I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
d
van
c
e
s
i
n
A
p
p
li
e
d
S
c
ie
n
c
e
s
(
I
JA
A
S
)
V
ol
.
14
, N
o.
3
,
S
e
pt
e
m
be
r
20
25
, pp.
671
~
681
I
S
S
N
:
2252
-
8814
,
D
O
I
:
10.11591/
ij
a
a
s
.
v14.
i
3
.
pp
671
-
681
671
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
aas
.i
ae
s
c
or
e
.c
om
H
ayst
ac
k
-
b
ase
d
Fac
e
b
ook
’
s d
at
a st
or
age
ar
c
h
i
t
e
c
t
u
r
e
:
st
or
e
,
d
i
r
e
c
t
or
y
,
an
d
c
ac
h
e
T
ol
e
S
u
t
ik
n
o
1
,2
, A
h
m
ad
H
e
r
yan
t
o
3
,
4
,
L
ak
s
an
a T
al
e
n
t
a A
h
m
ad
2,
5
1
F
a
c
ul
t
y of
I
ndus
t
r
i
a
l
T
e
c
hnol
ogy, U
ni
ve
r
s
i
t
a
s
A
hm
a
d D
a
hl
a
n, Y
ogya
ka
r
t
a
, I
ndone
s
i
a
2
E
m
be
dde
d S
ys
t
e
m
s
a
nd P
ow
e
r
E
l
e
c
t
r
oni
c
s
R
e
s
e
a
r
c
h G
r
oup, Y
ogya
ka
r
t
a
, I
ndone
s
i
a
3
F
a
c
ul
t
y of
E
ngi
ne
e
r
i
ng
,
U
ni
ve
r
s
i
t
a
s
S
r
i
w
i
j
a
ya
, P
a
l
e
m
ba
ng, I
ndone
s
i
a
4
F
a
c
ul
t
y of
C
om
put
e
r
S
c
i
e
nc
e
, U
ni
ve
r
s
i
t
a
s
S
r
i
w
i
j
a
ya
, P
a
l
e
m
ba
ng, I
ndone
s
i
a
5
D
e
pa
r
t
m
e
nt
of
I
nf
or
m
a
t
i
on
S
ys
t
e
m
s
, I
nt
e
r
na
t
i
ona
l
I
s
l
a
m
i
c
U
ni
ve
r
s
i
t
y M
a
l
a
ys
i
a
, K
ua
l
a
L
um
pur
, M
a
l
a
y
s
i
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
O
c
t
21
,
2024
R
e
vi
s
e
d
M
a
y
8
,
2025
A
c
c
e
pt
e
d
J
un
4
,
2025
Haystack
is
Facebook'
s
unique
way
of
managing
large
amounts
of
user
-
generated
content
like
photos.
The
architecture
prioritizes
perfor
mance,
reliabilit
y,
and
scalabilit
y
to
overcome
network
-
attached
storage
system
bottlenecks.
Haystack
speeds
data
access
and
ensures
data
integrity
during
hardware
failures
by
using
physical
and
logical
volumes.
This
study
examines
the
architectu
re
of
Facebook'
s
Haystack
data
storage
syste
m
and
its
effects
on
scalability
and
efficiency
in
handling
large
photo
data.
According
to
the
study,
the
store,
directory,
and
cache
functions
work
together
to
reduce
input/output
(
I/O
)
operations
and
improve
m
etadata
processing,
which
traditional
network
-
attached
storage
systems
can
not
do.
Haystack
manages
massive
photo
data
storage
and
retrieva
l,
s
olving
network
-
attached
storage
(NAS)
limit
ations.
It
balances
throughp
ut
and
latency
by
minimizing
disk
operations
and
optimizing
metadata
proc
essing.
Each
store,
directory,
and
cache
contribu
te
to
this
ecosystem.
The
Ha
ystack
architectu
re
reduces
disk
operation
s
and
metadata
processin
g
bottl
enecks
with
distributed
caching.
A
cache
allows
instant
access
to
frequently
requested
images
and
balances
read
and
write
operations
across
the
s
ystem.
We
should
study
adv
anced
storage
system
architectures
based
on
Face
book'
s
Haystack
architectu
re.
This
could
involv
e
investi
gating
faster
m
etadata
processing
algorithms,
using
artificial
intelligence
(AI)
to
improv
e
fault
detection
and
repair
systems,
and
assessing
the
economic
imp
act
of
distr
ibuted cac
hes.
K
e
y
w
o
r
d
s
:
A
r
c
hi
te
c
tu
r
e
C
a
c
he
D
a
ta
s
to
r
a
ge
D
ir
e
c
to
r
y
F
a
c
e
book
H
a
ys
ta
c
k
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
T
ol
e
S
ut
ik
no
F
a
c
ul
ty
of
I
ndus
tr
ia
l
T
e
c
hnol
ogy, Unive
r
s
it
a
s
A
hm
a
d
D
a
hl
a
n
A
hm
a
d Y
a
ni
S
tr
e
e
t
(
S
out
h of
R
in
g R
oa
d)
, T
a
m
a
na
n, Y
ogya
ka
r
t
a
55191, I
ndone
s
ia
E
m
a
il
:
to
le
@
te
.ua
d.a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
I
n t
he
f
a
s
t
-
c
ha
ngi
ng w
or
ld
o
f
da
ta
s
to
r
a
ge
t
e
c
hnol
ogy, ha
vi
ng s
ys
te
m
s
t
ha
t
c
a
n ha
ndl
e
l
a
r
ge
a
m
ount
s
of
in
f
or
m
a
ti
on
is
ve
r
y
im
por
ta
nt
.
F
a
c
e
book'
s
uni
que
w
a
y
of
ta
c
kl
in
g
th
is
pr
obl
e
m
is
s
how
n
th
r
ough
it
s
H
a
ys
ta
c
k
d
e
s
ig
n,
m
e
a
nt
to
m
a
na
ge
th
e
huge
s
c
a
l
e
of
us
e
r
-
c
r
e
a
te
d
c
ont
e
nt
li
ke
phot
os
.
T
hi
s
de
s
ig
n
s
ki
ll
f
ul
ly
a
voi
ds
c
om
m
on pr
obl
e
m
s
f
ound
i
n ne
twor
k
-
a
tt
a
c
he
d s
to
r
a
ge
s
ys
te
m
s
by putt
in
g t
oge
th
e
r
a
s
tr
ong s
ys
te
m
t
ha
t
f
oc
us
e
s
on
s
pe
e
d,
de
pe
nd
a
bi
li
ty
,
a
nd
gr
ow
th
.
B
y
us
in
g
bot
h
p
hys
ic
a
l
a
nd
lo
gi
c
a
l
vol
um
e
s
,
H
a
ys
ta
c
k
boos
t
s
da
ta
a
c
c
e
s
s
s
pe
e
d
a
nd
a
dds
im
por
ta
nt
r
e
dunda
nc
y
m
e
a
s
ur
e
s
to
ke
e
p
da
ta
s
a
f
e
dur
in
g
ha
r
dw
a
r
e
is
s
ue
s
.
T
he
a
r
c
hi
te
c
tu
r
e
ha
s
th
r
e
e
m
a
in
pa
r
ts
:
th
e
s
to
r
e
,
di
r
e
c
to
r
y,
a
nd
c
a
c
he
.
E
a
c
h
pa
r
t
is
c
r
uc
ia
l
f
or
opt
im
iz
in
g
phot
o
r
e
tr
ie
va
l,
a
ll
ow
in
g f
o
r
qui
c
k
r
e
a
d a
nd w
r
it
e
a
c
ti
ons
. T
he
s
to
r
e
ha
ndl
e
s
t
he
e
f
f
e
c
ti
ve
m
a
na
ge
m
e
nt
of
r
a
w
phot
o
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
3
,
S
e
pt
e
m
be
r
20
25
:
671
-
681
672
in
f
or
m
a
ti
on,
w
hi
le
th
e
di
r
e
c
to
r
y
m
a
ke
s
f
a
s
t
lo
okups
a
nd
or
ga
ni
z
e
s
th
e
s
to
r
e
d
pi
c
tu
r
e
s
.
A
t
th
e
s
a
m
e
ti
m
e
,
th
e
c
a
c
he
a
c
ts
a
s
a
s
p
e
e
dy
li
nk,
gr
e
a
tl
y
c
ut
ti
ng
dow
n
r
e
s
pon
s
e
ti
m
e
a
nd
e
nha
nc
in
g
u
s
e
r
e
xpe
r
ie
nc
e
.
T
he
c
om
bi
ne
d w
or
k of
t
he
s
e
pa
r
ts
c
ha
nge
s
t
he
w
a
y da
ta
s
to
r
a
ge
w
o
r
ks
, e
ns
ur
in
g F
a
c
e
book c
a
n m
a
na
ge
bi
ll
io
ns
of
phot
os
r
e
li
a
bl
y a
nd e
a
s
il
y. L
ooki
ng i
nt
o t
he
de
ta
il
s
of
H
a
ys
ta
c
k'
s
a
r
c
hi
te
c
tu
r
e
s
how
s
not
j
us
t
it
s
t
e
c
hni
c
a
l
s
ki
ll
but
a
ls
o
it
s
im
por
ta
nt
im
pa
c
t
on
th
e
f
ut
u
r
e
of
di
s
tr
ib
u
te
d
s
ys
te
m
s
.
A
s
th
e
di
gi
ta
l
w
or
ld
ke
e
ps
gr
ow
in
g,
knowing
how
H
a
ys
ta
c
k
is
de
s
ig
ne
d
pr
ovi
de
s
u
s
e
f
ul
in
s
ig
ht
s
f
or
c
r
e
a
ti
ng
s
c
a
la
bl
e
s
ol
ut
io
ns
in
m
a
ny
a
r
e
a
s
[
1]
–
[
8]
.
T
hi
s
s
tu
dy
w
il
l
br
e
a
k
dow
n
th
e
e
s
s
e
nt
ia
l
f
e
a
tu
r
e
s
of
th
e
a
r
c
hi
te
c
tu
r
e
,
lo
oki
ng
a
t
how
e
a
c
h
pa
r
t
he
lp
s
bui
ld
a
s
tr
ong
a
nd
e
f
f
e
c
ti
ve
s
ys
te
m
f
or
ha
ndl
in
g
la
r
ge
da
ta
a
m
ount
s
,
ul
ti
m
a
te
ly
s
e
r
vi
ng
a
s
a
m
ode
l
f
or
f
ut
ur
e
a
dva
nc
e
m
e
nt
s
i
n da
ta
s
to
r
a
ge
t
e
c
hnol
ogy.
H
a
ndl
in
g
a
la
r
ge
a
m
ount
of
c
ont
e
nt
m
a
de
by
us
e
r
s
r
e
qui
r
e
s
a
s
tr
ong
a
nd
s
m
a
r
t
w
a
y
to
s
to
r
e
da
ta
f
o
r
F
a
c
e
book.
W
it
h
m
or
e
th
a
n
260
bi
ll
io
n
phot
o
s
upl
oa
d
e
d
by
u
s
e
r
s
,
ha
vi
ng s
to
r
a
ge
th
a
t
c
a
n
gr
ow
a
nd w
or
k
w
e
ll
is
ve
r
y
im
por
ta
nt
.
O
ld
ne
twor
k
-
a
tt
a
c
he
d
s
to
r
a
ge
(
N
A
S
)
s
ys
te
m
s
w
e
r
e
not
e
nough
be
c
a
us
e
th
e
y
ha
d
i
s
s
ue
s
w
it
h
di
s
k
ope
r
a
ti
ons
a
nd
pr
oc
e
s
s
in
g
m
e
ta
da
ta
,
le
a
di
ng
F
a
c
e
b
ook
to
lo
ok
f
or
ne
w
m
e
th
ods
.
T
he
H
a
ys
ta
c
k
s
ys
te
m
c
a
m
e
a
s
a
s
ol
ut
io
n,
or
ga
ni
z
in
g
s
to
r
a
ge
in
to
m
a
ny
phy
s
ic
a
l
vol
um
e
s
,
e
a
c
h
a
bl
e
to
hol
d
m
il
li
ons
of
phot
os
[
9]
.
T
hi
s
s
e
tu
p
is
bui
lt
to
c
opy
da
ta
a
c
r
os
s
di
f
f
e
r
e
nt
s
to
r
a
ge
a
r
e
a
s
to
a
voi
d
lo
s
s
a
nd
e
ns
ur
e
r
e
li
a
bi
li
ty
.
A
ls
o,
ongoing
ba
c
kgr
ound
ta
s
ks
ke
e
p
c
he
c
ki
ng
a
nd
f
ix
in
g
pos
s
ib
le
pr
obl
e
m
s
,
pr
ot
e
c
ti
ng
us
e
r
da
ta
f
r
om
ha
r
dw
a
r
e
is
s
ue
s
[
10]
.
T
hus
,
F
a
c
e
book’
s
s
to
r
a
ge
ne
e
d
s
r
e
qui
r
e
a
ve
r
y
e
f
f
e
c
ti
ve
,
s
c
a
la
bl
e
,
a
nd
f
a
ul
t
-
r
e
s
is
ta
nt
s
ys
te
m
t
o ha
ndl
e
t
he
hug
e
a
m
ount
s
of
i
nf
or
m
a
ti
on c
r
e
a
te
d by it
s
w
or
ld
w
id
e
us
e
r
s
.
A
good
da
ta
s
to
r
a
ge
s
e
tu
p
is
c
r
uc
ia
l
f
o
r
boos
ti
ng
pe
r
f
or
m
a
nc
e
a
nd
e
ns
ur
in
g
r
e
li
a
bi
li
ty
w
he
n
m
a
na
gi
ng
la
r
ge
a
m
ount
s
of
da
ta
.
F
or
e
xa
m
pl
e
,
F
a
c
e
book'
s
H
a
ys
ta
c
k
s
y
s
te
m
w
a
s
m
a
de
to
ha
ndl
e
m
or
e
th
a
n
260
bi
ll
io
n
phot
os
.
A
w
e
ll
-
o
r
ga
ni
z
e
d
s
e
tu
p
r
e
duc
e
s
th
e
ne
e
d
f
or
di
s
k
ope
r
a
ti
ons
a
nd
m
e
ta
da
ta
pr
oc
e
s
s
in
g,
w
hi
c
h i
m
pr
ove
s
s
pe
e
d a
nd c
ut
s
dow
n on de
la
y. I
n di
s
tr
ib
ut
e
d s
ys
te
m
s
, t
hi
s
ki
nd of
s
e
tu
p pr
om
ot
e
s
s
c
a
la
bi
li
ty
,
he
lp
in
g
or
ga
ni
z
a
ti
ons
gr
ow
th
e
ir
s
to
r
a
ge
w
it
hout
m
a
ny
p
e
r
f
or
m
a
nc
e
is
s
ue
s
.
H
a
y
s
ta
c
k
s
how
s
how
us
in
g a
m
ix
of
s
to
r
a
ge
,
di
r
e
c
to
r
y,
a
nd
c
a
c
he
e
le
m
e
nt
s
c
a
n
e
nha
nc
e
f
a
ul
t
to
l
e
r
a
nc
e
a
nd
s
a
ve
c
o
s
ts
[
9]
.
S
uc
h
a
dv
a
nc
e
m
e
nt
s
s
how
th
a
t
s
m
a
r
t
s
to
r
a
ge
de
s
ig
n
not
onl
y
r
e
duc
e
s
r
is
ks
r
e
la
te
d
to
ha
r
dw
a
r
e
is
s
ue
s
but
a
ls
o
m
e
e
ts
th
e
gr
ow
in
g
de
m
a
nd f
or
qui
c
k da
ta
a
c
c
e
s
s
i
n a
w
or
ld
w
he
r
e
i
nf
or
m
a
ti
on i
s
e
xt
e
ns
iv
e
a
nd i
m
por
ta
nt
[
11]
.
T
he
H
a
ys
ta
c
k
s
ys
te
m
is
a
n
im
por
ta
nt
s
te
p
f
or
w
a
r
d
in
how
la
r
ge
da
ta
is
m
a
na
ge
d
a
nd
s
a
ve
d,
e
s
pe
c
ia
ll
y
m
a
d
e
to
m
e
e
t
F
a
c
e
book'
s
huge
phot
o
s
to
r
a
ge
ne
e
ds
.
A
t
it
s
c
or
e
,
th
e
s
y
s
te
m
ha
s
th
r
e
e
m
a
in
pa
r
ts
:
th
e
H
a
ys
ta
c
k
s
to
r
e
,
di
r
e
c
to
r
y,
a
nd
c
a
c
he
.
T
hi
s
th
r
e
e
-
pa
r
t
de
s
ig
n
he
lp
s
m
a
ke
da
ta
r
e
tr
ie
va
l
a
nd
s
to
r
a
ge
s
m
oot
he
r
,
gr
e
a
tl
y
im
pr
ovi
ng
ope
r
a
ti
ons
by
lo
w
e
r
in
g
di
s
k
in
put
/
out
put
(
I
/O
)
a
nd
boos
ti
ng
f
a
ul
t
to
le
r
a
nc
e
w
it
h
r
e
dunda
nc
y
a
nd
ongoing
ba
c
kgr
ound
ta
s
ks
.
B
y
s
or
ti
ng
da
ta
in
to
lo
gi
c
a
l
a
nd
phys
ic
a
l
vol
um
e
s
th
a
t
pr
ovi
de
phot
o ba
c
kup,
H
a
ys
ta
c
k
s
ol
ve
s
i
m
por
ta
nt
pr
obl
e
m
s
r
e
ga
r
di
ng ha
r
dw
a
r
e
br
e
a
kdown a
nd da
ta
l
os
s
, w
hi
c
h l
e
a
ds
to
be
tt
e
r
r
e
li
a
bi
li
ty
a
nd
pe
r
f
o
r
m
a
nc
e
[
9]
.
