ts
an
av
er
ag
e
r
ec
o
v
er
y
tim
e
im
p
r
o
v
em
en
t
o
f
8
7
.
7
%
ac
r
o
s
s
all
f
ailu
r
e
ca
teg
o
r
ies.
T
h
is
p
er
f
o
r
m
an
ce
g
ap
is
f
u
r
th
er
v
is
u
alize
d
in
Fig
u
r
e
2
,
illu
s
t
r
atin
g
h
o
w
th
e
au
to
n
o
m
o
u
s
ap
p
r
o
ac
h
co
m
p
r
ess
es
th
e
r
ec
o
v
er
y
tim
elin
e
b
y
ex
ec
u
tin
g
v
alid
atio
n
an
d
r
e
m
ed
iatio
n
task
s
in
p
ar
allel,
r
ed
u
cin
g
th
e
to
tal
d
u
r
atio
n
t
o
ju
s
t
3
–
8
s
ec
o
n
d
s
co
m
p
ar
ed
t
o
th
e
7
5
–
1
8
0
s
ec
o
n
d
s
r
eq
u
ir
ed
b
y
m
an
u
al
p
r
o
ce
d
u
r
es.
M
an
u
al
r
ec
o
v
er
y
ac
cu
m
u
lates
d
elay
at
ea
ch
s
eq
u
en
tial
h
an
d
o
f
f
p
o
in
t,
p
r
o
d
u
cin
g
a
s
tair
ca
s
e
p
atter
n
o
f
ela
p
s
ed
tim
e
b
ef
o
r
e
s
er
v
ice
is
r
esto
r
ed
;
au
to
n
o
m
o
u
s
r
ec
o
v
er
y
co
m
p
r
ess
es
th
at
s
am
e
tim
elin
e
b
y
r
u
n
n
in
g
v
alid
ati
o
n
an
d
r
em
e
d
iatio
n
task
s
in
p
ar
allel,
p
r
o
d
u
cin
g
a
m
u
ch
s
h
o
r
ter
p
ath
f
r
o
m
f
ail
u
r
e
d
etec
tio
n
to
co
n
f
ir
m
ed
r
esto
r
atio
n
.
Ov
er
h
ea
d
d
u
r
in
g
n
o
r
m
al
o
p
e
r
atio
n
s
r
em
ain
ed
m
in
im
al,
C
PU u
tili
z
atio
n
r
o
s
e
b
y
u
n
d
er
2
%,
m
e
m
o
r
y
co
n
s
u
m
p
tio
n
s
tay
ed
b
elo
w
1
GB
,
an
d
n
etwo
r
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
3
6
0
-
1
3
6
8
1366
b
an
d
wid
th
u
tili
za
tio
n
r
em
ain
e
d
u
n
d
e
r
1
%.
Du
r
i
n
g
ac
tiv
e
r
e
co
v
er
y
,
p
ea
k
C
PU
o
v
er
h
ea
d
r
ea
ch
ed
1
5
%
f
o
r
1
0
to
3
0
s
ec
o
n
d
s
,
a
tr
an
s
ien
t c
o
s
t
th
at
is
wh
o
lly
ac
ce
p
tab
le
g
i
v
e
n
th
e
g
ain
s
.
T
ab
le
4
.
R
ec
o
v
er
y
tim
e
co
m
p
ar
is
o
n
ac
r
o
s
s
f
ailu
r
e
ca
teg
o
r
ie
s
F
a
i
l
u
r
e
t
y
p
e
M
a
n
u
a
l
r
e
c
o
v
e
r
y
(
sec
o
n
d
s)
A
u
t
o
n
o
m
o
u
s rec
o
v
e
r
y
(
se
c
o
n
d
s)
I
mp
r
o
v
e
m
e
n
t
(
%)
N
o
d
e
c
r
a
sh
(
P
r
i
mar
y
)
1
8
0
8
9
5
.
6
N
o
d
e
c
r
a
sh
(
R
e
p
l
i
c
a
)
1
2
0
6
9
5
.
0
N
e
t
w
o
r
k
p
a
r
t
i
t
i
o
n
(
S
y
mm
e
t
r
i
c
)
1
5
0
18
8
8
.
0
N
e
t
w
o
r
k
p
a
r
t
i
t
i
o
n
(
A
s
y
mm
e
t
r
i
c
)
1
0
5
12
8
8
.
6
R
e
p
l
i
c
a
t
i
o
n
L
a
g
(
S
e
v
e
r
e
)
90
22
7
5
.
6
P
e
r
f
o
r
ma
n
c
e
d
e
g
r
a
d
a
t
i
o
n
60
10
8
3
.
3
A
v
e
r
a
g
e
1
1
7
.
5
1
2
.
7
8
7
.
7
Fig
u
r
e
2
.
R
ec
o
v
e
r
y
tim
elin
e
a
n
aly
s
is
3
.
3
.