A
ls
o,
th
e
s
ys
te
m
us
e
s
a
di
s
tr
ib
ut
e
d
c
a
c
he
to
c
ut
dow
n
on
w
a
it
ti
m
e
s
w
hi
le
ke
e
pi
ng
hi
gh
out
put
,
th
us
c
r
e
a
ti
ng
a
s
tr
ong
s
e
tu
p
th
a
t
ha
ndl
e
s
c
ur
r
e
nt
ne
e
ds
a
nd
is
r
e
a
dy
f
or
f
ut
ur
e
e
xpa
ns
io
n
[
12]
.
T
he
m
a
in
goa
ls
of
th
is
r
e
s
e
a
r
c
h
pa
pe
r
a
r
e
to
a
n
a
ly
z
e
th
e
de
s
i
gn
f
e
a
tu
r
e
s
of
F
a
c
e
book’
s
H
a
ys
t
a
c
k
da
ta
s
to
r
a
ge
s
ys
te
m
a
nd
to
e
xpl
a
in
how
th
e
y
a
f
f
e
c
t
s
c
a
la
bi
li
ty
a
nd
e
f
f
ic
ie
nc
y
w
he
n
de
a
li
ng
w
it
h
la
r
ge
a
m
ount
s
of
phot
o
da
ta
.
B
y
lo
oki
ng
c
lo
s
e
ly
a
t
th
e
s
to
r
e
,
di
r
e
c
to
r
y,
a
nd
c
a
c
he
f
unc
ti
ons
,
th
is
pa
pe
r
a
im
s
to
s
how
how
th
e
s
e
c
om
pone
nt
s
w
or
k
to
ge
th
e
r
to
r
e
duc
e
I
/O
ope
r
a
ti
ons
a
nd
im
pr
ove
m
e
ta
da
ta
pr
oc
e
s
s
in
g,
th
us
ta
c
kl
in
g
is
s
u
e
s
s
e
e
n
w
it
h
ol
de
r
n
e
twor
k
-
a
tt
a
c
he
d
s
to
r
a
ge
s
y
s
t
e
m
s
.
M
or
e
ov
e
r
,
th
e
r
e
s
e
a
r
c
h
w
il
l
di
s
c
u
s
s
th
e
e
f
f
e
c
ts
of
us
in
g
s
ta
nda
r
d
ha
r
dw
a
r
e
a
nd
r
e
dunda
nc
y
m
e
th
ods
on
s
ys
te
m
r
e
li
a
bi
li
ty
a
nd
f
a
ul
t
to
le
r
a
nc
e
[
9]
.
T
hr
ough c
a
s
e
s
tu
di
e
s
a
nd m
e
a
s
ur
a
bl
e
p
e
r
f
or
m
a
nc
e
m
e
tr
ic
s
, t
hi
s
pa
pe
r
i
nt
e
nds
t
o a
dd t
o t
he
c
ur
r
e
nt
knowle
dge
o
f
di
s
tr
ib
ut
e
d
s
ys
t
e
m
s
d
e
s
ig
n a
nd
pr
ovi
de
u
s
e
f
ul
gui
da
nc
e
f
or
d
e
ve
lo
pe
r
s
a
nd
a
r
c
hi
te
c
t
s
w
ho
w
a
nt
to
im
pr
ove
la
r
ge
-
s
c
a
le
da
ta
m
a
na
ge
m
e
nt
s
tr
a
te
gi
e
s
[
13]
.
I
n
th
e
e
nd,
th
e
r
e
s
e
a
r
c
h
hop
e
s
to
de
e
pe
n
th
e
und
e
r
s
ta
ndi
ng
of
s
c
a
la
bl
e
a
r
c
hi
te
c
tu
r
e
s
in
m
od
e
r
n
s
of
twa
r
e
s
e
tt
in
gs
.
I
t
is
v
e
r
y
im
por
ta
nt
to
e
xpl
a
in
H
a
ys
ta
c
k
'
s
da
ta
s
to
r
a
ge
s
tr
uc
tu
r
e
a
t
F
a
c
e
book.
T
he
in
tr
oduc
ti
on
gi
ve
s
a
n
ov
e
r
vi
e
w
of
w
hy
s
c
a
la
bl
e
s
to
r
a
ge
s
ys
t
e
m
s
m
a
tt
e
r
in
to
da
y'
s
w
or
ld
,
e
s
pe
c
ia
ll
y
f
or
c
ont
e
nt
-
he
a
vy
s
it
e
s
li
ke
F
a
c
e
book.
N
e
xt
,
it
lo
oks
c
lo
s
e
ly
a
t
th
e
m
a
in
pa
r
ts
s
to
r
e
,
di
r
e
c
to
r
y,
a
nd
c
a
c
he
-
e
xpl
a
in
in
g
how
e
a
c
h
pa
r
t
w
or
ks
a
nd
i
nt
e
r
a
c
ts
to
e
ns
ur
e
qui
c
k
phot
o
r
e
tr
ie
va
l
a
nd
r
e
li
a
bi
li
ty
.
T
he
c
onc
lu
s
io
n
br
in
gs
to
ge
th
e
r
th
e
s
e
p
a
r
ts
,
f
oc
us
i
ng
on
w
ha
t
H
a
ys
ta
c
k
s
m
e
a
ns
f
or
f
ut
ur
e
da
ta
s
to
r
a
ge
m
e
th
ods
a
nd
im
pr
ove
m
e
nt
s
in
di
s
tr
ib
ut
e
d
s
ys
t
e
m
de
s
ig
n.
B
y
c
le
a
r
ly
m
ovi
ng
f
r
om
ba
s
ic
id
e
a
s
to
m
or
e
c
om
pl
e
x
us
e
s
,
th
e
s
tu
dy
not
onl
y
s
ho
w
s
F
a
c
e
book'
s
in
nova
ti
ve
m
e
th
od
but
a
ls
o
pr
ovi
de
s
a
m
ode
l
f
or
c
r
e
a
ti
ng
s
tr
ong s
to
r
a
ge
s
ys
te
m
s
, a
s
poi
nt
e
d out i
n
[
9]
a
nd
[
11]
.
2.
U
N
D
E
R
S
T
A
N
D
I
N
G
H
A
Y
S
T
A
C
K
A
R
C
H
I
T
E
C
T
U
R
E
F
a
c
e
book'
s
H
a
ys
ta
c
k
s
y
s
te
m
ha
s
a
de
s
ig
n
th
a
t
m
a
na
ge
s
a
lo
t
of
phot
o
da
ta
s
to
r
a
ge
a
nd
r
e
tr
ie
va
l.
I
t
s
ol
ve
s
is
s
ue
s
s
e
e
n
in
s
ta
nda
r
d
N
A
S
s
ys
te
m
s
[
8]
,
[
14]
,
[
15]
a
s
s
how
n
in
F
ig
ur
e
1.
B
y
r
e
duc
in
g
di
s
k
ope
r
a
ti
ons
a
nd
im
pr
ovi
ng
m
e
ta
da
ta
ta
s
k
s
,
H
a
y
s
ta
c
k
f
in
ds
a
good
m
ix
of
hi
gh
s
pe
e
d
a
nd
lo
w
w
a
it
ti
m
e
,
w
hi
c
h
is
im
por
ta
nt
f
or
s
e
r
vi
c
e
s
w
it
h
ove
r
260
bi
ll
io
n
im
a
ge
s
[
9]
.
E
a
c
h
pa
r
t
,
s
to
r
e
,
di
r
e
c
to
r
y,
a
nd
c
a
c
he
h
a
s
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
dv A
ppl
S
c
i
I
S
S
N
:
2252
-
8814
H
ay
s
ta
c
k
-
bas
e
d F
ac
e
book
’
s
dat
a s
to
r
age
ar
c
hi
te
c
tu
r
e
:
s
to
r
e
, di
r
e
c
to
r
y
, and
c
ac
he
(
T
ol
e
Sut
ik
no)
673
s
pe
c
if
ic
f
unc
ti
on
in
th
i
s
s
y
s
te
m
.
T
he
H
a
ys
ta
c
k
s
to
r
e
a
r
r
a
nge
s
p
hys
ic
a
l
vol
um
e
s
in
to
lo
gi
c
a
l
s
tr
uc
tu
r
e
s
,
w
hi
c
h
he
lp
s
w
it
h
da
ta
ba
c
kup
a
nd
im
pr
ove
s
r
e
li
a
bi
li
ty
by
u
s
in
g
r
e
gul
a
r
ha
r
dw
a
r
e
,
a
s
s
how
n
in
F
ig
ur
e
2.
T
hi
s
m
e
th
od
he
lp
s
r
e
duc
e
pr
obl
e
m
s
f
r
om
ha
r
dw
a
r
e
f
a
il
ur
e
s
a
nd
is
s
uppor
te
d
by
ba
c
kgr
ound
ta
s
k
s
th
a
t
f
in
d
a
nd
f
ix
pr
ob
le
m
s
.
A
ddi
ti
ona
ll
y,
r
e
c
e
nt
s
tu
di
e
s
s
how
th
a
t
knowing
how
m
e
m
or
y
a
c
c
e
s
s
c
os
t
s
r
e
la
te
to
c
om
put
in
g
e
f
f
ic
ie
nc
y
is
im
por
ta
nt
;
us
in
g
to
ol
s
li
ke
H
a
ys
ta
c
k
he
lp
s
de
ve
lo
pe
r
s
le
a
r
n
how
to
im
pr
ove
th
e
ir
c
ode
w
it
h
c
a
c
h
e
s
y
s
te
m
s
,
le
a
di
ng
to
a
be
tt
e
r
s
to
r
a
ge
s
e
tu
p
ov
e
r
a
ll
[
16]
.
T
a
bl
e
1
s
how
s
th
e
N
A
S
v
s
H
a
y
s
ta
c
k
c
om
pa
r
is
on t
a
bl
e
.
F
ig
ur
e
1. N
A
S
ove
r
ne
twor
k f
il
e
s
ys
te
m
(
N
F
S
)
-
ba
s
e
d
s
to
r
a
ge
a
r
c
hi
te
c
tu
r
e
[
9]
F
ig
ur
e
2. H
a
ys
ta
c
k s
to
r
a
ge
a
r
c
hi
te
c
tu
r
e
[
9]
T
a
bl
e
1.
C
om
pa
r
is
on
t
a
bl
e
of
N
A
S
vs
.
H
a
y
s
ta
c
k
A
s
pe
c
t
NAS
H
a
ys
t
a
c
k
P
ur
pos
e
of
de
s
i
gn
G
e
ne
r
a
l
pur
pos
e
f
or
va
r
i
ous
t
ype
s
of
da
t
a
a
nd
a
ppl
i
c
a
t
i
ons
.
S
pe
c
i
f
i
c
a
l
l
y
de
s
i
gne
d
t
o
s
t
or
e
a
nd
m
a
na
ge
us
e
r
phot
os
e
f
f
i
c
i
e
nt
l
y.
A
r
c
hi
t
e
c
t
ur
e
F
i
l
e
-
ba
s
e
d;
us
e
s
pr
ot
oc
ol
s
l
i
ke
N
F
S
or
s
e
r
ve
r
m
e
s
s
a
ge
bl
o
c
k (
S
M
B
)
f
or
da
t
a
a
c
c
e
s
s
.
O
bj
e
c
t
-
ba
s
e
d;
us
e
s
a
uni
que
a
ppr
oa
c
h
t
o
s
t
or
e
m
e
t
a
da
t
a
a
nd da
t
a
t
oge
t
he
r
.
P
e
r
f
or
m
a
nc
e
H
i
ghe
r
l
a
t
e
nc
y f
or
f
r
e
que
nt
l
y a
c
c
e
s
s
e
d da
t
a
.
L
ow
l
a
t
e
nc
y
f
or
f
a
s
t
a
c
c
e
s
s
t
o
da
t
a
(
phot
os
)
w
i
t
h que
r
y opt
i
m
i
z
a
t
i
on.
S
t
or
a
ge
e
f
f
i
c
i
e
nc
y
M
e
t
a
da
t
a
a
nd
da
t
a
a
r
e
s
t
or
e
d
s
e
pa
r
a
t
e
l
y,
i
nc
r
e
a
s
i
ng m
e
t
a
da
t
a
ove
r
he
a
d.
M
e
t
a
da
t
a
a
nd
da
t
a
a
r
e
s
t
or
e
d
i
n
a
s
i
ngl
e
uni
t
,
r
e
duc
i
ng m
e
t
a
da
t
a
ove
r
he
a
d.
S
c
a
l
a
bi
l
i
t
y
M
or
e
c
ha
l
l
e
ngi
ng t
o s
c
a
l
e
a
t
a
l
a
r
ge
s
c
a
l
e
.
D
e
s
i
gne
d
f
or
l
a
r
ge
-
s
c
a
l
e
s
uppor
t
,
it
e
f
f
i
c
i
e
nt
l
y
ha
ndl
es
bi
l
l
i
ons
of
f
i
l
e
s
.
R
e
dunda
nc
y
D
e
pe
nds
on
r
e
dunda
nt
a
r
r
a
y
of
i
nde
pe
nde
nt
di
s
ks
(
R
A
I
D
)
s
ys
t
e
m
s
f
or
f
a
ul
t
t
ol
e
r
a
nc
e
.
U
s
e
s
di
s
t
r
i
but
e
d
r
e
pl
i
c
a
t
i
on
m
e
c
h
a
ni
s
m
s
f
or
da
t
a
r
e
l
i
a
bi
l
i
t
y a
nd
a
va
i
l
a
bi
l
i
t
y.
O
pt
i
m
i
z
a
t
i
on f
or
s
pe
c
i
f
i
c
da
t
a
N
ot
opt
i
m
i
z
e
d f
or
s
pe
c
i
f
i
c
da
t
a
t
ype
s
.
O
pt
i
m
i
z
e
d
f
or
phot
os
a
nd
ot
he
r
l
a
r
ge
m
e
di
a
,
i
nc
l
udi
ng m
a
na
ge
m
e
nt
of
l
a
r
ge
obj
e
c
t
s
i
z
e
s
.
M
e
t
a
da
t
a
m
a
na
g
e
m
e
nt
M
e
t
a
da
t
a
i
s
m
a
na
ge
d
s
e
pa
r
a
t
e
l
y,
r
e
qui
r
i
ng
a
ddi
t
i
ona
l
que
r
i
e
s
f
or
a
c
t
ua
l
da
t
a
.
M
e
t
a
da
t
a
i
s
s
t
or
e
d
w
i
t
h
da
t
a
i
n
a
s
i
ngl
e
i
nde
x,
a
c
c
e
l
e
r
a
t
i
ng da
t
a
r
e
t
r
i
e
va
l
.
O
pe
r
a
t
i
ona
l
c
os
t
R
e
l
a
t
i
ve
l
y hi
gh du
e
t
o r
e
l
i
a
nc
e
on
s
pe
c
i
a
l
i
z
e
d
ha
r
dw
a
r
e
.
L
ow
e
r
due
t
o
t
he
us
e
of
c
om
m
od
i
t
y
ha
r
dw
a
r
e
a
nd c
us
t
om
de
s
i
gn.
R
e
l
i
a
bi
l
i
t
y
R
e
l
i
e
s
on
t
r
a
di
t
i
ona
l
N
A
S
ha
r
dw
a
r
e
a
nd
s
of
t
w
a
r
e
.
M
or
e
r
e
l
i
a
bl
e
w
i
t
h
i
nt
e
gr
a
t
e
d
da
t
a
r
e
pl
i
c
a
t
i
on
a
nd f
a
ul
t
t
ol
e
r
a
nc
e
.
2.1.
D
e
f
in
it
io
n
an
d
c
om
p
on
e
n
t
s
of
H
ays
t
ac
k
I
n
F
a
c
e
book'
s
da
ta
s
to
r
a
ge
s
e
tu
p,
H
a
ys
ta
c
k
is
a
c
om
pl
e
x
s
y
s
te
m
th
a
t
he
lp
s
m
a
na
ge
la
r
ge
phot
o
c
ol
le
c
ti
ons
e
f
f
ic
ie
nt
ly
.
I
t
ha
s
th
r
e
e
m
a
in
pa
r
ts
:
th
e
s
to
r
e
,
di
r
e
c
t
or
y,
a
nd
c
a
c
he
,
e
a
c
h
im
por
ta
nt
f
or
im
pr
ovi
ng
how
da
ta
is
s
to
r
e
d
a
nd
r
e
tr
ie
ve
d.
T
he
s
to
r
e
hol
ds
m
os
t
phot
os
,
or
ga
ni
z
in
g
th
e
m
in
va
r
io
us
phys
ic
a
l
vol
um
e
s
to
pr
ovi
de
ba
c
kup
a
nd
de
pe
nda
bi
li
ty
a
ga
in
s
t
ha
r
dw
a
r
e
is
s
u
e
s
.
T
he
di
r
e
c
to
r
y
ta
ke
s
c
a
r
e
of
th
e
m
e
ta
da
ta
,
a
ll
ow
in
g
f
a
s
t
a
c
c
e
s
s
to
ne
c
e
s
s
a
r
y
in
f
or
m
a
ti
on
w
hi
le
r
e
duc
in
g
di
s
k
us
e
a
nd
m
e
ta
d
a
ta
pr
oc
e
s
s
in
g.
T
hi
s
gr
e
a
tl
y
im
pr
ove
s
pe
r
f
or
m
a
nc
e
.
L
a
s
tl
y,
th
e
c
a
c
he
a
c
ts
a
s
a
f
a
s
t
li
nk,
lo
w
e
r
in
g
de
la
y
by
qui
c
kl
y
pr
ovi
di
ng
f
r
e
que
nt
ly
a
c
c
e
s
s
e
d
da
ta
.
T
hi
s
th
r
e
e
-
pa
r
t
s
ys
te
m
not
onl
y
im
pr
ove
s
e
f
f
ic
ie
nc
y
but
a
ls
o
a
id
s
F
a
c
e
book
in
m
a
na
gi
ng
a
huge
a
m
ount
of
c
ont
e
nt
w
i
th
s
tr
e
ngt
h a
nd
f
le
xi
bi
li
ty
,
e
ns
ur
in
g
a
good us
e
r
e
xpe
r
ie
nc
e
e
ve
n a
s
t
he
de
m
a
nd f
or
s
to
r
a
ge
i
nc
r
e
a
s
e
s
[
10]
.
2.2.
H
is
t
or
ic
al
c
on
t
e
xt
an
d
d
e
ve
lo
p
m
e
n
t
of
H
ays
t
ac
k
T
he
r
is
e
of
H
a
ys
ta
c
k
a
s
a
ke
y da
ta
s
to
r
a
ge
de
s
ig
n i
s
c
lo
s
e
ly
t
ie
d
t
o F
a
c
e
book'
s
r
a
pi
d i
nc
r
e
a
s
e
i
n us
e
r
-
ge
ne
r
a
te
d
c
ont
e
nt
,
e
s
p
e
c
ia
ll
y
phot
os
.
A
t
f
ir
s
t,
F
a
c
e
book
u
s
e
d
N
A
S
s
y
s
te
m
s
[
8]
,
[
14]
,
[
15]
,
w
hi
c
h
be
c
a
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
3
,
S
e
pt
e
m
be
r
20
25
:
671
-
681
674
in
s
uf
f
ic
ie
nt
w
he
n
th
e
pl
a
tf
or
m
gr
e
w
to
hous
e
ove
r
260
bi
ll
i
on
phot
os
.
T
hi
s
in
a
d
e
qua
c
y
hi
ghl
ig
ht
e
d
th
e
ne
c
e
s
s
it
y
f
or
a
ne
w
s
to
r
a
ge
s
ys
te
m
to
m
e
e
t
th
e
gr
ow
in
g
ne
e
ds
f
or
s
pe
e
d,
r
e
li
a
bi
li
ty
,
a
nd
da
ta
in
te
gr
it
y.
T
he
de
ve
lo
pm
e
nt
of
H
a
ys
ta
c
k
dr
e
w
f
r
om
de
ta
il
e
d
c
a
s
e
s
tu
di
e
s
in
s
c
a
la
bl
e
s
y
s
te
m
de
s
ig
n,
unc
ove
r
in
g
im
por
ta
nt
w
a
ys
to
m
in
im
iz
e
di
s
k
ope
r
a
ti
ons
a
nd
m
a
k
e
m
e
ta
da
ta
pr
oc
e
s
s
i
ng
m
or
e
e
f
f
ic
ie
nt
,
bot
h
vi
ta
l
f
o
r
ha
ndl
in
g
la
r
ge
da
ta
s
e
ts
.
H
a
ys
ta
c
k
w
a
s
bui
lt
w
it
h
a
w
e
ll
-
s
tr
uc
tu
r
e
d
de
s
ig
n
th
a
t
in
c
lu
de
s
a
di
s
tr
ib
ut
e
d
c
a
c
he
,
s
to
r
a
ge
m
e
th
od
s
,
a
nd
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
,
w
hi
c
h
ta
c
kl
e
d
c
om
m
on
is
s
ue
s
,
im
pr
ovi
ng
th
r
oughput
a
nd
f
a
ul
t
to
le
r
a
nc
e
w
hi
le
us
in
g
in
e
xpe
ns
iv
e
ha
r
dw
a
r
e
[
9]
, [
17]
.
T
hi
s
ba
c
kgr
ound not onl
y
i
nf
o
r
m
e
d
H
a
ys
ta
c
k
'
s
de
s
ig
n pr
in
c
ip
le
s
but
a
ls
o l
a
id
th
e
gr
oundwor
k f
or
f
ut
u
r
e
s
c
a
la
bl
e
s
y
s
te
m
m
ode
ls
t
ha
t
c
oul
d ha
ndl
e
l
a
r
ge
a
m
ount
s
of
da
ta
.
2.3.
C
om
p
ar
is
on
w
it
h
t
r
ad
it
io
n
al
d
at
a s
t
or
age
s
ys
t
e
m
s
T
r
a
di
ti
ona
l
da
ta
s
to
r
a
ge
s
ys
te
m
s
of
te
n
us
e
N
A
S
s
e
tu
ps
,
w
hi
c
h
ha
ve
pr
obl
e
m
s
w
it
h
s
c
a
la
bi
li
ty
a
nd
pe
r
f
or
m
a
nc
e
.
H
a
ys
ta
c
k
'
s
a
r
c
hi
te
c
tu
r
e
s
m
a
r
tl
y
ta
c
kl
e
s
th
e
s
e
c
h
a
ll
e
nge
s
w
it
h
it
s
uni
que
s
to
r
e
,
di
r
e
c
to
r
y,
a
nd
c
a
c
he
f
e
a
tu
r
e
s
.
C
onve
nt
io
na
l
s
ys
te
m
s
of
te
n
e
xpe
r
ie
nc
e
s
lo
w
do
w
ns
due
to
is
s
ue
s
w
it
h
m
e
ta
da
ta
ha
ndl
in
g
a
nd
he
a
vy
di
s
k
I
/O,
m
a
ki
ng
it
ha
r
d
to
e
f
f
ic
ie
nt
ly
m
a
na
ge
la
r
ge
a
m
ount
s
of
da
ta
,
li
ke
F
a
c
e
book'
s
huge
phot
o
c
ol
le
c
ti
on.
O
n
th
e
ot
he
r
ha
nd,
H
a
y
s
ta
c
k
r
e
duc
e
s
th
e
ne
e
d
f
or
di
s
k
a
c
c
e
s
s
a
nd
u
s
e
s
a
di
s
tr
ib
ut
e
d
c
a
c
hi
ng
s
ys
te
m
to
pr
ovi
de
bot
h
hi
gh
th
r
oughput
a
nd
lo
w
la
t
e
nc
y,
w
hi
c
h
a
r
e
e
s
s
e
nt
i
a
l
f
or
r
e
a
l
-
ti
m
e
in
te
r
a
c
ti
ons
.
A
ddi
ti
ona
ll
y,
by
or
ga
ni
z
in
g
s
to
r
a
ge
in
to
lo
gi
c
a
l
a
nd
phys
ic
a
l
vol
um
e
s
w
it
h
s
tr
ong
r
e
dunda
nc
y,
H
a
ys
ta
c
k
boos
ts
f
a
ul
t
to
le
r
a
n
c
e
,
gr
e
a
tl
y
e
nha
nc
in
g
r
e
li
a
bi
li
ty
c
om
pa
r
e
d
to
tr
a
di
ti
ona
l
s
ys
te
m
s
[
9]
.
O
ve
r
a
ll
,
th
i
s
a
r
c
hi
te
c
tu
r
e
m
a
r
ks
a
c
le
a
r
c
ha
ng
e
f
r
om
ol
d
s
to
r
a
ge
m
e
th
od
s
,
m
a
ki
ng
it
a
gr
e
a
t
c
hoi
c
e
f
or
to
da
y’
s
da
ta
-
he
a
vy
a
ppl
ic
a
ti
ons
[
18]
.
2.4.
K
e
y f
e
at
u
r
e
s
o
f
H
ays
t
ac
k
ar
c
h
it
e
c
t
u
r
e
A
uni
que
pa
r
t
of
th
e
H
a
ys
ta
c
k
de
s
ig
n
is
it
s
ne
w
w
a
y
of
im
pr
o
vi
ng
phot
o
s
to
r
a
ge
a
nd
a
c
c
e
s
s
,
w
hi
c
h
is
c
r
uc
ia
l
f
or
ha
ndl
in
g
la
r
ge
di
gi
ta
l
c
ol
le
c
ti
ons
.
T
hi
s
s
y
s
te
m
is
m
a
de
to
us
e
s
e
v
e
r
a
l
phys
ic
a
l
vol
um
e
s
,
e
a
c
h
hol
di
ng
m
il
li
ons
of
im
a
ge
s
,
to
m
a
ke
s
ur
e
it
c
a
n
gr
ow
a
nd
s
ta
y
de
pe
nda
bl
e
.
T
he
w
a
y
vol
um
e
s
a
r
e
a
r
r
a
nge
d
in
H
a
ys
ta
c
k
a
ll
ow
s
phot
o
s
to
be
c
opi
e
d
a
c
r
os
s
di
f
f
e
r
e
nt
phys
ic
a
l
s
to
r
a
ge
,
gr
e
a
tl
y
r
e
duc
in
g
th
e
c
ha
nc
e
of
lo
s
in
g
da
ta
be
c
a
us
e
of
ha
r
dw
a
r
e
i
s
s
ue
s
.
A
ddi
ti
ona
ll
y,
by
us
in
g
a
di
s
tr
ib
ut
e
d
c
a
c
he
a
nd
c
ut
ti
ng
dow
n
on
di
s
k
ope
r
a
ti
ons
, t
he
de
s
ig
n a
c
hi
e
ve
s
a
good mi
x of
hi
gh s
pe
e
d a
nd l
ow
de
la
y, m
e
e
ti
ng t
he
hi
gh de
m
a
nd f
r
om
us
e
r
s
a
c
c
e
s
s
in
g
m
or
e
th
a
n
260
bi
ll
io
n
phot
os
.
T
he
s
e
im
pr
ove
m
e
nt
s
not
onl
y
m
a
ke
a
c
c
e
s
s
e
a
s
ie
r
but
a
ls
o
f
it
a
budge
t
-
f
r
ie
ndl
y
a
ppr
oa
c
h
f
or
m
a
na
gi
ng
c
lo
ud
da
ta
,
hi
ghl
ig
ht
in
g
th
e
ne
e
d
to
e
nh
a
nc
e
s
to
r
a
ge
s
ol
ut
io
ns
in
to
da
y'
s
da
ta
-
dr
iv
e
n
e
nvi
r
onm
e
nt
.
O
ve
r
a
ll
,
th
e
H
a
ys
ta
c
k
de
s
ig
n
s
how
s
im
por
ta
nt
f
e
a
tu
r
e
s
th
a
t
ta
c
kl
e
th
e
c
ha
ll
e
nge
s
of
la
r
ge
-
s
c
a
l
e
da
ta
s
to
r
a
ge
,
be
in
g
vi
ta
l
f
or
pl
a
tf
or
m
s
li
ke
F
a
c
e
book,
w
hi
c
h
ne
e
d
s
tr
ong
a
nd
e
f
f
e
c
ti
ve
s
ys
te
m
s
f
or
ha
ndl
in
g us
e
r
-
ge
ne
r
a
te
d
c
ont
e
nt
.
2.5.
R
ol
e
of
H
ays
t
ac
k
in
F
ac
e
b
ook
's
ove
r
al
l
in
f
r
a
s
t
r
u
c
t
u
r
e
I
n
F
a
c
e
book'
s
bi
g
te
c
h
e
nvi
r
onm
e
nt
,
H
a
y
s
ta
c
k
is
ke
y
in
s
ol
vi
ng
pr
obl
e
m
s
w
it
h
phot
o
s
to
r
a
ge
a
nd
r
e
tr
ie
va
l,
im
pr
ovi
ng
th
e
s
it
e
'
s
ove
r
a
ll
pe
r
f
or
m
a
nc
e
a
nd
u
s
e
r
e
x
pe
r
ie
nc
e
.
B
y
m
a
ki
ng
d
a
ta
m
a
na
g
e
m
e
nt
m
or
e
e
f
f
ic
ie
nt
,
H
a
ys
ta
c
k
he
lp
s
r
e
duc
e
th
e
s
lo
w
dow
ns
s
e
e
n
in
r
e
gul
a
r
s
to
r
a
ge
s
ys
te
m
s
,
s
how
n
by
it
s
c
le
ve
r
de
s
ig
n
th
a
t
c
ut
s
dow
n
on
di
s
k
op
e
r
a
ti
ons
a
nd
pr
oc
e
s
s
in
g
de
m
a
nd
s
[
9]
.
T
hi
s
bi
g
im
pr
ove
m
e
nt
is
not
ju
s
t
a
te
c
h
upgr
a
de
;
it
a
li
gns
w
it
h
F
a
c
e
book'
s
la
r
ge
r
goa
l
of
e
f
f
ic
ie
nt
ly
m
a
na
gi
ng
it
s
e
xt
e
ns
iv
e
ope
r
a
ti
ons
.
W
it
h
H
a
ys
ta
c
k
,
u
s
e
r
s
c
a
n e
a
s
il
y a
c
c
e
s
s
ov
e
r
260
bi
ll
io
n
phot
os
qui
c
k
ly
,
w
hi
c
h
is
c
r
uc
ia
l
f
or
ke
e
pi
ng
u
s
e
r
s
e
ng
a
ge
d
a
nd
s
a
ti
s
f
ie
d.
A
s
F
a
c
e
book
w
or
k
s
to
upgr
a
de
it
s
s
ys
te
m
s
to
s
up
por
t
gr
ow
in
g
a
m
ount
s
of
da
ta
,
H
a
ys
ta
c
k
'
s
r
ol
e
in
m
a
in
ta
in
in
g
f
a
s
t
a
nd
e
f
f
ic
ie
nt
op
e
r
a
ti
ons
s
how
s
it
s
vi
ta
l
im
por
ta
nc
e
in
th
e
c
om
pa
ny’
s
s
e
tu
p
[
19]
.
I
n
th
e
e
nd,
H
a
ys
ta
c
k
s
how
s
how
f
oc
us
e
d
s
ol
ut
io
ns
c
a
n
im
pr
ove
th
e
ove
r
a
ll
e
f
f
e
c
ti
ve
ne
s
s
of
th
e
in
f
r
a
s
tr
uc
tu
r
e
w
hi
l
e
pr
ovi
di
ng a
s
tr
ong a
nd s
c
a
la
bl
e
w
a
y t
o m
a
na
ge
da
ta
s
to
r
a
ge
.
3.
D
A
T
A
S
T
O
R
A
G
E
M
E
C
H
A
N
I
S
M
S
I
N
H
A
Y
S
T
A
C
K
T
he
s
to
r
a
ge
m
e
th
od
s
us
e
d
by
H
a
ys
ta
c
k
a
r
e
a
ke
y
pa
r
t
of
how
it
is
bui
lt
,
a
im
e
d
a
t
ha
ndl
in
g
F
a
c
e
book'
s
la
r
ge
phot
o
c
ol
le
c
ti
on
w
hi
le
s
ol
vi
ng
p
e
r
f
or
m
a
nc
e
is
s
ue
s
.
A
k
e
y
f
e
a
tu
r
e
of
th
is
s
e
tu
p
is
th
e
s
pl
it
in
to
phys
ic
a
l
vol
um
e
s
t
ha
t
s
to
r
e
m
il
li
ons
of
photos
, w
hi
c
h a
r
e
g
r
oupe
d i
nt
o l
ogi
c
a
l
vol
um
e
s
f
or
be
tt
e
r
ba
c
kup.
T
hi
s
a
ppr
oa
c
h he
lp
s
r
e
duc
e
t
he
c
ha
n
c
e
of
l
os
in
g da
ta
f
r
om
ha
r
dw
a
r
e
pr
obl
e
m
s
, m
a
ki
ng s
ur
e
i
t
w
or
ks
w
e
ll
i
n a
di
s
tr
ib
ut
e
d s
e
tu
p, w
hi
c
h i
s
i
m
por
ta
nt
f
o
r
da
ta
a
c
c
e
s
s
a
c
r
os
s
di
f
f
e
r
e
nt
c
lo
uds
a
nd r
e
gi
ons
[
20]
. T
he
s
ys
te
m
a
ls
o
in
c
lu
de
s
a
di
s
tr
ib
ut
e
d
c
a
c
he
to
im
pr
ove
s
pe
e
d
a
nd
lo
w
e
r
de
la
ys
,
m
a
ki
ng
da
ta
r
e
tr
ie
va
l
e
a
s
ie
r
.
R
e
c
e
nt
upda
te
s
s
how
th
a
t
knowing
m
e
m
or
y s
tr
uc
tu
r
e
s
is
im
por
ta
nt
,
a
s
it
r
e
v
e
a
l
s
th
e
hi
dde
n c
os
ts
r
e
la
te
d
to
d
a
ta
m
ove
m
e
nt
, a
de
ta
il
of
te
n
m
is
s
e
d
by
de
ve
lo
pe
r
s
[
16]
.
B
y
di
s
tr
ib
ut
in
g
r
e
a
d
a
nd
w
r
it
e
ta
s
k
s
a
m
ong
it
s
p
a
r
ts
,
H
a
y
s
ta
c
k
de
m
ons
tr
a
te
s
a
f
le
xi
bl
e
da
ta
s
to
r
a
ge
s
ol
ut
io
n
th
a
t
f
ul
f
il
ls
F
a
c
e
book'
s
la
r
ge
ope
r
a
ti
ona
l
r
e
qui
r
e
m
e
nt
s
[
9]
.