Co
m
pa
ra
t
iv
e
a
na
ly
s
is
a
nd
a
pp
lica
bi
lity
T
h
e
p
er
f
o
r
m
an
ce
g
ain
s
ar
e
r
ea
l,
b
u
t
th
ey
d
o
n
o
t
ar
r
iv
e
with
o
u
t
tr
ad
eo
f
f
s
th
at
d
eser
v
e
h
o
n
est
ac
k
n
o
wled
g
m
en
t.
T
h
e
m
o
n
i
to
r
in
g
an
d
d
ec
is
io
n
-
m
ak
i
n
g
in
f
r
astru
ctu
r
e
ad
d
s
s
y
s
tem
co
m
p
lex
ity
-
m
o
r
e
co
m
p
o
n
en
ts
m
ea
n
m
o
r
e
p
o
ten
tial
f
ailu
r
e
p
o
in
ts
,
an
d
th
e
r
ec
o
v
er
y
m
ac
h
in
er
y
its
elf
m
u
s
t
b
e
d
esig
n
ed
s
o
th
at
its
o
wn
f
ailu
r
es
d
o
n
o
t
a
m
p
l
if
y
th
e
in
cid
en
ts
it
was
b
u
ilt
to
r
eso
lv
e
[
1
1
]
,
[
2
5
]
.
ML
c
o
m
p
o
n
en
ts
r
eq
u
ir
e
o
n
g
o
in
g
m
ain
ten
a
n
ce
:
tr
ain
in
g
d
ata
p
ip
elin
es,
q
u
ality
o
v
e
r
s
ig
h
t,
an
d
p
er
io
d
ic
r
ev
alid
at
io
n
as
th
e
f
ailu
r
e
lan
d
s
ca
p
e
ev
o
lv
es.
Failu
r
e
s
c
en
ar
io
s
th
at
g
en
u
i
n
ely
f
all
o
u
ts
id
e
th
e
class
if
icatio
n
s
y
s
te
m
'
s
ex
p
er
ien
ce
wil
l
escalate
to
h
u
m
an
o
p
er
at
o
r
s
-
wh
ich
m
ea
n
s
escalatio
n
p
at
h
way
s
n
ee
d
to
b
e
m
ain
tain
e
d
,
t
ested
,
an
d
s
taf
f
ed
[
3
]
,
[
8
]
,
[
2
1
]
.
T
h
e
u
n
d
e
r
l
y
i
n
g
p
r
i
n
c
i
p
l
e
s
a
r
e
n
o
t
s
p
e
c
i
f
i
c
t
o
p
a
y
m
e
n
t
r
a
i
ls
,
e
v
e
n
t
h
o
u
g
h
t
h
a
t
d
o
m
a
i
n
m
o
t
i
v
a
t
e
d
t
h
e
d
e
s
i
g
n
.
A
n
y
l
a
r
g
e
-
s
c
al
e
d
i
s
tr
i
b
u
t
e
d
s
y
s
t
e
m
w
h
e
r
e
r
e
c
o
v
e
r
y
l
a
t
e
n
c
y
h
a
s
d
i
r
e
c
t
o
p
e
r
a
t
i
o
n
a
l
o
r
f
i
n
a
n
c
i
a
l
c
o
n
s
e
q
u
e
n
c
e
s
-
cl
o
u
d
i
n
f
r
a
s
t
r
u
ctu
r
e
,
c
o
n
t
e
n
t
d
i
s
t
r
i
b
u
t
i
o
n
n
et
w
o
r
k
s
,
t
e
l
ec
o
m
m
u
n
i
c
a
t
i
o
n
s
s
w
i
tch
i
n
g
,
m
a
n
u
f
a
c
t
u
r
i
n
g
c
o
n
t
r
o
l
e
n
v
i
r
o
n
m
e
n
t
s
-
s
t
a
n
d
s
t
o
b
e
n
e
f
i
t
f
r
o
m
t
h
e
s
a
m
e
c
o
m
b
i
n
a
t
i
o
n
o
f
p
r
e
d
i
c
t
i
v
e
d
e
t
e
ct
i
o
n
,
d
e
t
e
r
m
i
n
is
t
i
c
a
c
ti
o
n
m
a
p
p
i
n
g
,
a
n
d
d
o
m
a
i
n
-
c
o
n
s
t
r
a
i
n
t
i
n
t
e
g
r
a
t
i
o
n
[
4
]
,
[
1
3
]
.
T
h
e
s
e
t
t
l
e
m
e
n
t
f
i
n
al
i
t
y
c
o
n
s
t
r
a
i
n
t
is
d
o
m
a
i
n
-
s
p
e
c
i
f
ic
;
t
h
e
a
r
c
h
i
t
e
c
t
u
r
al
p
a
t
t
e
r
n
t
h
at
a
c
c
o
m
m
o
d
a
t
e
s
f
i
r
s
t
-
c
la
s
s
c
o
n
s
t
r
a
i
n
ts
o
f
t
h
a
t
k
i
n
d
is
n
o
t
[
6
]
,
[
2
4
]
.