F
ig
ur
e
3
de
pi
c
ts
th
e
phot
o
-
s
e
r
vi
ng
s
t
a
c
k,
w
hi
c
h
c
ons
is
t
s
of
f
our
la
ye
r
s
:
br
ow
s
e
r
c
a
c
h
e
s
,
e
dge
c
a
c
he
s
,
or
ig
in
c
a
c
he
, a
nd
H
a
ys
ta
c
k
b
a
c
ke
nd.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
dv A
ppl
S
c
i
I
S
S
N
:
2252
-
8814
H
ay
s
ta
c
k
-
bas
e
d F
ac
e
book
’
s
dat
a s
to
r
age
ar
c
hi
te
c
tu
r
e
:
s
to
r
e
, di
r
e
c
to
r
y
, and
c
ac
he
(
T
ol
e
Sut
ik
no)
675
F
ig
ur
e
3. D
a
ta
s
to
r
a
ge
m
e
c
ha
ni
s
m
s
in
H
a
ys
t
a
c
k
F
ig
ur
e
3
il
lu
s
tr
a
te
s
th
e
d
a
ta
s
to
r
a
ge
m
e
c
ha
ni
s
m
in
th
e
H
a
ys
t
a
c
k
s
y
s
te
m
,
w
hi
c
h
le
ve
r
a
ge
s
va
r
io
us
ty
pe
s
of
c
a
c
he
s
pl
a
yi
ng
a
c
r
it
ic
a
l
r
ol
e
in
e
nha
nc
in
g
ove
r
a
ll
s
ys
te
m
pe
r
f
or
m
a
nc
e
a
nd
e
f
f
ic
ie
nc
y.
T
he
br
ow
s
e
r
c
a
c
he
s
to
r
e
s
w
e
bpa
g
e
e
le
m
e
nt
s
,
s
uc
h
a
s
im
a
ge
s
a
nd
ot
he
r
f
il
e
s
,
on
th
e
us
e
r
’
s
s
id
e
,
a
im
in
g
to
a
c
c
e
le
r
a
te
pa
g
e
lo
a
d
ti
m
e
s
a
nd
r
e
duc
e
s
e
r
ve
r
lo
a
d
by
ut
il
iz
in
g
p
r
e
vi
ous
ly
dow
nl
oa
de
d
da
ta
.
T
he
pop
-
e
nd
(
e
dge
)
c
a
c
he
ope
r
a
te
s
on
s
e
r
ve
r
s
lo
c
a
te
d
c
lo
s
e
to
us
e
r
s
,
s
uc
h
a
s
th
os
e
in
a
c
o
nt
e
nt
de
li
ve
r
y
ne
twor
k
(
C
D
N
)
,
s
to
r
in
g
c
opi
e
s
of
da
ta
to
r
e
duc
e
l
a
te
nc
y
a
nd
im
pr
ove
a
c
c
e
s
s
s
pe
e
d
by
de
li
ve
r
i
ng
c
ont
e
nt
f
r
om
s
e
r
ve
r
s
g
e
ogr
a
phi
c
a
ll
y
c
lo
s
e
r
to
th
e
us
e
r
.
M
e
a
nw
hi
le
,
th
e
or
ig
in
c
a
c
h
e
f
unc
ti
ons
on
th
e
or
ig
in
s
e
r
ve
r
by
s
to
r
in
g
f
r
e
que
nt
ly
a
c
c
e
s
s
e
d
d
a
ta
c
opi
e
s
,
de
s
ig
ne
d
to
e
xpe
di
te
da
ta
a
va
il
a
bi
li
ty
dur
in
g
r
e
que
s
ts
a
nd
a
ll
e
vi
a
te
th
e
bur
de
n
on
th
e
p
r
im
a
r
y
s
to
r
a
ge
s
e
r
ve
r
. A
ddi
ti
ona
ll
y, t
he
ba
c
ke
nd c
a
c
he
i
s
e
m
pl
oye
d w
it
hi
n ba
c
ke
nd s
ys
te
m
s
,
s
uc
h a
s
H
a
y
s
ta
c
k, t
o s
to
r
e
l
a
r
ge
da
ta
obj
e
c
ts
,
s
uc
h
a
s
im
a
ge
s
a
nd
ot
he
r
m
e
di
a
,
w
it
h
th
e
pur
po
s
e
of
a
c
c
e
le
r
a
ti
ng
da
ta
r
e
tr
ie
va
l,
r
e
duc
in
g
th
e
lo
a
d
on
s
lo
w
e
r
s
to
r
a
ge
s
ys
t
e
m
s
,
a
nd
e
nha
n
c
in
g
da
ta
r
e
li
a
bi
li
ty
a
nd
a
va
il
a
bi
li
ty
th
r
ough
r
e
pl
ic
a
ti
on
a
nd
di
s
tr
ib
ut
e
d
s
to
r
a
ge
m
e
c
ha
ni
s
m
s
.
A
ll
th
e
s
e
c
a
c
hi
ng
m
e
c
h
a
ni
s
m
s
w
or
k
s
yne
r
gi
s
ti
c
a
ll
y
to
e
ns
ur
e
f
a
s
t,
e
f
f
ic
ie
nt
,
a
nd
r
e
li
a
bl
e
da
ta
a
c
c
e
s
s
a
c
r
os
s
a
ll
la
ye
r
s
of
th
e
s
to
r
a
ge
s
ys
te
m
.
T
he
in
te
r
a
c
ti
on
a
m
ong
di
f
f
e
r
e
nt
ty
pe
s
of
c
a
c
he
s
,
br
ow
s
e
r
,
e
dge
,
or
ig
in
,
a
nd
ba
c
ke
nd
s
uppor
ts
la
r
ge
-
s
c
a
l
e
da
ta
di
s
tr
ib
ut
io
n
opt
im
iz
a
ti
on
by
m
in
im
iz
in
g
la
te
nc
y a
nd i
m
pr
ovi
ng ove
r
a
ll
da
ta
a
c
c
e
s
s
s
pe
e
d a
nd
e
f
f
ic
ie
nc
y.
3.1.
O
ve
r
vi
e
w
of
d
at
a s
t
or
age
t
e
c
h
n
iq
u
e
s
u
s
e
d
I
n
m
ode
r
n
da
ta
s
to
r
a
ge
s
ys
te
m
s
,
di
f
f
e
r
e
nt
m
e
th
ods
a
r
e
im
por
ta
nt
f
or
m
a
ki
ng
th
in
gs
e
f
f
ic
ie
nt
,
r
e
li
a
bl
e
,
a
nd
s
c
a
la
bl
e
.
O
ne
ke
y
m
e
th
od
is
u
s
in
g
di
s
tr
ib
ut
e
d
s
t
or
a
ge
s
e
tu
ps
,
w
hi
c
h
u
s
e
s
t
a
nda
r
d
ha
r
dw
a
r
e
to
bui
ld
a
s
tr
ong
in
f
r
a
s
tr
uc
tu
r
e
th
a
t
c
a
n
ha
ndl
e
la
r
ge
a
m
ount
s
of
da
ta
.
A
n
e
xa
m
pl
e
is
F
a
c
e
book'
s
H
a
y
s
ta
c
k
s
ys
te
m
,
w
hi
c
h
e
f
f
e
c
ti
ve
ly
de
a
l
s
w
it
h
th
e
is
s
u
e
s
of
s
to
r
in
g
a
lo
t
of
phot
os
.
I
t
doe
s
th
is
by
us
in
g
bot
h
phys
ic
a
l
a
nd
lo
gi
c
a
l
vol
um
e
m
a
na
ge
m
e
nt
,
a
lo
ng
w
it
h
r
e
dunda
nc
y
m
e
th
ods
to
a
voi
d
da
ta
lo
s
s
.
B
y
s
pl
it
ti
ng
phot
os
in
to
s
e
ve
r
a
l
vol
um
e
s
a
nd
u
s
in
g
a
di
s
tr
ib
ut
e
d
c
a
c
he
f
or
qui
c
k
a
c
c
e
s
s
,
H
a
y
s
ta
c
k
gr
e
a
tl
y
e
nha
nc
e
s
r
e
a
d
a
nd
w
r
it
e
pr
oc
e
s
s
e
s
,
le
a
di
ng
to
be
tt
e
r
pe
r
f
or
m
a
nc
e
a
nd
s
ys
t
e
m
s
tr
e
ngt
h
[
9]
.
A
ddi
ti
ona
ll
y,
ongoing
ba
c
kgr
ound
ta
s
ks
in
th
is
s
ys
te
m
ke
e
p
a
n
e
ye
on
pot
e
nt
ia
l
pr
obl
e
m
s
a
nd
f
ix
th
e
m
,
th
us
m
a
in
ta
in
in
g
da
ta
in
te
g
r
it
y
a
nd
s
m
oot
h
ope
r
a
ti
ons
[
11]
.
T
hi
s
va
r
ie
d
a
ppr
oa
c
h
to
d
a
ta
s
to
r
a
ge
hi
ghl
ig
ht
s
th
e
n
e
e
d
f
or
c
r
e
a
ti
ve
de
s
ig
n
in
de
a
li
ng
w
it
h
to
da
y’
s
di
gi
ta
l
c
ha
ll
e
nge
s
, e
s
pe
c
i
a
ll
y i
n a
r
e
a
s
w
it
h r
a
pi
d da
ta
gr
ow
th
.
3.2.
F
il
e
s
t
or
age
an
d
m
an
age
m
e
n
t
i
n
H
ays
t
ac
k
E
f
f
ic
ie
nt
f
il
e
s
to
r
a
ge
a
nd
m
a
na
ge
m
e
nt
in
H
a
y
s
ta
c
k
'
s
s
y
s
te
m
h
e
lp
s
le
s
s
e
n
pr
obl
e
m
s
of
de
a
li
ng
w
it
h
la
r
ge
da
ta
a
m
ount
s
s
uc
h a
s
F
a
c
e
book’
s
bi
g
phot
o
li
br
a
r
y.
B
y
or
ga
ni
z
in
g
s
to
r
a
ge
in
to
m
a
ny
phy
s
ic
a
l
vol
um
e
s
,
e
a
c
h
hol
di
ng
m
il
li
ons
of
phot
os
,
H
a
ys
ta
c
k
pr
ovi
de
s
a
n
or
de
r
ly
a
nd
s
tr
ong
w
a
y
to
ha
ndl
e
da
ta
.
K
e
y
pa
r
ts
of
th
e
s
ys
te
m
in
c
lu
de
th
e
H
a
y
s
ta
c
k
s
to
r
e
,
di
r
e
c
to
r
y,
a
nd
c
a
c
h
e
,
w
hi
c
h
w
or
k
to
ge
th
e
r
to
im
pr
ove
da
ta
a
c
c
e
s
s
.
R
e
dunda
nc
y
m
e
th
od
s
in
phys
i
c
a
l
vol
um
e
s
pr
ot
e
c
t
a
ga
in
s
t
da
ta
lo
s
s
f
r
om
ha
r
dw
a
r
e
is
s
ue
s
,
s
ho
w
in
g
a
f
or
w
a
r
d
-
th
in
ki
ng
a
ppr
oa
c
h
to
f
il
e
m
a
na
ge
m
e
nt
.
E
a
c
h
lo
gi
c
a
l
v
ol
um
e
is
de
s
ig
ne
d
to
dupl
ic
a
te
phot
o
s
a
c
r
os
s
phys
ic
a
l
s
to
r
a
ge
uni
ts
,
m
a
ki
ng
it
m
or
e
r
e
li
a
bl
e
a
nd
a
bl
e
to
r
e
s
i
s
t
f
a
ul
ts
[
9]
.
A
ls
o,
H
a
ys
ta
c
k
a
ll
ow
s
f
or
qui
c
k
a
c
c
e
s
s
to
im
a
ge
s
,
im
pr
ovi
ng
us
e
r
e
xpe
r
ie
nc
e
w
hi
le
r
e
du
c
in
g
d
is
k
ope
r
a
ti
ons
a
nd
m
e
ta
da
ta
lo
a
d
[
16]
.
T
he
s
e
in
nova
ti
ons
hi
ghl
ig
ht
how
im
por
ta
nt
H
a
ys
ta
c
k
i
s
in
s
uppor
ti
ng
F
a
c
e
book’
s
a
bi
li
ty
to
m
a
na
ge
a
nd
s
to
r
e
d
a
ta
s
m
oot
hl
y.
3.3.
D
at
a r
e
p
li
c
at
io
n
s
t
r
at
e
gi
e
s
e
m
p
lo
ye
d
R
e
li
a
bi
li
ty
i
n da
ta
s
to
r
a
ge
s
y
s
te
m
s
gr
e
a
tl
y de
pe
nd
s
on good r
e
p
li
c
a
ti
on me
th
ods
, pa
r
ti
c
ul
a
r
ly
i
n l
a
r
ge
s
e
tu
ps
li
ke
F
a
c
e
book'
s
H
a
y
s
ta
c
k
.
A
t
th
e
c
e
nt
e
r
of
H
a
ys
ta
c
k
'
s
de
s
ig
n
is
a
s
tr
ong
da
ta
r
e
pl
ic
a
ti
on
s
ys
te
m
th
a
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
3
,
S
e
pt
e
m
be
r
20
25
:
671
-
681
676
pr
ovi
de
s
r
e
dunda
nc
y
ove
r
s
e
ve
r
a
l
phy
s
ic
a
l
vol
um
e
s
,
w
hi
c
h
h
e
lp
s
pr
e
ve
nt
da
ta
lo
s
s
f
r
om
ha
r
dw
a
r
e
f
a
il
ur
e
s
.
E
a
c
h
vol
um
e
hol
ds
m
il
li
ons
o
f
phot
os
,
a
r
r
a
nge
d
in
to
lo
g
ic
a
l
gr
oups
f
or
be
tt
e
r
a
c
c
e
s
s
a
nd
m
a
in
te
na
nc
e
,
w
hi
le
a
ls
o
im
pr
ovi
ng
r
e
li
a
bi
li
ty
th
r
ough
c
a
r
e
f
ul
dupl
ic
a
ti
on. T
hi
s
m
e
t
hod
r
e
duc
e
s
th
e
r
is
k
of
s
in
gl
e
poi
nt
s
of
f
a
il
ur
e
a
nd
e
nha
nc
e
s
r
e
a
d
a
nd
w
r
it
e
pr
oc
e
s
s
e
s
,
a
ll
ow
in
g
r
e
que
s
t
s
to
be
s
pr
e
a
d
out
a
c
r
os
s
di
f
f
e
r
e
nt
r
e
pl
ic
a
s
.
H
a
ys
ta
c
k
us
e
s
s
ta
nda
r
d
ha
r
dw
a
r
e
to
c
a
r
r
y
out
th
e
s
e
m
e
th
ods
,
e
na
bl
in
g
c
os
t
-
e
f
f
e
c
ti
ve
gr
ow
th
w
hi
le
e
ns
ur
in
g
hi
gh
th
r
oughput
a
nd
lo
w
de
la
y
s
in
m
e
ta
da
ta
h
a
ndl
in
g
a
nd
a
c
c
e
s
s
.
W
it
h
ongoing
ba
c
kgr
ound
ta
s
k
s
f
oc
us
e
d
on
c
he
c
ki
ng
a
nd
f
ix
in
g
da
ta
in
te
gr
it
y
pr
ob
le
m
s
,
H
a
ys
ta
c
k
s
e
r
ve
s
a
s
a
s
ol
id
r
e
pl
ic
a
ti
on
m
ode
l
th
a
t
va
lu
e
s
bot
h
pe
r
f
or
m
a
nc
e
a
nd r
e
li
a
bi
li
ty
[
11]
.
3.4.
S
c
al
ab
il
it
y c
on
s
id
e
r
at
io
n
s
i
n
d
a
t
a s
t
or
age
I
n
da
ta
s
to
r
a
ge
,
th
in
ki
ng
a
bout
s
c
a
la
bi
li
ty
is
ve
r
y
im
por
ta
nt
f
or
ha
ndl
in
g
th
e
r
a
pi
d
in
c
r
e
a
s
e
of
us
e
r
-
ge
ne
r
a
te
d
c
ont
e
nt
,
e
s
p
e
c
ia
ll
y
on
pl
a
tf
or
m
s
li
ke
F
a
c
e
book, whic
h
de
a
ls
w
it
h
bi
ll
io
ns
of
phot
os
e
ve
r
y
da
y.
T
he
de
s
ig
n us
e
d i
n
H
a
ys
ta
c
k
s
how
s
i
m
por
ta
nt
w
a
ys
t
o a
c
hi
e
v
e
s
c
a
la
bi
li
ty
, l
ik
e
r
e
m
ovi
ng bott
le
ne
c
ks
f
ound in
tr
a
di
ti
ona
l
N
A
S
s
ys
te
m
s
a
nd
us
in
g
a
di
s
tr
ib
ut
e
d
c
a
c
hi
ng
s
y
s
te
m
th
a
t
im
pr
ove
s
bot
h
s
pe
e
d
a
nd
r
e
s
pons
e
ti
m
e
.
B
y
s
pr
e
a
di
ng
da
ta
ove
r
s
e
ve
r
a
l
phys
ic
a
l
vol
um
e
s
,
e
a
c
h
w
it
h
m
il
li
ons
of
phot
os
,
H
a
ys
ta
c
k
a
ll
ow
s
th
e
s
ys
te
m
to
ha
ndl
e
la
r
ge
da
ta
s
e
ts
e
f
f
e
c
ti
ve
ly
w
hi
le
ke
e
pi
ng
ba
c
kups
to
gua
r
d
a
ga
in
s
t
ha
r
dw
a
r
e
is
s
ue
s
.
T
hi
s
de
s
ig
n
s
uppor
ts
r
e
li
a
bi
li
ty
a
nd
f
a
ul
t
to
le
r
a
nc
e
s
in
c
e
b
a
c
kgr
ound
ta
s
k
s
ke
e
p
a
n
e
ye
on
a
nd
f
ix
pos
s
ib
le
pr
obl
e
m
s
,
e
ns
ur
in
g
th
e
s
ys
te
m
w
or
ks
w
e
ll
w
it
hout
s
lo
w
in
g
dow
n
[
9]
.
A
s
th
e
ne
e
d
f
or
s
to
r
a
ge
gr
ow
s
,
th
e
in
s
ig
ht
s
f
r
om
H
a
ys
ta
c
k
'
s
a
ppr
oa
c
h
hi
ghl
ig
ht
th
e
ne
e
d
f
or
a
s
tr
ong, s
c
a
la
bl
e
f
r
a
m
e
w
or
k
th
a
t
c
a
n
c
ha
nge
to
m
e
e
t
th
e
ne
e
ds
of
da
ta
-
he
a
vy a
ppl
ic
a
ti
ons
[
11]
.
3.5.
P
e
r
f
or
m
an
c
e
m
e
t
r
ic
s
f
or
d
at
a s
t
or
age
e
f
f
ic
ie
n
c
y
K
now
in
g
how
to
m
e
a
s
ur
e
da
ta
s
to
r
a
ge
e
f
f
ic
ie
nc
y
is
im
por
ta
nt
f
or
a
s
s
e
s
s
in
g
s
ys
te
m
s
li
ke
F
a
c
e
book'
s
H
a
ys
ta
c
k
.
T
h
e
r
e
a
r
e
m
a
in
m
e
tr
ic
s
to
c
ons
id
e
r
,
li
ke
th
r
ough
put
,
la
te
nc
y,
a
nd
r
e
dunda
nc
y
le
ve
ls
,
w
hi
c
h
to
ge
th
e
r
a
na
ly
z
e
ho
w
w
e
ll
d
a
ta
is
s
to
r
e
d
a
nd
r
e
tr
ie
ve
d.
T
hr
oughput
s
how
s
th
e
s
p
e
e
d
of
r
e
a
di
ng
or
w
r
it
in
g
da
ta
,
im
pa
c
ti
ng
us
e
r
e
xpe
r
ie
nc
e
a
nd
s
y
s
te
m
pe
r
f
or
m
a
nc
e
.
L
a
te
nc
y
in
di
c
a
te
s
how
lo
ng
it
ta
ke
s
to
ge
t
da
ta
,
a
de
la
y t
ha
t
c
a
n be
i
m
pr
ove
d w
it
h good c
a
c
hi
ng t
e
c
hni
que
s
f
ou
nd i
n
H
a
ys
ta
c
k
, l
e
a
di
ng t
o be
tt
e
r
pe
r
f
or
m
a
nc
e
.
A
ddi
ti
ona
ll
y,
m
e
tr
ic
s
on
r
e
dunda
nc
y
a
r
e
c
r
uc
ia
l
f
or
ke
e
pi
ng
da
ta
s
a
f
e
a
nd
a
va
il
a
bl
e
;
th
e
r
e
pl
ic
a
ti
on
m
e
th
ods
us
e
d
in
H
a
y
s
ta
c
k
he
lp
pr
e
ve
nt
d
a
ta
lo
s
s
f
r
om
ha
r
dw
a
r
e
is
s
ue
s
,
in
c
r
e
a
s
in
g
r
e
li
a
bi
li
ty
[
9]
.
B
y
ba
la
nc
in
g
th
e
s
e
m
e
a
s
ur
e
m
e
nt
s
,
s
y
s
te
m
de
s
ig
ne
r
s
c
a
n
e
ns
ur
e
th
a
t
th
e
da
ta
s
to
r
a
ge
s
e
tu
p
m
e
e
ts
c
ur
r
e
nt
ne
e
d
s
a
nd
c
a
n
gr
ow
e
f
f
ic
ie
nt
ly
f
or
m
or
e
us
e
r
s
[
12]
.
4.
D
I
R
E
C
T
O
R
Y
S
E
R
V
I
C
E
S
I
N
H
A
Y
S
T
A
C
K
I
nt
e
gr
a
ti
ng
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
in
th
e
H
a
ys
ta
c
k
s
tr
uc
tu
r
e
is
im
por
ta
nt
f
or
im
pr
ovi
ng
how
da
ta
is
r
e
tr
ie
ve
d
a
nd
e
ns
ur
in
g
good
a
c
c
e
s
s
to
th
e
la
r
ge
a
m
ount
s
o
f
phot
os
s
to
r
e
d
on
F
a
c
e
book'
s
s
ys
te
m
.
T
he
di
r
e
c
to
r
y a
c
ts
a
s
a
l
in
k be
tw
e
e
n da
ta
s
to
r
a
ge
a
nd c
li
e
nt
r
e
que
s
ts
,
he
lp
in
g t
o quic
kl
y f
in
d phys
ic
a
l
vol
um
e
s
t
ha
t
c
ont
a
in
pa
r
ti
c
ul
a
r
phot
o
da
ta
,
w
hi
c
h
r
e
duc
e
s
w
a
it
in
g
ti
m
e
s
.
T
hi
s
s
e
tu
p
m
e
e
ts
th
e
s
ys
te
m
'
s
m
a
in
a
im
of
m
a
ki
ng
ope
r
a
ti
ons
s
m
oot
he
r
by
lo
w
e
r
in
g
th
e
c
om
put
in
g
lo
a
d
li
nke
d
to
m
e
ta
da
ta
pr
oc
e
s
s
in
g,
w
hi
c
h
is
a
bi
g
upgr
a
de
c
om
pa
r
e
d
to
ol
de
r
NAS
opt
io
ns
.
A
ls
o,
th
e
di
r
e
c
to
r
y
u
s
e
s
a
lo
gi
c
a
l
la
y
e
r
th
a
t
boos
t
s
f
a
ul
t
to
le
r
a
nc
e
,
m
a
ki
ng
s
ur
e
th
a
t
e
ve
n
if
th
e
r
e
is
a
ha
r
dw
a
r
e
pr
obl
e
m
,
da
ta
c
a
n
s
ti
ll
be
a
c
c
e
s
s
e
d
a
nd
is
s
e
c
ur
e
.
B
y
u
s
in
g
th
e
s
e
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
,
H
a
ys
ta
c
k
not
onl
y
im
pr
ove
s
e
f
f
ic
ie
nc
y
but
a
ls
o
m
a
ke
s
be
tt
e
r
us
e
of
r
e
s
our
c
e
s
,
w
hi
c
h
is
im
por
ta
nt
gi
ve
n
th
e
r
is
in
g
de
m
a
nds
f
or
c
r
os
s
-
c
lo
ud
da
ta
a
c
c
e
s
s
s
e
e
n
in
r
e
c
e
nt
s
tu
di
e
s
[
20]
.
F
ur
th
e
r
m
or
e
,
th
e
di
r
e
c
to
r
y'
s
r
ol
e
f
i
ts
w
it
h c
ur
r
e
nt
c
lo
ud
opt
im
iz
a
ti
on t
r
e
nds
, a
s
hi
ghl
ig
ht
e
d i
n r
e
c
e
nt
s
tu
di
e
s
t
ha
t
s
tr
e
s
s
t
he
ne
e
d
f
or
f
le
xi
bl
e
s
ys
te
m
s
th
a
t
c
a
n
lo
w
e
r
r
e
s
our
c
e
c
ons
um
pt
io
n
w
hi
le
ke
e
pi
ng
up
pe
r
f
or
m
a
nc
e
[
21]
.
H
e
n
c
e
,
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
in
H
a
ys
ta
c
k
a
r
e
c
r
uc
ia
l
f
or
pr
ovi
di
ng
a
f
le
xi
bl
e
a
nd
s
tr
ong
da
ta
s
to
r
a
ge
s
ol
ut
io
n,
c
onf
ir
m
in
g i
ts
s
ig
ni
f
ic
a
nc
e
i
n t
oda
y'
s
c
lo
ud c
om
put
in
g s
tr
a
te
gi
e
s
.
4.1.
F
u
n
c
t
io
n
al
it
y of
d
ir
e
c
t
or
y s
e
r
vi
c
e
s
i
n
H
ays
t
ac
k
T
he
e
f
f
e
c
ti
ve
ne
s
s
of
da
ta
m
a
na
ge
m
e
nt
in
H
a
y
s
ta
c
k
la
r
ge
ly
de
p
e
nds
on
it
s
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
,
w
hi
c
h
a
r
e
c
r
uc
ia
l
f
or
na
vi
ga
ti
ng
th
e
m
a
ny
s
to
r
e
d
it
e
m
s
.
T
he
di
r
e
c
to
r
y
s
e
r
ve
s
a
s
a
m
e
ta
da
ta
l
a
ye
r
th
a
t
e
f
f
e
c
ti
ve
ly
li
nks
us
e
r
s
e
a
r
c
he
s
to
a
c
tu
a
l
s
to
r
a
ge
lo
c
a
ti
ons
,
th
u
s
le
s
s
e
ni
ng
t
he
us
ua
l
bur
de
n
s
r
e
la
te
d
to
m
e
ta
da
ta
s
e
a
r
c
h
e
s
.
U
nl
ik
e
s
ta
nda
r
d
di
r
e
c
to
r
y
s
ys
te
m
s
,
H
a
ys
t
a
c
k
'
s
di
r
e
c
to
r
y
is
m
a
d
e
to
c
ut
dow
n
on
di
s
k
ope
r
a
ti
ons
,
r
e
s
ul
ti
ng
in
m
uc
h
be
tt
e
r
pe
r
f
or
m
a
nc
e
f
or
r
e
a
di
ng
a
nd
w
r
it
in
g.
E
a
c
h
e
nt
r
y
in
th
e
di
r
e
c
to
r
y
is
c
a
r
e
f
ul
ly
c
r
a
f
te
d
to
ha
ndl
e
hi
gh
de
m
a
nd
f
or
s
pe
e
d
a
nd
lo
w
w
a
it
ti
m
e
s
,
m
e
e
ti
ng
F
a
c
e
book’
s
ne
e
ds
f
or
qui
c
kl
y
a
c
c
e
s
s
in
g
a
nd
d
e
li
ve
r
in
g
a
va
s
t
num
be
r
of
phot
os
.
A
ddi
ti
ona
ll
y,
th
e
s
m
oot
h
in
te
gr
a
ti
on
of
th
e
di
r
e
c
to
r
y
w
it
h
th
e
H
a
ys
ta
c
k
s
to
r
e
a
nd
c
a
c
he
s
ys
t
e
m
s
[
9]
gua
r
a
nt
e
e
s
th
a
t
r
e
tr
ie
vi
ng
da
ta
is
bot
h
f
a
s
t
a
nd
r
e
li
a
bl
e
,
s
uppor
ti
ng
th
e
ove
r
a
ll
goa
l
o
f
r
e
duc
in
g
c
os
ts
w
hi
le
r
e
m
a
in
in
g
s
c
a
la
bl
e
[
20]
.
T
hr
ough
it
s
un
iq
ue
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
,
H
a
y
s
ta
c
k
s
how
s
how
s
m
a
r
t
de
s
ig
n c
a
n he
lp
s
ol
ve
da
ta
a
c
c
e
s
s
i
s
s
ue
s
a
nd i
m
pr
ove
t
he
e
f
f
ic
ie
nc
y of
s
to
r
a
ge
s
ys
te
m
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
dv A
ppl
S
c
i
I
S
S
N
:
2252
-
8814
H
ay
s
ta
c
k
-
bas
e
d F
ac
e
book
’
s
dat
a s
to
r
age
ar
c
hi
te
c
tu
r
e
:
s
to
r
e
, di
r
e
c
to
r
y
, and
c
ac
he
(
T
ol
e
Sut
ik
no)
677
4.2.
D
at
a i
n
d
e
xi
n
g an
d
r
e
t
r
ie
val
p
r
oc
e
s
s
e
s
D
a
ta
in
de
xi
ng
a
nd
r
e
tr
ie
va
l
a
r
e
ve
r
y
im
po
r
ta
nt
f
o
r
la
r
ge
s
to
r
a
g
e
s
ys
te
m
s
li
ke
F
a
c
e
book’
s
H
a
y
s
ta
c
k
,
w
hi
c
h
he
lp
s
m
a
na
ge
bi
ll
io
ns
of
phot
os
.
H
a
ys
ta
c
k
’
s
s
tr
uc
tu
r
e
in
c
lu
de
s
s
pe
c
i
a
l
in
de
xi
ng
m
e
th
ods
th
a
t
m
a
ke
r
e
tr
ie
va
l
f
a
s
te
r
a
nd
c
ut
dow
n
on
s
lo
w
di
s
k
ope
r
a
ti
ons
.
B
y
c
o
m
bi
ni
ng
how
da
ta
is
lo
gi
c
a
ll
y
a
nd
phy
s
ic
a
ll
y
or
ga
ni
z
e
d,
H
a
ys
ta
c
k
ke
e
ps
da
ta
s
to
r
e
d
w
e
ll
a
nd
e
a
s
y
to
ge
t
w
h
e
n
us
e
r
s
ne
e
d
it
.
T
h
e
s
ys
te
m
a
im
s
to
lo
w
e
r
th
e
w
or
kl
oa
d
on
m
e
ta
da
ta
pr
oc
e
s
s
in
g
to
im
pr
ove
pe
r
f
or
m
a
nc
e
,
e
na
bl
in
g
qui
c
k
a
c
c
e
s
s
w
it
hout
s
tr
e
s
s
in
g
th
e
s
to
r
a
ge
ha
r
dw
a
r
e
.
A
dd
it
i
on
a
ll
y,
it
u
s
e
s
a
di
s
t
r
ib
ut
e
d
c
a
c
h
e
s
ys
te
m
to
e
n
s
ur
e
f
a
s
t
d
a
t
a
r
e
tr
i
e
v
a
l
w
it
h
h
ig
h
th
r
o
ugh
put
a
nd
l
ow
de
la
y
s
, w
hi
c
h
a
l
s
o
h
e
l
ps
t
h
e
s
y
s
t
e
m
r
e
s
i
s
t
f
a
ul
t
s
[
9]
,
[
2
2]
. O
v
e
r
a
ll
,
t
he
s
e
a
dv
a
n
c
e
d
i
nd
e
xi
ng
a
nd
r
e
tr
i
e
v
a
l
m
e
th
od
s
a
r
e
c
r
u
c
i
a
l
f
or
ke
e
p
in
g F
a
c
e
b
oo
k’
s
l
a
r
g
e
p
hot
o
s
to
r
a
ge
s
y
s
t
e
m
r
e
s
p
on
s
i
ve
a
nd
r
e
l
ia
bl
e
.
4.3.
R
ol
e
of
m
e
t
ad
at
a i
n
d
ir
e
c
t
or
y s
e
r
vi
c
e
s
I
n
m
ode
r
n
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
,
m
e
ta
da
ta
pl
a
ys
a
n
im
por
ta
nt
r
ol
e
in
m
a
ki
ng
da
ta
r
e
t
r
ie
va
l
s
ys
te
m
s
w
or
k be
tt
e
r
.
P
r
ovi
di
ng
in
f
or
m
a
ti
on a
bout
th
e
da
ta
-
li
ke
w
he
r
e
i
t
c
om
e
s
f
r
om
, i
ts
l
a
yout
, a
nd how it conne
c
ts
t
o
ot
he
r
da
ta
-
m
e
ta
da
ta
,
he
lp
s
us
e
r
s
a
c
c
e
s
s
r
e
s
our
c
e
s
qui
c
kl
y
a
nd
s
uppor
ts
da
ta
in
te
gr
it
y
a
nd
m
a
na
ge
m
e
nt
e
f
f
or
ts
.
F
or
e
xa
m
pl
e
,
in
F
a
c
e
book’
s
H
a
ys
ta
c
k
s
ys
te
m
,
m
a
n
a
g
in
g
m
e
ta
da
ta
is
ke
y
to
s
or
ti
ng
th
r
ough
la
r
ge
phot
o
c
ol
le
c
ti
ons
s
to
r
e
d
in
di
f
f
e
r
e
nt
phys
ic
a
l
a
nd
lo
gi
c
a
l
lo
c
a
ti
ons
,
w
hi
c
h
h
e
lp
s
pr
e
ve
nt
d
a
ta
lo
s
s
f
r
om
ha
r
dw
a
r
e
f
a
il
ur
e
s
.
T
hi
s
m
e
th
od,
w
he
r
e
m
e
ta
d
a
ta
f
unc
ti
ons
a
s
a
m
a
in
di
r
e
c
to
r
y,
a
s
s
is
ts
in
f
a
s
t
lo
okups
a
nd
ke
e
ps
a
hi
gh
r
a
te
of
da
ta
pr
oc
e
s
s
in
g
w
hi
le
c
ut
ti
ng
dow
n
on
de
la
ys
in
a
c
c
e
s
s
in
g
da
ta
.
A
ddi
ti
ona
ll
y,
us
in
g
m
e
ta
da
ta
e
f
f
e
c
ti
ve
ly
he
lp
s
to
opt
im
iz
e
s
to
r
a
ge
a
nd
e
nha
nc
e
f
a
ul
t
to
le
r
a
nc
e
s
in
c
e
it
e
na
bl
e
s
ba
c
kgr
ound
pr
oc
e
s
s
e
s
to
c
h
e
c
k
f
or
a
nd
f
ix
is
s
ue
s
e
a
r
ly
,
th
us
im
pr
ovi
ng
ov
e
r
a
ll
s
ys
te
m
r
e
li
a
bi
li
ty
[
9]
,
[
11]
.