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
-
8
7
0
8
Desig
n
in
g
s
elf
-
h
ea
lin
g
d
a
ta
b
a
s
e
fa
b
r
ics fo
r
r
ea
l
-
time
p
a
yme
n
t ra
ils
(
R
a
g
h
u
G
o
lla
p
u
d
i
)
1367
4.
CO
NCLU
SI
O
N
T
h
e
co
r
e
ar
g
u
m
en
t
th
is
p
a
p
e
r
m
ak
es
is
th
at
s
ettlem
en
t
f
i
n
ality
an
d
au
to
n
o
m
o
u
s
r
ec
o
v
er
y
ar
e
n
o
t
co
m
p
etin
g
o
b
jectiv
es
—
th
ey
b
ec
o
m
e
m
u
tu
ally
r
ein
f
o
r
cin
g
o
n
ce
f
in
ality
is
b
u
ilt
in
to
th
e
r
ec
o
v
er
y
lo
g
ic
as
a
f
ir
s
t
-
class
co
n
s
tr
ain
t
r
ath
er
th
an
ap
p
en
d
e
d
as
a
p
o
s
t
-
r
ec
o
v
e
r
y
ch
ec
k
.
T
h
e
AR
FA
ar
ch
itectu
r
e
d
em
o
n
s
tr
ates
th
is
th
r
o
u
g
h
b
o
th
d
esig
n
a
n
d
m
ea
s
u
r
em
en
t.
R
ec
o
v
er
y
d
ec
is
io
n
s
g
o
v
er
n
ed
b
y
ex
p
licit
f
in
ality
co
n
s
tr
ain
ts
s
tr
u
ctu
r
ally
p
r
ev
e
n
t
th
e
s
p
lit
-
b
r
ain
co
n
d
itio
n
s
th
at
th
r
e
aten
p
ay
m
en
t
s
ettlem
en
t
in
teg
r
ity
,
wh
ile
th
e
co
m
b
in
atio
n
o
f
p
ar
allel
v
alid
a
tio
n
an
d
p
r
e
-
v
alid
ated
r
ec
o
v
er
y
p
ath
s
d
eliv
er
s
th
e
s
p
ee
d
th
at
m
an
u
al
p
r
o
ce
d
u
r
es
ca
n
n
o
t
m
atc
h
.
Acr
o
s
s
n
o
d
e
cr
ash
es,
n
etwo
r
k
p
a
r
titi
o
n
s
,
r
ep
l
icatio
n
lag
ev
e
n
ts
,
an
d
p
er
f
o
r
m
an
ce
d
e
g
r
ad
atio
n
s
ce
n
ar
io
s
,
av
er
a
g
e
r
ec
o
v
er
y
ti
m
es
f
ell
b
y
8
8
%
r
elativ
e
t
o
m
an
u
al
b
aselin
es.
C
r
itically
,
th
e
p
r
ec
is
io
n
h
eld
u
p
alo
n
g
s
id
e
th
e
s
p
ee
d
:
f
alse
p
o
s
itiv
e
r
ates
s
tay
ed
b
elo
w
2
%
a
cr
o
s
s
all
test
ed
f
ailu
r
e
ca
teg
o
r
ies,
wh
ich
m
atter
s
p
r
ac
tically
b
ec
a
u
s
e
u
n
n
ec
ess
ar
y
r
ec
o
v
er
y
ac
tio
n
s
in
a
p
r
o
d
u
ctio
n
p
ay
m
en
t
en
v
ir
o
n
m
e
n
t
ca
n
b
e
as
d
is
r
u
p
tiv
e
as th
e
f
ailu
r
es th
ey
ar
e
m
ea
n
t t
o
ad
d
r
ess
.
Sev
er
al
d
ir
ec
tio
n
s
r
em
ain
o
p
en
.
Mu
lti
-
r
eg
io
n
co
o
r
d
in
atio
n
u
n
d
er
Fed
No
w
an
d
DORA
r
eg
u
lato
r
y
r
eg
im
es
in
tr
o
d
u
ce
s
cr
o
s
s
-
b
o
r
d
er
s
ettlem
en
t
co
n
s
tr
ain
ts
th
at
th
e
cu
r
r
en
t
ar
ch
itectu
r
e
d
o
es
n
o
t
y
et
ad
d
r
ess
—
ex
ten
d
in
g
AR
FA
to
th
at
en
v
ir
o
n
m
e
n
t
is
th
e
m
o
s
t
im
m
e
d
iate
p
r
io
r
ity
.
Valid
atin
g
th
e
ap
p
r
o
ac
h
o
n
liv
e
p
ay
m
en
t
in
f
r
astru
ctu
r
e,
r
ath
er
th
an
s
im
u
latio
n
,
is
a
n
ec
ess
ar
y
s
tep
to
war
d
p
r
o
d
u
ctio
n
co
n
f
id
en
ce
.