I
f
th
e
r
e
w
e
r
e
no
s
ol
id
m
e
ta
da
ta
s
y
s
te
m
s
,
th
e
e
f
f
ic
ie
nc
y
a
nd
gr
ow
th
of
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
w
oul
d
b
e
s
e
r
io
us
ly
a
f
f
e
c
te
d,
ne
ga
ti
ve
ly
i
m
pa
c
ti
ng us
e
r
e
xpe
r
ie
nc
e
a
nd c
om
pa
ny pr
oduc
ti
vi
t
y.
4.4.
C
h
al
le
n
ge
s
i
n
d
ir
e
c
t
or
y m
an
age
m
e
n
t
M
a
na
gi
ng
di
r
e
c
to
r
ie
s
in
bi
g
da
ta
s
to
r
a
ge
s
ys
te
m
s
,
li
ke
F
a
c
e
book'
s
H
a
ys
t
a
c
k
,
c
om
e
s
w
it
h
m
a
ny
c
ha
ll
e
nge
s
th
a
t
ne
e
d
ha
ndl
in
g
f
or
s
ys
te
m
e
f
f
ic
ie
nc
y
a
nd
r
e
li
a
bi
li
ty
.
A
m
a
jo
r
pr
obl
e
m
is
th
e
ne
e
d
f
or
qui
c
k
a
c
c
e
s
s
to
m
e
ta
da
ta
,
w
hi
c
h
is
e
s
s
e
nt
ia
l
f
or
or
ga
ni
z
in
g
a
nd
r
e
tr
ie
vi
ng
la
r
ge
da
ta
s
e
ts
.
W
it
h
m
or
e
th
a
n
260
bi
ll
io
n
s
to
r
e
d
phot
os
in
H
a
y
s
ta
c
k
,
th
e
di
r
e
c
to
r
y
m
us
t
h
a
ndl
e
a
nd
in
de
x
th
is
d
a
ta
w
e
ll
to
a
ll
ow
f
a
s
t
a
c
c
e
s
s
,
r
e
duc
in
g
de
la
ys
a
nd
boos
ti
ng
pe
r
f
or
m
a
nc
e
.
A
ls
o,
ke
e
pi
ng
da
ta
c
ons
is
te
nt
a
nd
e
ns
ur
in
g
in
te
gr
it
y
a
c
r
os
s
m
a
ny
phys
ic
a
l
a
nd l
ogi
c
a
l
vol
um
e
s
g
e
ts
t
r
ic
ki
e
r
a
s
di
r
e
c
to
r
y s
iz
e
s
gr
o
w
;
s
ys
te
m
f
a
il
ur
e
s
c
a
n t
hr
e
a
t
e
n t
hi
s
i
nt
e
gr
it
y i
f
s
tr
ong
ba
c
kup
m
e
th
ods
a
r
e
not
in
pl
a
c
e
[
9]
.
M
or
e
ove
r
,
th
e
c
ha
ngi
ng
na
tu
r
e
of
us
e
r
a
c
ti
on
s
,
s
u
c
h
a
s
upl
oa
ds
a
nd
de
le
ti
ons
,
r
e
qui
r
e
s
ongoing
upda
te
s
to
th
e
di
r
e
c
to
r
y,
m
a
ki
n
g
m
a
na
ge
m
e
nt
m
or
e
c
om
pl
ic
a
te
d
a
nd
n
e
e
di
ng
a
dva
nc
e
d a
lg
or
it
hm
s
t
o m
a
in
ta
in
l
oa
d a
nd pe
r
f
or
m
a
nc
e
[
23]
.
4.5.
I
n
n
ovat
io
n
s
i
n
d
ir
e
c
t
or
y s
e
r
vi
c
e
s
w
it
h
in
H
ays
t
ac
k
T
he
im
pr
ove
m
e
nt
s
in
di
r
e
c
to
r
y
s
e
r
vi
c
e
s
in
th
e
H
a
ys
ta
c
k
a
r
c
hi
te
c
tu
r
e
gr
e
a
tl
y
im
pr
ove
how
phot
o
s
to
r
a
ge
is
m
a
na
ge
d
a
t
F
a
c
e
book.
T
he
di
r
e
c
to
r
y,
w
hi
c
h
is
c
r
uc
ia
l
f
or
f
in
di
ng
a
nd
r
e
tr
ie
vi
ng
im
a
ge
s
,
is
c
a
r
e
f
ul
ly
c
r
e
a
te
d
to
r
e
duc
e
la
te
nc
y
a
nd
in
c
r
e
a
s
e
th
r
oughput.
B
y
us
in
g
a
di
s
tr
ib
ut
e
d
s
e
tu
p,
th
e
di
r
e
c
to
r
y
s
e
r
vi
c
e
m
a
na
ge
s
m
il
li
ons
of
photos
w
hi
le
pr
ovi
di
ng
r
e
dunda
nc
y a
c
r
os
s
phys
ic
a
l
vol
um
e
s
t
o pr
e
ve
nt
da
ta
l
os
s
,
a
ke
y
poi
nt
in
th
e
H
a
y
s
ta
c
k
m
ode
l.
A
ddi
ti
ona
ll
y,
th
e
di
r
e
c
to
r
y
in
c
lu
de
s
ba
c
kgr
ound
ta
s
ks
th
a
t
s
uppor
t
e
a
r
ly
de
te
c
ti
on
a
nd
f
ix
in
g
of
f
a
il
ur
e
s
,
w
hi
c
h
e
nha
nc
e
s
th
e
s
tr
e
ngt
h
of
th
e
w
hol
e
s
ys
te
m
[
9]
.
T
hi
s
a
r
c
hi
te
c
tu
r
e
'
s
a
bi
li
ty
to
ba
la
nc
e
r
e
a
d
a
nd
w
r
it
e
ta
s
ks
not
onl
y
m
a
k
e
s
it
e
a
s
ie
r
to
a
c
c
e
s
s
s
to
r
e
d
im
a
g
e
s
but
a
ls
o
he
lp
s
w
it
h
c
os
t
-
e
f
f
e
c
ti
ve
gr
ow
th
,
m
e
e
ti
ng
th
e
n
e
e
ds
of
a
n
e
xpa
ndi
ng
u
s
e
r
ba
s
e
[
11]
.
T
h
e
r
e
f
or
e
,
th
e
ne
w
m
e
th
od
us
e
d
in
H
a
ys
ta
c
k
’
s
di
r
e
c
to
r
y s
e
r
vi
c
e
s
s
how
s
a
dv
a
nc
e
d t
e
c
hni
qu
e
s
i
n t
he
de
s
ig
n of
di
s
tr
ib
ut
e
d s
ys
te
m
s
.
5.
C
A
C
H
I
N
G
S
T
R
A
T
E
G
I
E
S
I
N
H
A
Y
S
T
A
C
K
T
he
u
s
e
of
c
a
c
hi
ng
m
e
th
ods
in
th
e
H
a
ys
ta
c
k
a
r
c
hi
te
c
tu
r
e
i
s
e
s
s
e
nt
ia
l
f
or
im
pr
ovi
ng
pe
r
f
or
m
a
nc
e
a
nd
us
e
r
e
xpe
r
ie
nc
e
in
F
a
c
e
book'
s
la
r
ge
phot
o
s
to
r
a
ge
s
ys
te
m
.
B
y
us
in
g
a
di
s
tr
ib
ut
e
d
c
a
c
he
,
th
e
a
r
c
hi
te
c
tu
r
e
de
c
r
e
a
s
e
s
th
e
num
be
r
of
d
is
k
ope
r
a
ti
ons
,
w
hi
c
h
he
lp
s
r
e
duc
e
d
e
la
ys
f
r
om
m
e
ta
da
ta
pr
oc
e
s
s
in
g,
a
s
m
e
nt
io
ne
d
in
[
9]
. T
hi
s
de
c
r
e
a
s
e
i
s
i
m
por
ta
nt
, c
ons
id
e
r
in
g t
he
hi
gh numbe
r
of
photo r
e
tr
ie
va
l
r
e
que
s
ts
t
ha
t
H
a
ys
ta
c
k
d
e
a
ls
w
it
h
e
a
c
h
d
a
y. T
he
a
ddi
ti
on
of
a
c
a
c
he
not
onl
y
a
ll
ow
s
qui
c
k
a
c
c
e
s
s
to
popula
r
im
a
ge
s
but
a
ls
o
he
lp
s
m
a
na
ge
r
e
a
d
a
nd
w
r
it
e
ope
r
a
ti
ons
th
r
oughout
th
e
s
ys
te
m
.
M
or
e
ove
r
,
th
e
de
s
ig
n
of
th
e
a
r
c
hi
te
c
tu
r
e
f
oc
us
e
s
on
c
os
t
-
e
f
f
e
c
ti
ve
ne
s
s
w
hi
le
e
n
s
ur
in
g
r
e
li
a
bi
li
ty
,
hi
ghl
ig
ht
in
g
it
s
e
f
f
ic
ie
nc
y.
O
ve
r
a
ll
,
th
e
c
a
c
hi
ng
m
e
th
od
s
in
H
a
ys
ta
c
k
r
e
pr
e
s
e
nt
a
s
m
a
r
t
w
a
y
to
m
a
na
ge
da
ta
,
w
hi
c
h
is
vi
ta
l
f
or
F
a
c
e
book'
s
goa
l
of
pr
ovi
di
ng
s
m
oot
h
phot
o a
c
c
e
s
s
on a
l
a
r
ge
s
c
a
le
[
24]
, [
25]
.
5.1.
I
m
p
or
t
an
c
e
o
f
c
ac
h
in
g i
n
d
at
a r
e
t
r
ie
val
I
n
da
ta
r
e
tr
ie
va
l
s
ys
te
m
s
th
a
t
a
r
e
c
om
pl
e
x,
c
a
c
hi
ng
is
a
n
im
por
ta
nt
to
ol
th
a
t
he
lp
s
w
it
h
pe
r
f
or
m
a
nc
e
a
nd
e
f
f
ic
ie
nc
y.
C
a
c
hi
ng
c
ut
s
do
w
n
on
a
c
c
e
s
s
de
la
y
s
by
te
m
por
a
r
il
y
s
a
vi
ng
da
ta
th
a
t
is
a
s
k
e
d
f
or
of
te
n,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
3
,
S
e
pt
e
m
be
r
20
25
:
671
-
681
678
r
e
duc
in
g
th
e
lo
a
d
on
m
a
in
s
to
r
a
ge
s
y
s
te
m
s
.
T
hi
s
c
a
n
be
s
e
e
n
in
s
ys
te
m
s
li
ke
F
a
c
e
book'
s
H
a
y
s
ta
c
k
,
w
hi
c
h
ne
e
ds
f
a
s
t
a
nd
d
e
pe
nda
bl
e
a
c
c
e
s
s
to
a
va
s
t
num
be
r
of
im
a
ge
s
.
B
y
us
in
g
a
di
s
tr
ib
ut
e
d
c
a
c
h
e
,
H
a
y
s
ta
c
k
ge
ts
hi
gh
pe
r
f
or
m
a
nc
e
w
hi
le
ke
e
pi
ng
lo
w
de
la
ys
,
e
f
f
e
c
ti
ve
ly
m
a
na
gi
ng
th
e
im
por
ta
nt
r
e
a
d
a
nd
w
r
it
e
ta
s
ks
w
hi
le
m
a
ki
ng
s
ur
e
us
e
r
s
c
a
n
a
c
c
e
s
s
c
ont
e
nt
[
9]
.
A
ddi
ti
ona
ll
y,
c
a
c
hi
n
g
pl
a
ys
a
bi
g
r
ol
e
in
le
s
s
e
ni
ng
th
e
to
ta
l
lo
a
d
on
s
to
r
a
ge
s
ys
te
m
s
be
hi
nd
th
e
s
c
e
n
e
s
,
w
hi
c
h
le
ts
ta
s
k
s
li
ke
m
e
ta
da
ta
ha
ndl
in
g
ha
ppe
n
m
or
e
qui
c
kl
y
[
11]
.
T
he
r
e
f
or
e
,
good
c
a
c
hi
ng
m
e
th
ods
not
onl
y
s
p
e
e
d
up
da
ta
r
e
tr
i
e
va
l
but
a
ls
o
im
pr
ove
how
s
c
a
l
a
bl
e
,
r
e
li
a
bl
e
,
a
nd f
a
ul
t
-
to
le
r
a
nt
m
ode
r
n di
s
tr
ib
ut
e
d s
ys
te
m
s
a
r
e
.
5.2.
T
yp
e
s
of
c
ac
h
in
g m
e
c
h
an
is
m
s
u
s
e
d
C
a
c
hi
ng
m
e
th
od
s
a
r
e
im
por
ta
nt
f
or
im
pr
ovi
ng
how
qui
c
kl
y
d
a
ta
c
a
n
be
r
e
tr
ie
ve
d
in
s
y
s
te
m
s
li
ke
F
a
c
e
book’
s
H
a
ys
ta
c
k
,
w
hi
c
h
ne
e
d
s
f
a
s
t
a
c
c
e
s
s
to
s
to
r
e
d
im
a
ge
s
.
O
ne
ty
pe
of
c
a
c
hi
ng
,
c
a
ll
e
d
di
s
tr
ib
ut
e
d
cach
in
g
,
he
lp
s
de
c
r
e
a
s
e
de
la
y
s
by
ke
e
pi
ng
of
te
n
-
us
e
d
d
a
ta
o
n
m
a
ny
node
s
,
m
a
ki
ng
us
e
r
r
e
que
s
t
s
qui
c
ke
r
.
A
ls
o, i
n s
e
tu
ps
s
im
il
a
r
t
o t
hos
e
m
e
nt
io
ne
d i
n
[
20]
, c
om
bi
ni
ng both
dyna
m
ic
r
a
ndom
-
a
c
c
e
s
s
m
e
m
or
y (
D
R
A
M
)
a
nd
obj
e
c
t
s
to
r
a
ge
in
c
a
c
hi
ng
s
ys
te
m
s
he
lp
s
b
a
la
nc
e
s
pe
e
d
a
nd
c
os
t.
T
h
e
s
e
ki
nd
s
of
hybr
id
c
a
c
hi
ng
a
ppr
oa
c
he
s
s
how
how
d
a
ta
c
a
n
be
m
a
na
g
e
d
w
e
ll
to
a
dj
u
s
t
to
c
ha
ngi
ng
de
m
a
nd,
e
m
pha
s
iz
in
g
th
e
ne
e
d
f
or
f
le
xi
bl
e
c
a
c
he
s
iz
e
a
nd
ty
pe
.
A
s
c
lo
ud
a
ppl
ic
a
ti
ons
c
h
a
nge
of
te
n,
ne
w
te
c
hni
que
s
li
ke
m
e
m
or
y
opt
im
iz
a
ti
on,
in
c
lu
di
ng
id
e
a
s
f
r
om
e
vol
ut
io
na
r
y
a
lg
or
it
hm
s
m
e
nt
io
ne
d
in
[
21]
,
s
how
pos
s
ib
il
it
ie
s
f
or
be
tt
e
r
r
e
s
our
c
e
m
a
na
ge
m
e
nt
w
it
hout
hur
ti
ng s
ys
te
m
pe
r
f
or
m
a
nc
e
. T
he
s
e
m
e
th
o
ds
a
r
e
ke
y t
o ke
e
pi
ng l
a
r
ge
-
s
c
a
le
da
ta
s
ys
te
m
s
w
or
ki
ng w
e
ll
a
nd a
c
c
e
s
s
ib
le
w
hi
le
de
a
li
ng w
it
h t
he
c
om
pl
e
xi
ti
e
s
of
m
ode
r
n da
ta
s
tr
uc
tu
r
e
s
.
5.3.
C
ac
h
e
c
on
s
is
t
e
n
c
y an
d
i
n
val
id
at
io
n
t
e
c
h
n
i
q
u
e
s
I
n
s
ys
te
m
s
th
a
t
s
to
r
e
da
ta
a
c
r
os
s
m
a
ny
lo
c
a
ti
ons
,
li
ke
H
a
ys
ta
c
k
,
ke
e
pi
ng
c
a
c
he
c
on
s
is
te
nt
a
nd
ha
ndl
in
g
in
va
li
da
ti
on
m
e
th
ods
w
e
ll
is
ve
r
y
im
por
ta
nt
f
or
da
t
a
c
or
r
e
c
tn
e
s
s
a
nd
pe
r
f
or
m
a
nc
e
.
W
he
n
m
a
ny
c
li
e
nt
s
us
e
th
e
s
a
m
e
da
ta
a
t
th
e
s
a
m
e
ti
m
e
,
pr
obl
e
m
s
c
a
n
c
om
e
up,
w
hi
c
h
m
a
y
r
e
s
ul
t
in
out
da
te
d
or
in
c
or
r
e
c
t
da
ta
be
in
g
s
how
n.
T
o
s
ol
ve
th
is
,
m
e
th
ods
s
uc
h
a
s
w
r
it
e
-
th
r
ough
a
nd
w
r
it
e
-
ba
c
k
c
a
c
hi
ng
c
a
n
be
us
e
d,
w
it
h
e
a
c
h
ha
vi
ng
di
s
ti
nc
t
pr
os
a
nd
c
ons
r
e
la
te
d
to
s
pe
e
d
a
nd
th
e
u
s
e
of
r
e
s
our
c
e
s
.
A
ls
o,
th
e
r
e
a
r
e
s
tr
a
te
gi
e
s
f
or
in
va
li
da
ti
on
th
a
t
a
r
e
e
it
he
r
pr
oa
c
ti
ve
or
r
e
a
c
ti
ve
,
w
hi
c
h
a
r
e
vi
ta
l
f
or
m
a
na
gi
ng
th
e
c
a
c
he
.