Ad
a
p
tiv
e
ML
r
etr
ain
in
g
p
ip
elin
es
f
o
r
e
v
o
lv
in
g
f
ailu
r
e
s
ig
n
atu
r
es,
in
t
eg
r
atio
n
with
clo
u
d
-
n
ativ
e
d
a
tab
ase
ar
ch
itectu
r
es
b
u
ilt
ar
o
u
n
d
g
eo
-
d
is
tr
ib
u
ted
c
o
n
s
en
s
u
s
p
r
o
to
c
o
ls
,
an
d
ap
p
li
ca
b
ilit
y
s
tu
d
ies
in
telec
o
m
m
u
n
icatio
n
s
,
h
ea
lth
ca
r
e
tr
an
s
ac
tio
n
s
y
s
tem
s
,
an
d
d
ig
it
al
ass
et
s
ett
lem
en
t
p
latf
o
r
m
s
e
ac
h
r
ep
r
esen
t
p
r
o
d
u
ctiv
e
d
ir
ec
tio
n
s
f
o
r
e
x
ten
d
in
g
th
is
wo
r
k
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
au
th
o
r
s
ac
k
n
o
wled
g
e
t
h
e
tech
n
ical
in
f
r
astru
ctu
r
e
p
r
o
v
id
ed
b
y
Fis
er
v
I
n
c
f
o
r
c
o
n
d
u
ctin
g
th
e
s
im
u
latio
n
-
b
ased
ev
alu
atio
n
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
r
ec
eiv
ed
n
o
s
p
e
cif
ic
r
esear
ch
g
r
an
t
o
r
co
n
tr
ac
t
f
r
o
m
an
y
f
u
n
d
i
n
g
ag
en
c
y
in
th
e
p
u
b
lic,
co
m
m
er
cial,
o
r
n
o
t
-
f
o
r
-
p
r
o
f
it secto
r
s
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
R
ag
h
u
Go
llap
u
d
i
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
s
im
u
latio
n
d
ata
an
d
r
ec
o
v
er
y
m
et
r
ics
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
a
v
ai
lab
le
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
au
th
o
r
u
p
o
n
r
ea
s
o
n
ab
le
r
e
q
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
D
.
S
a
h
u
e
t
a
l
.
,
“
A
d
a
p
t
i
v
e
f
a
u
l
t
t
o
l
e
r
a
n
c
e
me
c
h
a
n
i
sms
f
o
r
e
n
su
r
i
n
g
h
i
g
h
a
v
a
i
l
a
b
i
l
i
t
y
o
f
d
i
g
i
t
a
l
t
w
i
n
s
i
n
d
i
st
r
i
b
u
t
e
d
e
d
g
e
c
o
m
p
u
t
i
n
g
sy
st
e
ms,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
r
t
s
,
v
o
l
.
1
5
,
n
o
.
1
,
p
.
4
1
6
7
6
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
5
-
2
5
5
9
0
-
4.
[
2
]
J.
R
.
C
a
mp
o
s,
E.
C
o
s
t
a
,
a
n
d
M
.
V
i
e
i
r
a
,
“
O
n
l
i
n
e
f
a
i
l
u
r
e
p
r
e
d
i
c
t
i
o
n
t
h
r
o
u
g
h
f
a
u
l
t
i
n
j
e
c
t
i
o
n
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
:
M
e
t
h
o
d
o
l
o
g
y
a
nd
c
a
se
s
t
u
d
y
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
I
n
t
e
rn
a
t
i
o
n
a
l
S
y
m
p
o
s
i
u
m
o
n
S
o
f
t
w
a
re
Re
l
i
a
b
i
l
i
t
y
En
g
i
n
e
e
ri
n
g
,
I
S
S
R
E
,
2
0
2
3
,
p
p
.
4
5
1
–
4
6
1
,
d
o
i
:
1
0
.
1
1
0
9
/
I
S
S
R
E
5
9
8
4
8
.
2
0
2
3
.
0
0
0
2
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
1
6
,
No
.
3
,
J
u
n
e
20
2
6
:
1
3
6
0
-
1
3
6
8
1368
[
3
]
R
.
A
l
a
m
l
e
h
a
n
d
N
.
El
Em
a
m,
“
A
s
u
r
v
e
y
o
f
f
a
u
l
t
t
o
l
e
r
a
n
c
e
t
e
c
h
n
i
q
u
e
s
i
n
d
i
st
r
i
b
u
t
e
d
s
y
s
t
e
ms,
”
i
n
Ea
s
y
C
h
a
i
r
Pre
p
ri
n
t
s
,
2
0
2
4
,
v
o
l
.
3
4
0
6
,
p
.
2
0
0
0
7
,
d
o
i
:
1
0
.
1
0
6
3
/
5
.
0
3
1
8
9
2
9
.
[
4
]
R
.
M
o
h
a
m
me
d
,
“
Th
e
f
u
t
u
r
e
o
f
s
i
t
e
r
e
l
i
a
b
i
l
i
t
y
e
n
g
i
n
e
e
r
i
n
g
i
n
f
i
n
a
n
c
i
a
l
p
l
a
t
f
o
r
ms:
e
n
s
u
r
i
n
g
u
p
t
i
me
f
o
r
m
u
l
t
i
-
b
i
l
l
i
o
n
-
d
o
l
l
a
r
t
r
a
n
s
a
c
t
i
o
n
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Em
e
r
g
i
n
g
T
r
e
n
d
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
I
T
,
v
o
l
.
7
,
n
o
.
1
,
2
0
2
6
,
d
o
i
:
1
0
.
6
3
2
8
2
/
3
0
5
0
-
9
2
4
6
.
I
JETC
S
I
T
-
V
7
I
1
P
1
1
0
.
[
5
]
A
.
S
.
M
.
N
o
o
r
,
A
.
F
.
C
.
F
a
u
z
i
,
A
.
A
.
C
.
F
a
u
z
i
,
N
.
A
h
ma
d
,
a
n
d
M
.
S
.
M
.
A
r
i
f
i
n
,
“
A
n
a
u
t
o
ma
t
e
f
a
i
l
u
r
e
r
e
c
o
v
e
r
y
f
o
r
s
y
n
c
h
r
o
n
o
u
s
d
i
s
t
r
i
b
u
t
e
d
d
a
t
a
b
a
se
s
y
s
t
e
m
,
”
I
n
d
o
n
e
s
i
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
8
,
n
o
.
2
,
p
p
.
9
7
3
–
9
7
9
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
2
8
.
i
2
.
p
p
9
7
3
-
9
7
9
.
[
6
]
D
.
K
.
Y
a
d
a
v
,
A
.
K
a
u
s
h
i
k
,
a
n
d
N
.
Y
a
d
a
v
,
“
P
r
e
d
i
c
t
i
n
g
ma
c
h
i
n
e
f
a
i
l
u
r
e
s
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,
”
S
u
s
t
a
i
n
a
b
l
e
Ma
n
u
f
a
c
t
u
r
i
n
g
a
n
d
S
e
rv
i
c
e
E
c
o
n
o
m
i
c
s
,
v
o
l
.
3
,
p
.
1
0
0
0
2
9
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
sms
e
.
2
0
2
4
.
1
0
0
0
2
9
.
[
7
]
F
.
Ta
n
g
,
J.
P
e
n
g
,
P
.
W
a
n
g
,
H
.
Z
h
u
,
a
n
d
T.
X
u
,
“
I
mp
r
o
v
e
d
d
y
n
a
m
i
c
B
y
z
a
n
t
i
n
e
f
a
u
l
t
t
o
l
e
r
a
n
t
c
o
n
se
n
s
u
s
me
c
h
a
n
i
sm
,
”
C
o
m
p
u
t
e
r
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
2
2
6
–
2
2
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mc
o
m.
2
0
2
4
.
0
8
.
0
0
4
.
[
8
]
Z.
Y
a
z
d
a
n
p
a
r
a
st
,
“
A
s
u
r
v
e
y
o
n
se
l
f
-
h
e
a
l
i
n
g
s
o
f
t
w
a
r
e
s
y
st
e
m,
”
a
rX
i
v
p
re
p
r
i
n
t
a
rX
i
v
:
2
4
0
3
.
0
0
4
5
5
,
2
0
2
4
.
[
9
]
Y
.
Z
h
u
,
K
.
C
.
H
o
w
,
H
.
J.
W
u
,
a
n
d
Q
.
C
a
o
,
“
A
I
-
b
a
sed
p
r
o
a
c
t
i
v
e
st
o
r
a
g
e
f
a
i
l
u
r
e
m
a
n
a
g
e
me
n
t
i
n
s
o
f
t
w
a
r
e
-
d
e
f
i
n
e
d
d
a
t
a
c
e
n
t
r
e
s
,
”
i
n
AC
M
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
Pr
o
c
e
e
d
i
n
g
S
e
r
i
e
s
,
2
0
2
3
,
p
p
.
2
3
1
–
2
3
7
,
d
o
i
:
1
0
.
1
1
4
5
/
3
6
2
5
1
5
6
.
3
6
2
5
1
9
0
.
[
1
0
]
O
r
a
c
l
e
,
“
Tr
a
n
sf
o
r
m
i
n
g
r
e
a
l
-
t
i
m
e
p
a
y
men
t
s
w
i
t
h
o
r
a
c
l
e
’
s
d
i
st
r
i
b
u
t
e
d
d
a
t
a
b
a
se:
so
v
e
r
e
i
g
n
t
y
,
s
c
a
l
a
b
i
l
i
t
y
,
a
n
d
s
u
r
v
i
v
a
b
i
l
i
t
y
,
”
O
r
a
c
l
e
2
0
2
5
.
[
1
1
]
S
i
v
a
P
r
a
s
a
d
N
a
n
d
i
,
“
A
I
-
p
o
w
e
r
e
d
a
u
t
o
n
o
my
:
Th
e
e
v
o
l
u
t
i
o
n
a
n
d
f
u
t
u
r
e
o
f
i
n
t
e
l
l
i
g
e
n
t
d
a
t
a
b
a
s
e
m
a
n
a
g
e
me
n
t
sy
st
e
ms,
”
Wo
r
l
d
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
E
n
g
i
n
e
e
r
i
n
g
T
e
c
h
n
o
l
o
g
y
a
n
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
5
,
n
o
.