P
r
oa
c
ti
ve
in
va
li
da
ti
on
upda
te
s
or
de
le
te
s
c
a
c
h
e
d
da
ta
a
he
a
d
of
ti
m
e
ba
s
e
d
on
s
e
t
r
ul
e
s
,
w
h
e
r
e
a
s
r
e
a
c
ti
ve
s
tr
a
te
gi
e
s
r
e
ly
on
a
le
r
ts
th
a
t
oc
c
ur
w
he
n
da
ta
c
h
a
nge
s
,
th
u
s
he
lp
in
g
w
it
h
ba
n
dw
id
th
us
e
a
nd
lo
w
e
r
in
g
th
e
c
os
t
s
of
ke
e
pi
ng
c
a
c
he
c
on
s
is
te
nt
.
B
ui
ld
in
g
a
s
tr
ong
c
a
c
hi
ng
s
y
s
te
m
,
li
ke
th
e
o
ne
in
H
a
ys
ta
c
k
,
not
onl
y
boos
t
s
pe
r
f
or
m
a
nc
e
a
nd
r
e
duc
e
s
de
la
y
but
a
ls
o
he
lp
s
ta
c
kl
e
th
e
is
s
ue
s
th
a
t
c
om
e
w
it
h
e
ns
ur
in
g
r
e
li
a
bl
e
a
c
c
e
s
s
to
la
r
ge
da
ta
c
ol
le
c
ti
ons
, ba
c
ki
ng t
he
s
ys
te
m
’
s
m
a
in
a
im
of
be
in
g e
f
f
ic
ie
nt
a
n
d s
c
a
la
bl
e
[
11]
.
5.4.
I
m
p
ac
t
of
c
ac
h
in
g on
s
ys
t
e
m
p
e
r
f
or
m
an
c
e
C
a
c
hi
ng
m
e
th
od
s
a
r
e
ve
r
y
im
por
ta
nt
f
or
m
a
ki
ng
s
y
s
te
m
s
r
un
be
tt
e
r
by
lo
w
e
r
in
g
d
e
la
y
a
nd
us
in
g
r
e
s
our
c
e
s
m
or
e
e
f
f
ic
ie
nt
ly
.
B
y
ke
e
pi
ng
da
ta
th
a
t
is
a
c
c
e
s
s
e
d
o
f
te
n
in
a
qui
c
k
-
a
c
c
e
s
s
c
a
c
h
e
,
a
ppl
ic
a
ti
ons
c
a
n
c
ut
dow
n
on
th
e
e
xt
r
a
w
or
k
th
a
t
c
om
e
s
w
it
h
r
e
a
di
ng
f
r
om
di
s
ks
a
nd
boos
t
ove
r
a
ll
pe
r
f
or
m
a
nc
e
.
F
or
e
xa
m
pl
e
,
in
F
a
c
e
book’
s
H
a
ys
ta
c
k
s
e
tu
p,
a
ddi
ng
a
di
s
tr
ib
ut
e
d
c
a
c
he
he
lp
e
d
th
e
s
ys
te
m
s
e
nd
out
ove
r
260
bi
ll
io
n
phot
os
w
it
h
li
tt
le
de
la
y,
f
ix
in
g
bi
g
pr
obl
e
m
s
f
ound
in
ol
de
r
N
A
S
.
T
hi
s
s
m
a
r
t
c
a
c
hi
ng
u
s
e
not
onl
y
s
pe
e
ds
up
da
t
a
a
c
c
e
s
s
but
a
ls
o
r
e
duc
e
s
pr
e
s
s
ur
e
on
ba
c
ke
nd
s
to
r
a
ge
to
ol
s
,
le
a
di
ng
to
be
tt
e
r
r
e
s
pons
e
ti
m
e
s
a
nd
us
e
r
s
a
ti
s
f
a
c
ti
on
[
9]
.
F
ur
th
e
r
m
or
e
,
a
good
c
a
c
hi
ng
pl
a
n
he
lp
s
c
r
e
a
te
a
be
tt
e
r
ba
la
nc
e
be
twe
e
n
r
e
a
di
ng
a
nd
w
r
it
in
g
da
ta
,
w
hi
c
h
in
c
r
e
a
s
e
s
r
e
li
a
bi
li
ty
a
nd
ope
r
a
ti
ona
l
e
f
f
e
c
ti
ve
ne
s
s
[
13]
.
T
he
r
e
f
or
e
,
us
in
g
c
a
c
hi
ng
is
k
e
y
f
or
s
tr
ong
da
ta
ha
ndl
in
g i
n l
a
r
ge
, s
pr
e
a
d
-
out
s
ys
t
e
m
s
.
5.5.
F
u
t
u
r
e
t
r
e
n
d
s
i
n
c
ac
h
in
g s
t
r
at
e
gi
e
s
f
or
H
ays
t
ac
k
T
he
ne
e
d
f
or
e
f
f
e
c
ti
ve
da
ta
a
c
c
e
s
s
i
s
gr
ow
in
g,
a
nd
f
ut
ur
e
tr
e
nds
in
c
a
c
hi
ng
f
or
H
a
ys
ta
c
k
ne
e
d
to
c
ons
id
e
r
th
e
c
ha
ngi
ng
s
to
r
a
ge
s
ol
ut
io
n
la
nds
c
a
pe
.
O
ne
us
e
f
ul
a
ppr
oa
c
h
is
us
in
g
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
a
nd
m
a
c
hi
ne
le
a
r
ni
ng
to
im
pr
ove
c
a
c
he
m
a
na
ge
m
e
nt
.
T
hi
s
w
oul
d
l
e
t
us
a
na
ly
z
e
da
ta
a
c
c
e
s
s
pa
tt
e
r
ns
in
r
e
a
l
ti
m
e
a
nd
f
il
l
c
a
c
he
s
pr
oa
c
ti
ve
ly
.
T
hi
s
ki
nd
of
pr
e
di
c
ti
ve
c
a
c
hi
ng
c
ou
ld
lo
w
e
r
la
te
nc
y
a
nd
boo
s
t
th
r
oughput,
w
hi
c
h
f
it
s
w
e
ll
w
it
h
H
a
ys
ta
c
k
’
s
c
ur
r
e
nt
de
s
ig
n
th
a
t
s
uppor
ts
bi
ll
i
ons
of
phot
os
w
hi
le
c
ut
ti
ng
dow
n
on
di
s
k
ope
r
a
ti
ons
a
nd
m
e
ta
da
ta
ha
ndl
in
g.
I
n
a
ddi
ti
on,
us
in
g
a
dva
nc
e
d
di
s
tr
ib
ut
e
d
c
a
c
hi
ng
m
e
th
ods
,
li
ke
m
ul
ti
-
ti
e
r
a
r
c
hi
te
c
tu
r
e
s
,
c
a
n
he
lp
s
pr
e
a
d
out
th
e
lo
a
d
a
m
ong
nod
e
s
,
m
a
ki
ng
th
e
s
y
s
te
m
m
or
e
f
a
ul
t
-
to
le
r
a
nt
a
nd
r
e
s
il
ie
nt
to
ha
r
dw
a
r
e
is
s
ue
s
[
9]
.
L
a
s
tl
y,
lo
oki
ng
in
to
e
dge
c
om
put
in
g
c
oul
d
e
nha
nc
e
H
a
ys
ta
c
k
'
s
a
bi
li
ti
e
s
by
pl
a
c
in
g
c
a
c
he
d r
e
s
our
c
e
s
ne
a
r
e
r
t
o us
e
r
s
f
or
qui
c
ke
r
da
ta
a
c
c
e
s
s
, l
e
a
di
n
g t
o a
m
or
e
e
f
f
ic
ie
nt
s
ys
te
m
ove
r
a
ll
[
26]
.
6.
C
O
N
C
L
U
S
I
O
N
I
n
f
in
is
hi
ng
th
e
lo
ok
a
t
F
a
c
e
book'
s
H
a
ys
ta
c
k
-
ba
s
e
d
da
ta
s
to
r
a
ge
s
ys
te
m
,
it
is
c
le
a
r
th
a
t
th
is
ne
w
s
e
tu
p
m
a
r
ks
a
bi
g
s
te
p
f
or
w
a
r
d
in
m
a
na
gi
ng
d
a
ta
a
t
s
c
a
le
.
T
he
a
r
c
hi
te
c
tu
r
e
s
uc
c
e
s
s
f
ul
ly
de
a
ls
w
it
h
th
e
is
s
u
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
dv A
ppl
S
c
i
I
S
S
N
:
2252
-
8814
H
ay
s
ta
c
k
-
bas
e
d F
ac
e
book
’
s
dat
a s
to
r
age
ar
c
hi
te
c
tu
r
e
:
s
to
r
e
, di
r
e
c
to
r
y
, and
c
ac
he
(
T
ol
e
Sut
ik
no)
679
c
a
us
e
d
by
th
e
huge
a
m
ount
of
u
s
e
r
-
c
r
e
a
te
d
c
ont
e
nt
,
pa
r
ti
c
ul
a
r
ly
in
r
e
la
ti
on
to
s
to
r
in
g
a
nd
f
in
di
ng
phot
os
.
B
y
s
pr
e
a
di
ng
s
to
r
a
ge
ove
r
di
f
f
e
r
e
nt
phys
ic
a
l
vol
um
e
s
a
nd
us
in
g
a
s
tr
ong
c
a
c
hi
ng
s
y
s
te
m
,
H
a
ys
ta
c
k
r
e
duc
e
s
m
e
ta
da
ta
lo
a
d
a
nd
boos
t
s
th
r
oughput,
w
hi
c
h
he
lp
s
in
a
c
c
e
s
s
in
g
bi
ll
io
ns
of
phot
os
.
A
ls
o,
th
e
us
e
of
pr
oa
c
ti
ve
ba
c
kgr
ound
ta
s
k
s
to
c
he
c
k a
nd
f
ix
s
ys
te
m
f
a
il
ur
e
s
s
how
s
th
e
a
r
c
hi
te
c
tu
r
e
'
s
f
oc
us
on
be
in
g
r
e
li
a
bl
e
a
nd
a
bl
e
to
ha
ndl
e
f
a
ul
ts
. T
he
s
e
f
e
a
tu
r
e
s
not
onl
y i
m
pr
ove
how
w
e
ll
i
t
w
or
ks
but
a
ls
o he
lp
ke
e
p c
o
s
ts
dow
n, w
hi
c
h i
s
pa
r
t
of
a
bi
gge
r
tr
e
nd
in
c
lo
ud
c
om
put
in
g
to
w
a
r
d
us
in
g
c
he
a
p,
r
e
gul
a
r
ha
r
dw
a
r
e
.
I
n
th
e
e
nd,
H
a
ys
ta
c
k
s
ta
nds
out
a
s
a
s
tr
ong
e
x
a
m
pl
e
of
us
in
g
s
m
a
r
t
a
r
c
hi
te
c
tu
r
a
l
m
e
th
ods
to
s
a
ti
s
f
y
th
e
c
ha
ngi
ng
ne
e
d
s
of
a
ppl
ic
a
ti
ons
th
a
t
us
e
a
lo
t
of
da
ta
.
F
a
c
e
book'
s
H
a
ys
ta
c
k
da
ta
s
to
r
a
ge
s
ys
te
m
i
s
a
gr
oundbr
e
a
ki
ng
s
ol
ut
io
n
f
or
m
a
na
gi
ng
la
r
ge
a
m
ount
s
of
di
gi
ta
l
m
e
di
a
.
I
ts
de
s
ig
n
c
om
bi
ne
s
s
to
r
a
ge
,
di
r
e
c
t
or
y,
a
nd
c
a
c
he
pa
r
ts
,
r
e
s
ul
ti
ng
in
hi
ghe
r
da
ta
f
lo
w
a
nd
le
s
s
w
a
it
in
g
ti
m
e
f
or
a
c
c
e
s
s
in
g
phot
os
.
T
he
s
ys
t
e
m
r
e
duc
e
s
di
s
k
us
e
a
nd
im
pr
ove
s
m
e
ta
da
ta
ha
ndl
in
g,
a
ddr
e
s
s
in
g
de
la
ys
in
ol
de
r
ne
twor
k
-
a
tt
a
c
he
d
s
to
r
a
g
e
s
ys
te
m
s
.
T
he
s
y
s
te
m
a
ls
o
e
m
pl
oys
r
obus
t
ba
c
kup
m
e
th
ods
to
e
ns
ur
e
da
ta
s
a
f
e
ty
,
r
e
gul
a
r
ly
c
opyi
ng
m
il
li
ons
of
phot
os
to
pr
e
ve
nt
ha
r
dw
a
r
e
is
s
ue
s
.
T
he
H
a
ys
ta
c
k
a
r
c
hi
te
c
tu
r
e
s
ig
ni
f
ic
a
nt
ly
e
nha
n
c
e
s
di
s
tr
ib
ut
e
d
s
y
s
te
m
s
d
e
s
ig
n,
s
e
tt
in
g
a
s
ta
nd
a
r
d
f
or
f
ut
ur
e
a
dva
nc
e
m
e
nt
s
in
da
t
a
s
to
r
a
ge
m
e
th
od
s
.
I
t
out
pe
r
f
or
m
s
ol
d
m
e
th
ods
li
ke
N
A
S
,
w
hi
c
h
c
a
n
f
a
c
e
s
lo
w
dow
ns
a
nd
is
s
ue
s
unde
r
he
a
vy
us
e
.
T
h
e
c
om
bi
na
ti
on
of
th
e
H
a
ys
ta
c
k
s
to
r
e
,
di
r
e
c
to
r
y,
a
nd
c
a
c
he
m
a
ke
s
phot
o
r
e
tr
ie
va
l
e
a
s
ie
r
, i
nc
r
e
a
s
in
g s
pe
e
d a
nd l
ow
e
r
in
g de
la
ys
. T
hi
s
s
ys
t
e
m
m
e
e
ts
t
oda
y'
s
r
e
qui
r
e
m
e
nt
s
f
or
qui
c
k da
ta
a
c
c
e
s
s
i
n
c
lo
ud
c
om
put
in
g,
hi
ghl
ig
ht
in
g
th
e
ne
e
d
f
or
f
le
xi
bl
e
c
a
c
hi
ng
m
e
th
ods
.
H
ow
e
ve
r
,
th
e
r
e
a
r
e
s
hor
tc
om
in
gs
in
c
ur
r
e
nt
r
e
s
e
a
r
c
h,
s
u
c
h
a
s
th
e
la
c
k
of
r
e
a
l
-
w
or
ld
te
s
ti
ng
on
th
e
s
ys
te
m
'
s
s
c
a
la
bi
li
ty
unde
r
he
a
vy
u
s
a
ge
a
nd
th
e
c
om
pl
e
xi
ti
e
s
of
va
r
io
us
da
ta
ty
pe
s
a
nd
a
c
c
e
s
s
be
ha
vi
or
s
in
a
f
a
s
t
-
c
ha
ngi
ng
s
o
c
ia
l
m
e
di
a
e
nvi
r
onm
e
nt
.
F
ut
ur
e
s
tu
di
e
s
s
houl
d
e
xpl
or
e
be
tt
e
r
de
s
ig
ns
f
or
bi
g
s
to
r
a
ge
s
ys
te
m
s
,
f
oc
us
in
g
on
ne
w
m
e
th
ods
th
a
t
r
e
duc
e
m
e
ta
da
ta
pr
oc
e
s
s
in
g, a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
f
or
f
a
ul
t
de
te
c
ti
on a
nd r
e
pa
ir
s
,
a
nd t
he
c
os
t
e
f
f
e
c
ts
of
di
s
tr
ib
ut
e
d c
a
c
h
e
s
.
T
he
s
hi
f
t
f
r
om
ol
d
da
ta
s
to
r
a
ge
m
ode
ls
to
ne
w
s
ys
t
e
m
s
hi
ghl
ig
ht
s
th
e
ne
e
d
f
or
f
le
xi
bl
e
s
ol
ut
io
ns
th
a
t
c
a
n
s
ol
ve
to
da
y'
s
s
to
r
a
ge
pr
obl
e
m
s
a
nd
pr
e
pa
r
e
f
or
f
ut
ur
e
r
e
qui
r
e
m
e
nt
s
a
s
da
ta
gr
ow
s
.
F
oc
us
in
g
on
ba
c
kup
m
e
a
s
ur
e
s
a
nd
pr
e
ve
nt
iv
e
upke
e
p
in
th
e
s
e
s
ys
te
m
s
h
e
lp
s
pr
ot
e
c
t
a
g
a
in
s
t
ha
r
dw
a
r
e
is
s
ue
s
,
e
ns
ur
in
g
da
t
a
r
e
li
a
bi
li
ty
a
nd
a
va
il
a
bi
li
ty
i
n m
or
e
c
om
pl
e
x di
gi
ta
l
e
nvi
r
onm
e
nt
s
.