2
,
2
0
2
5
,
d
o
i
:
1
0
.
3
0
5
7
4
/
w
j
a
e
t
s.
2
0
2
5
.
1
5
.
2
.
0
5
2
4
.
[
1
2
]
N
.
N
.
Jh
a
a
n
d
P
.
M
a
n
w
a
n
i
,
“
S
e
l
f
-
h
e
a
l
i
n
g
p
a
y
m
e
n
t
s
y
s
t
e
ms
v
i
a
A
I
-
d
r
i
v
e
n
a
n
o
m
a
l
y
r
e
c
o
v
e
r
y
:
a
z
e
r
o
-
d
o
w
n
t
i
me
f
r
a
m
e
w
o
r
k
f
o
r
sec
u
r
e
a
n
d
r
e
l
i
a
b
l
e
t
r
a
n
s
a
c
t
i
o
n
s
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
6
,
n
o
.
2
,
p
p
.
2
0
0
–
2
1
2
,
2
0
2
5
,
d
o
i
:
1
0
.
3
4
2
1
8
/
I
JC
ET
_
1
6
_
0
2
_
0
1
4
.
[
1
3
]
A
.
K
.
M
o
k
a
a
n
d
S
.
K
.
La
g
i
set
t
y
,
“
R
e
a
l
-
t
i
me
a
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
:
a
h
y
b
r
i
d
a
p
p
r
o
a
c
h
c
o
m
b
i
n
i
n
g
s
t
r
e
a
m
i
n
g
S
Q
L
a
n
d
h
a
r
d
w
a
r
e
a
c
c
e
l
e
r
a
t
i
o
n
,
”
i
n
2
0
2
5
5
t
h
I
n
t
e
l
l
i
g
e
n
t
C
y
b
e
rs
e
c
u
ri
t
y
C
o
n
f
e
r
e
n
c
e
,
2
0
2
5
,
p
p
.
3
8
5
–
3
9
2
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
C
6
5
5
9
6
.
2
0
2
5
.
1
1
1
4
0
0
8
5
.
[
1
4
]
Q
.
B
a
o
,
J.
W
a
n
g
,
H
.
G
o
n
g
,
Y
.
Zh
a
n
g
,
X
.
G
u
o
,
a
n
d
H
.
F
e
n
g
,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
t
o
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
i
n
h
i
g
h
-
f
r
e
q
u
e
n
c
y
t
r
a
d
i
n
g
d
a
t
a
,
”
2
0
2
5
4
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
S
y
m
p
o
s
i
u
m
o
n
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s
a
n
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
I
S
C
AI
T
2
0
2
5
.
p
p
.
2
8
7
–
2
9
1
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
S
C
A
I
T6
4
9
1
6
.
2
0
2
5
.
1
1
0
1
0
6
8
0
.
[
1
5
]
B
.
W
.
W
u
b
e
t
e
,
B
.
Esf
a
n
d
i
a
r
i
,
a
n
d
T
.
K
u
n
z
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
f
o
r
p
r
e
d
i
c
t
i
n
g
l
i
n
k
f
a
i
l
u
r
e
s
i
n
p
r
o
d
u
c
t
i
o
n
n
e
t
w
o
r
k
s
,
”
C
o
m
p
u
t
e
r
N
e
t
w
o
r
k
s
,
v
o
l
.
2
5
9
,
p
.
1
1
1
0
9
8
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mn
e
t
.
2
0
2
5
.
1
1
1
0
9
8
.
[
1
6
]
S
.
L
i
u
,
“
A
su
r
v
e
y
o
n
f
a
u
l
t
-
t
o
l
e
r
a
n
c
e
i
n
d
i
s
t
r
i
b
u
t
e
d
o
p
t
i
m
i
z
a
t
i
o
n
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
a
rXi
v
p
r
e
p
r
i
n
t
a
r
Xi
v
:
2
1
0
6
.
0
8
5
4
5
,
2
0
2
1
.
[
1
7
]
W
.
Zh
o
n
g
e
t
a
l
.
,
“
B
y
z
a
n
t
i
n
e
f
a
u
l
t
-
t
o
l
e
r
a
n
t
c
o
n
se
n
su
s
a
l
g
o
r
i
t
h
ms
:
A
s
u
r
v
e
y
,
”
E
l
e
c
t
r
o
n
i
c
s
(
S
w
i
t
z
e
rl
a
n
d
)
,
v
o
l
.
1
2
,
n
o
.
1
8
,
p
.
3
8
0
1
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
2
1
8
3
8
0
1
.
[
1
8
]
H
.
A
.
M
a
h
mo
u
d
a
n
d
H
.
M
a
s
e
e
h
Y
a
s
i
n
,
“
D
a
t
a
i
n
t
e
g
r
i
t
y
a
n
d
c
o
n
s
i
st
e
n
c
y
c
h
a
l
l
e
n
g
e
s
i
n
d
i
s
t
r
i
b
u
t
e
d
d
a
t
a
b
a
s
e
s
y
st
e
ms,”
E
n
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
J
o
u
r
n
a
l
,
v
o
l
.
1
0
,
n
o
.
0
5
,
p
p
.
5
0
7
7
–
5
0
8
6
,
2
0
2
5
,
d
o
i
:
1
0
.
4
7
1
9
1
/
e
t
j
/
v
1
0
i
0
5
.
3
6
.
[
1
9
]
R
.
Ta
f
t
e
t
a
l
.
,
“
C
o
c
k
r
o
a
c
h
D
B
:
T
h
e
r
e
si
l
i
e
n
t
g
e
o
-
d
i
st
r
i
b
u
t
e
d
S
Q
L
d
a
t
a
b
a
se
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
A
C
M
S
I
G
M
O
D
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
a
n
a
g
e
m
e
n
t
o
f
D
a
t
a
,
2
0
2
0
,
p
p
.
1
4
9
3
–
1
5
0
9
,
d
o
i
:
1
0
.
1
1
4
5
/
3
3
1
8
4
6
4
.
3
3
8
6
1
3
4
.
[
2
0
]
S
.
C
.
R
a
j
e
s
h
a
n
d
R
.
K
u
m
a
r
,
“
H
i
g
h
a
v
a
i
l
a
b
i
l
i
t
y
st
r
a
t
e
g
i
e
s
i
n
d
i
s
t
r
i
b
u
t
e
d
s
y
st
e
ms:
A
p
r
a
c
t
i
c
a
l
g
u
i
d
e
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
,
Re
f
e
rre
d
,
P
e
e
r
Re
v
i
e
w
e
d
&
I
n
d
e
x
e
d
M
o
n
t
h
l
y
J
o
u
rn
a
l
w
w
w
.
i
j
rsm
l
.
o
r
g
R
e
s
a
g
a
t
e
G
l
o
b
a
l
-
Ac
a
d
e
m
y
f
o
r
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
s
o
f
Mu
l
t
i
d
i
sc
i
p
l
i
n
a
r
y
Re
se
a
rc
h
,
v
o
l
.
1
.
p
p
.
1
1
0
–
1
3
0
,
2
0
2
5
.
[
2
1
]
B
.
A
n
b
a
l
a
g
a
n
,
“
P
r
o
a
c
t
i
v
e
f
a
i
l
o
v
e
r
a
n
d
a
u
t
o
m
a
t
i
o
n
f
r
a
m
e
w
o
r
k
s
f
o
r
m
i
ss
i
o
n
-
c
r
i
t
i
c
a
l
w
o
r
k
l
o
a
d
s
:
l
e
ss
o
n
s
f
r
o
m
ma
n
u
f
a
c
t
u
r
i
n
g
i
n
d
u
st
r
y
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Re
sea
r
c
h
a
n
d
A
p
p
l
i
e
d
I
n
n
o
v
a
t
i
o
n
s
,
v
o
l
.
0
6
,
n
o
.
0
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
5
6
6
2
/
i
j
r
a
i
.
2
0
2
3
.
0
6
0
1
0
0
4
.
[
2
2
]
Q
.
C
a
i
,
C
.
C
u
i
,
Y
.
X
i
o
n
g
,
W
.
W
a
n
g
,
Z.
X
i
e
,
a
n
d
M
.
Zh
a
n
g
,
“
A
s
u
r
v
e
y
o
n
d
e
e
p
r
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
f
o
r
d
a
t
a
p
r
o
c
e
ss
i
n
g
a
n
d
a
n
a
l
y
t
i
c
s,
”
I
EE
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
K
n
o
w
l
e
d
g
e
a
n
d
D
a
t
a
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
3
5
,
n
o
.
5
,
p
p
.
4
4
4
6
–
4
4
6
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
T
K
D
E.
2
0
2
2
.
3
1
5
5
1
9
6
.
[
2
3
]
A
.
V
i
k
i
r
u
,
M
.
M
u
i
r
u
r
i
,
a
n
d
I
.
A
t
e
y
a
,
“
A
n
o
v
e
r
v
i
e
w
o
n
c
l
o
u
d
d
i
st
r
i
b
u
t
e
d
d
a
t
a
b
a
se
s
f
o
r
b
u
s
i
n
e
ss
e
n
v
i
r
o
n
me
n
t
s,
”
a
rX
i
v
p
r
e
p
r
i
n
t
a
rXi
v
:
2
3
0
1
.
1
0
6
7
3
,
2
0
2
3
.
[
2
4
]
J.
Li
,
W
.
R
e
n
,
a
n
d
X
.
W
u
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
s
e
d
n
e
t
w
o
r
k
p
e
r
f
o
r
m
a
n
c
e
mo
n
i
t
o
r
i
n
g
a
n
d
p
r
e
d
i
c
t
i
o
n
f
o
r
d
i
st
r
i
b
u
t
e
d
A
I
t
r
a
i
n
i
n
g
w
o
r
k
l
o
a
d
s
,
”
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
i
n
g
S
y
st
e
m
s
,
v
o
l
.
5
,
n
o
.
1
0
.
p
p
.
1
–
1
7
,
2
0
2
5
,
d
o
i
:
1
0
.
6
9
9
8
7
/
j
a
c
s
.
2
0
2
5
.
5
1
0
0
1
.
[
2
5
]
V
.
S
u
b
r
a
ma
n
i
,
“
R
e
si
l
i
e
n
c
e
b
y
d
e
si
g
n
:
si
t
e
r
e
l
i
a
b
i
l
i
t
y
e
n
g
i
n
e
e
r
i
n
g
i
n
f
i
n
a
n
c
i
a
l
p
l
a
t
f
o
r
ms
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s i
n
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
3
,
n
o
.
2
s
,
p
p
.
8
7
–
9
5
,
2
0
2
5
.
[
2
6
]
F
.
G
o
m
e
s,
P
.
R
e
g
o
,
a
n
d
F
.
Tr
i
n
t
a
,
“
A
sy
s
t
e
m
a
t
i
c
ma
p
p
i
n
g
st
u
d
y
o
n
o
b
ser
v
a
b
i
l
i
t
y
o
f
m
i
c
r
o
s
e
r
v
i
c
e
s
-
b
a
s
e
d
a
p
p
l
i
c
a
t
i
o
n
s
:
f
u
n
d
a
m
e
n
t
a
l
s,
c
l
a
ssi
f
i
c
a
t
i
o
n
s,
a
n
d
c
h
a
l
l
e
n
g
e
s,
”
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
0
7
,
n
o
.
9
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s0
0
6
0
7
-
0
2
5
-
0
1
5
4
0
-
w.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ra
g
h
u
G
o
ll
a
p
u
d
i
is
a
S
e
n
io
r
Ora
c
le
Da
tab
a
se
Ad
m
in
istr
a
to
r
a
n
d
Da
tab
a
s
e
Re
li
a
b
il
it
y
En
g
in
e
e
r
a
t
F
ise
rv
,
wh
e
re
h
e
a
rc
h
it
e
c
ts
a
n
d
su
p
p
o
r
ts
m
issio
n
-
c
rit
ica
l
fin
a
n
c
ial
in
fra
stru
c
tu
re
fo
r
p
latfo
rm
s
su
c
h
a
s
Zelle
a
n
d
Ba
n
k
o
f
Am
e
rica
.
He
sp
e
c
ialize
s
in
re
li
a
b
il
it
y
e
n
g
in
e
e
rin
g
fo
r
e
n
terp
rise
d
a
tab
a
se
s,
with
a
stro
n
g
f
o
c
u
s
o
n
d
e
s
ig
n
i
n
g
se
lf
-
h
e
a
li
n
g
d
a
tab
a
se
fa
b
rics
a
n
d
AI
-
d
riv
e
n
p
re
d
ictiv
e
fa
il
u
re
m
o
d
e
ls
f
o
r
sy
ste
m
s
t
h
a
t
p
ro
c
e
ss
b
il
li
o
n
s
o
f
d
o
ll
a
rs
i
n
d
a
il
y
tran
sa
c
ti
o
n
s.
He
h
o
ld
s
a
m
a
ste
r’s
d
e
g
re
e
in
i
n
fo
rm
a
ti
o
n
tec
h
n
o
l
o
g
y
fr
o
m
Va
l
p
a
ra
iso
Un
iv
e
rsity
a
n
d
is
a
n
a
c
ti
v
e
re
se
a
rc
h
e
r
in
t
h
e
fi
e
l
d
o
f
a
u
to
n
o
m
o
u
s
d
a
tab
a
se
re
sili
e
n
c
e
.
His
re
se
a
rc
h
fo
c
u
se
s
o
n
t
h
e
i
n
ters
e
c
ti
o
n
o
f
tra
d
it
i
o
n
a
l
d
a
tab
a
se
c
o
n
s
isten
c
y
m
o
d
e
ls
a
n
d
m
o
d
e
rn
fin
a
n
c
ial
se
tt
lem
e
n
t
f
in
a
li
t
y
re
q
u
irem
e
n
ts,
c
o
n
tr
ib
u
ti
n
g
t
o
th
e
a
d
v
a
n
c
e
m
e
n
t
o
f
i
n
telli
g
e
n
t
a
n
d
re
sili
e
n
t
fin
a
n
c
ial
sy
ste
m
s.
Ra
g
h
u
m
a
in
tain
s
a
n
a
c
ti
v
e
a
c
a
d
e
m
ic
p
re
se
n
c
e
th
ro
u
g
h
h
is
ORCID
a
n
d
G
o
o
g
le
S
c
h
o
lar p
r
o
fil
e
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
re
a
c
h
ra
g
h
u
2
5
1
@
g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.