A
C
K
N
O
WL
E
D
G
M
E
N
T
S
T
he
a
ut
hor
s
w
oul
d
li
ke
to
e
xpr
e
s
s
th
e
ir
gr
a
ti
tu
de
to
U
ni
ve
r
s
it
a
s
A
hm
a
d
D
a
hl
a
n
(
U
A
D
)
,
th
e
E
m
be
dde
d
S
ys
te
m
a
nd
P
ow
e
r
E
le
c
tr
oni
c
s
R
e
s
e
a
r
c
h
G
r
ou
p
(
E
S
P
E
R
G
)
,
U
ni
ve
r
s
it
a
s
P
a
le
m
ba
ng,
a
nd
I
nt
e
r
na
ti
ona
l
I
s
la
m
ic
U
ni
ve
r
s
it
y
M
a
la
y
s
ia
(
I
I
U
M
)
f
or
th
e
ir
in
va
lu
a
bl
e
s
uppor
t
a
nd
c
ont
r
ib
ut
io
ns
to
th
is
r
e
s
e
a
r
c
h. T
he
f
a
c
il
it
ie
s
, gui
da
nc
e
, a
nd c
ol
la
bor
a
ti
ve
e
nvi
r
onm
e
nt
pr
ovi
de
d by the
s
e
i
ns
ti
tu
ti
ons
w
e
r
e
e
s
s
e
nt
ia
l
to
t
he
s
uc
c
e
s
s
f
ul
c
om
pl
e
ti
on a
nd publi
c
a
ti
on of
t
hi
s
w
or
k.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
T
he
r
e
s
e
a
r
c
h
w
a
s
f
unde
d
by
P
T
.
I
nt
e
le
kt
ua
l
P
us
ta
ka
M
e
di
a
U
ta
m
a
(
I
P
M
U
)
unde
r
c
ont
r
a
c
t
num
be
r
09/
R
S
T
-
E
/I
P
M
U
/I
/2
024, whic
h s
uppor
te
d t
he
f
a
c
il
it
a
ti
on of
t
hi
s
w
or
k.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
T
ol
e
S
ut
ik
no
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
A
hm
a
d H
e
r
ya
nt
o
✓
✓
✓
✓
✓
✓
L
a
ks
a
na
T
a
le
nt
a
A
hm
a
d
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8814
I
nt
J
A
dv A
ppl
S
c
i
,
V
ol
.
14
, N
o.
3
,
S
e
pt
e
m
be
r
20
25
:
671
-
681
680
I
N
F
O
R
M
E
D
C
O
N
S
E
N
T
T
hi
s
s
tu
dy
doe
s
not
in
vol
ve
di
r
e
c
t
in
te
r
a
c
ti
on
w
it
h
in
di
vi
dua
ls
.
T
he
r
e
f
or
e
,
in
f
or
m
e
d
c
ons
e
nt
is
not
ne
c
e
s
s
a
r
y, a
s
t
he
r
e
s
e
a
r
c
h r
e
li
e
s
on publi
c
ly
a
va
il
a
bl
e
da
ta
.
E
T
H
I
C
A
L
A
P
P
R
O
V
A
L
T
hi
s
s
tu
dy
doe
s
not
in
c
lu
de
a
ny
hum
a
n
or
a
ni
m
a
l
pa
r
ti
c
ip
a
n
ts
;
th
e
r
e
f
or
e
,
e
th
ic
a
l
a
ppr
ova
l
is
not
r
e
qui
r
e
d f
or
t
hi
s
a
r
ti
c
le
.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
D
a
ta
a
va
il
a
bi
li
ty
is
not
a
ppl
ic
a
bl
e
to
th
is
pa
pe
r
a
s
no
ne
w
da
ta
w
e
r
e
c
r
e
a
te
d
or
a
na
ly
z
e
d
in
th
is
s
tu
dy.
R
E
F
E
R
E
N
C
E
S
[
1]
S
.
M
ur
a
l
i
dha
r
e
t
al
.
,
“
F
4:
F
a
c
e
book’
s
w
a
r
m
bl
ob
s
t
or
a
ge
s
ys
t
e
m
,”
i
n
P
r
oc
e
e
di
ngs
of
t
he
11t
h
U
SE
N
I
X
Sy
m
pos
i
um
on
O
pe
r
at
i
ng
Sy
s
t
e
m
s
D
e
s
i
gn
and
I
m
pl
e
m
e
nt
at
i
on,
O
SD
I
2014
,
F
a
c
e
book
I
nc
.:
U
S
E
N
I
X
A
s
s
oc
i
a
t
i
on,
2014,
pp.
383
–
398.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
w
w
w
.s
c
opus
.
c
om
/
i
nw
a
r
d/
r
e
c
or
d.ur
i
?
e
i
d=2
-
s
2.0
-
85075667407&
pa
r
t
ne
r
I
D
=40&m
d5=e
9934c
5a
1b372c
b2303f
a
5729c
992512
[
2]
M
.
V
.
B
a
r
be
r
a
,
S
.
B
r
onz
i
ni
,
A
.
M
e
i
,
a
nd
V
.
C
.
P
e
r
t
a
,
“
A
ne
e
dl
e
i
n
t
h
e
ha
ys
t
a
c
k
-
de
l
a
y
ba
s
e
d
us
e
r
i
de
nt
i
f
i
c
a
t
i
on
i
n
c
e
l
l
ul
a
r
ne
t
w
or
ks
,”
i
n
L
e
c
t
ur
e
N
ot
e
s
i
n
C
om
put
e
r
Sc
i
e
nc
e
(
i
nc
l
udi
ng
s
ubs
e
r
i
e
s
L
e
c
t
ur
e
N
ot
e
s
i
n
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
and
L
e
c
t
ur
e
N
ot
e
s
i
n B
i
oi
nf
or
m
at
i
c
s
)
, S
a
pi
e
nz
a
U
ni
ve
r
s
i
t
y, R
om
e
, I
t
a
l
y:
S
pr
i
nge
r
V
e
r
l
a
g, 2014, pp. 265
–
267. doi
:
10.1007/
978
-
3
-
319
-
04918
-
2_27.
[
3]
A
.
R
.
S
æ
t
na
n,
“
T
he
H
a
ys
t
a
c
k
f
a
l
l
a
c
y,
or
w
hy
bi
g
da
t
a
p
r
ovi
de
s
l
i
t
t
l
e
s
e
c
ur
i
t
y,”
i
n
T
he
P
o
l
i
t
i
c
s
of
B
i
g
D
at
a:
B
i
g
D
at
a,
B
i
g
B
r
ot
he
r
?
,
N
or
w
e
gi
a
n
U
ni
v
e
r
s
i
t
y
of
S
c
i
e
nc
e
a
nd
T
e
c
hnol
ogy,
T
r
ondhe
i
m
,
N
or
w
a
y:
T
a
yl
or
a
nd
F
r
a
nc
i
s
,
2018,
pp.
21
–
38.
doi
:
10.4324/
9781315231938
-
2.
[
4]
J
.
M
a
c
C
or
m
i
c
k,
N
i
n
e
al
gor
i
t
hm
s
t
hat
c
hange
d
t
he
f
ut
ur
e
:
t
he
i
nge
ni
ous
i
de
as
t
hat
dr
i
v
e
t
oday
’
S
c
om
put
e
r
s
.
D
i
c
ki
ns
on
C
ol
l
e
ge
,
U
ni
t
e
d S
t
a
t
e
s
:
P
r
i
nc
e
t
on U
ni
ve
r
s
i
t
y P
r
e
s
s
, 2020. doi
:
10.5860/
c
hoi
c
e
.49
-
5106.
[
5]
J
.
Y
i
n,
H
.
Z
hu,
a
nd
P
.
C
.
V
i
nh,
“
F
or
m
a
l
i
z
a
t
i
on
a
nd
a
na
l
ys
i
s
of
H
a
y
s
t
a
c
k
a
r
c
h
i
t
e
c
t
ur
e
f
r
om
pr
oc
e
s
s
a
l
ge
br
a
pe
r
s
pe
c
t
i
ve
,”
M
obi
l
e
N
e
t
w
or
k
s
and A
ppl
i
c
at
i
ons
, vol
. 25, no. 3, pp. 1125
–
1139, 2020, doi
:
10.1007/
s
11036
-
019
-
01433
-
1.
[
6]
D
. B
e
a
ve
r
, S
. K
um
a
r
, H
.
C
.
L
i
, J
. S
obe
l
, a
nd P
. V
a
j
ge
l
,
“
F
i
ndi
ng a
ne
e
dl
e
i
n
H
a
ys
t
a
c
k
:
F
a
c
e
book’
s
phot
o s
t
or
a
ge
,”
i
n
P
r
oc
e
e
di
ng
s
of
t
he
9t
h
U
SE
N
I
X
Sy
m
pos
i
um
on
O
pe
r
at
i
ng
Sy
s
t
e
m
s
D
e
s
i
gn
and
I
m
pl
e
m
e
nt
at
i
on,
O
SD
I
2010
,
F
a
c
e
book
I
nc
.
,
U
ni
t
e
d
S
t
a
t
e
s
:
U
S
E
N
I
X
A
s
s
oc
i
a
t
i
on,
2019,
pp.
47
–
60.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
w
w
w
.s
c
opu
s
.c
om
/
i
nw
a
r
d/
r
e
c
or
d.ur
i
?
e
i
d=2
-
s
2.0
-
85076926134&
pa
r
t
ne
r
I
D
=40&m
d5=c
5b6b2216da
bf
1e
3748392c
69d237b5e
[
7]
S
.
C
om
f
or
t
e
t
al
.
,
“
S
o
r
t
i
ng
t
hr
ough
t
he
s
a
f
e
t
y
da
t
a
H
a
ys
t
a
c
k
:
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
t
o
i
de
nt
i
f
y
i
ndi
v
i
dua
l
c
a
s
e
s
a
f
e
t
y
r
e
por
t
s
i
n
s
oc
i
a
l
-
di
gi
t
a
l
m
e
di
a
,”
D
r
ug Saf
e
t
y
, vol
. 41, no. 6, pp. 579
–
590, 2018, doi
:
10.1
007/
s
40264
-
018
-
0641
-
7.
[
8]
D
.
C
unha
,
N
.
G
ui
m
a
r
ã
e
s
,
a
nd
Á
.
F
i
gue
i
r
a
,
“
A
n
a
r
c
hi
t
e
c
t
ur
e
f
or
a
c
ont
i
nuo
us
a
nd
e
xpl
or
a
t
or
y
a
na
l
ys
i
s
on
s
oc
i
a
l
m
e
di
a
,”
i
n
P
r
oc
e
e
di
ngs
of
t
he
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
s
on
C
om
put
e
r
G
r
aphi
c
s
,
V
i
s
ua
l
i
z
at
i
on,
C
om
put
e
r
V
i
s
i
on
and
I
m
age
P
r
oc
e
s
s
i
ng
2017
and
B
i
g
D
at
a
A
nal
y
t
i
c
s
,
D
at
a
M
i
ni
ng
and
C
o
m
put
at
i
onal
I
nt
e
l
l
i
ge
nc
e
2017
-
P
ar
t
of
t
he
M
ul
t
i
C
onf
e
r
e
n
c
e
on
C
om
put
e
r
Sc
i
e
nc
e
and
I
nf
o
,
R
.
L
.,
X
.
Y
.,
a
nd
A
.
A
.P
.,
E
ds
.,
C
R
A
C
S
/
I
N
E
S
C
T
E
C
a
nd
U
ni
ve
r
s
i
da
de
D
o
P
or
t
o,
R
ua
do
C
a
m
po
A
l
e
gr
e
1021/
1055,
P
or
t
o,
4169
-
007,
P
o
r
t
uga
l
:
I
A
D
I
S
,
2017,
pp.
339
–
342.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
w
w
w
.s
c
opus
.
c
om
/
i
nw
a
r
d/
r
e
c
or
d.ur
i
?
e
i
d=2
-
s
2.0
-
85040191575&
pa
r
t
ne
r
I
D
=40&m
d5=bf
e
7796f
7223a
45e
3bc
65df
f
1f
a
5a
929
[
9]
S
hi
va
ng, “
F
a
c
e
book’
s
phot
o s
t
or
a
ge
a
r
c
hi
t
e
c
t
ur
e
.”
2024. [
O
nl
i
ne
]
. A
va
i
l
a
bl
e
:
h
t
t
ps
:
/
/
s
hor
t
ur
l
.a
t
/
8X
V
z
Y
[
10]
T
.
N
.
M
our
a
t
i
di
s
,
“
O
pt
i
m
i
z
i
ng
t
he
r
e
c
ove
r
y
of
da
t
a
c
ons
i
s
t
e
nc
y
gos
s
i
p
a
l
gor
i
t
hm
s
on
di
s
t
r
i
but
e
d
obj
e
c
t
-
s
t
or
e
s
ys
t
e
m
s
(
c
e
ph
)
β
ε
λ
τ
ι
σ
τ
οπ
οί
η
σ
η
α
λ
γ
ορ
ί
θ
μ
ου
φ
λ
υ
α
ρ
ί
α
ς
γ
ι
α
σ
υ
ν
ε
π
ή
δε
δομ
έ
ν
α
σ
ε
κ
α
τ
α
ν
ε
μ
η
μ
έ
ν
α
σ
υ
σ
τ
ή
μ
α
τ
α
α
π
οθ
ή
κ
ε
υ
σ
η
ς
α
ν
τ
ι
κ
ε
ι
μ
έ
ν
ων
(
c
e
ph
)
,”
N
a
t
i
ona
l
a
nd K
a
podi
s
t
r
i
a
n U
ni
ve
r
s
i
t
y of
A
t
he
ns
, 2020.
[
11]
L
.
G
e
or
ge
,
H
B
as
e
t
he
de
f
i
ni
t
i
v
e
gui
de
.
r
andom
ac
c
e
s
s
t
o
y
ou
r
pl
an
e
t
-
s
i
z
e
dat
a
.
O
’
R
e
i
l
l
y
M
e
di
a
,
2011.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
books
.googl
e
.c
o.i
d/
books
?
i
d=nU
hi
Q
xU
X
V
pM
C
[
12]
B
.
E
l
l
i
s
,
R
e
al
-
t
i
m
e
anal
y
t
i
c
s
:
t
e
c
hni
que
s
t
o
anal
y
z
e
and
v
i
s
ual
i
z
e
s
t
r
e
am
i
ng
dat
a
.
W
i
l
e
y,
2014.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
books
.googl
e
.c
o.i
d/
books
?
i
d=D
F
nO
A
w
A
A
Q
B
A
J
[
13]
M
.
J
.
F
r
a
nkl
i
n,
C
l
i
e
nt
dat
a
c
ac
hi
ng:
a
f
oundat
i
on
f
o
r
hi
gh
pe
r
f
or
m
anc
e
obj
e
c
t
dat
abas
e
s
y
s
t
e
m
s
.
i
n
T
he
S
pr
i
nge
r
I
nt
e
r
na
t
i
ona
l
S
e
r
i
e
s
i
n
E
ngi
ne
e
r
i
ng
a
nd
C
om
put
e
r
S
c
i
e
nc
e
.
S
pr
i
nge
r
U
S
,
2011.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
books
.googl
e
.c
o.i
d/
books
?
i
d=dr
c
K
s
w
E
A
C
A
A
J
[
14]
M
.
H
.
F
u
e
t
al
.
,
“
S
t
udy
a
nd
i
m
pl
e
m
e
nt
a
t
i
on
of
know
l
e
dge
-
ba
s
e
d
i
nnova
t
i
ve
s
oc
i
a
l
m
e
di
a
s
e
r
vi
c
e
s
,”
I
C
I
C
E
x
pr
e
s
s
L
e
t
t
e
r
s
,
vol
.
9,
no.
3,
pp.
759
–
765,
2015,
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
p
s
:
/
/
w
w
w
.s
c
opu
s
.c
om
/
i
nw
a
r
d/
r
e
c
or
d.ur
i
?
e
i
d=2
-
s
2.0
-
84924067908&
pa
r
t
ne
r
I
D
=40&m
d5=3298373e
92d8299b8e
4c
4230ddb01d88
[
15]
K
.
R
.
K
r
i
s
h,
A
.
A
nw
a
r
,
a
nd
A
.
R
.
B
ut
t
,
“
H
a
t
S
:
a
he
t
e
r
oge
ne
i
t
y
-
a
w
a
r
e
t
i
e
r
e
d
s
t
or
a
ge
f
or
ha
doop,”
i
n
P
r
oc
e
e
di
ngs
-
14t
h
I
E
E
E
/
A
C
M
I
nt
e
r
nat
i
onal
Sy
m
pos
i
um
on
C
l
us
t
e
r
,
C
l
oud,
and
G
r
i
d
C
om
put
i
ng,
C
C
G
r
i
d
2014
,
D
e
p
a
r
t
m
e
nt
of
C
om
put
e
r
S
c
i
e
nc
e
,
V
i
r
gi
ni
a
T
e
c
h, B
l
a
c
ks
bur
g, V
A
, U
ni
t
e
d S
t
a
t
e
s
:
I
E
E
E
C
om
put
e
r
S
oc
i
e
t
y, 2014,
pp. 502
–
511. doi
:
10.1109/
C
C
G
r
i
d.2014.51.
[
16]
P
.
F
om
pe
yr
i
ne
,
“
C
a
c
he
m
ode
l
pl
ugi
n
f
or
m
e
m
or
y
hi
e
r
a
r
c
hy
a
w
a
r
e
pr
ogr
a
m
m
i
ng,”
E
T
H
Z
ur
i
c
h,
2020.
doi
:
ht
t
ps
:
/
/
doi
.or
g/
10.3929/
e
t
hz
-
b
-
000431675.
[
17]
S
.
C
ha
,
R
.
N
.
T
a
yl
or
,
a
nd
K
.
K
a
ng,
H
andbook
of
s
of
t
w
a
r
e
e
ngi
ne
e
r
i
ng
.
C
ha
m
:
S
pr
i
nge
r
I
nt
e
r
na
t
i
ona
l
P
ubl
i
s
hi
ng,
2019.
doi
:
10.1007/
978
-
3
-
030
-
00262
-
6.
[
18]
K
.
D
.
M
c
C
r
a
c
ke
n,
F
l
e
x
i
bl
e
and
r
obus
t
dat
a
s
t
or
age
and
r
e
t
r
i
e
v
al
i
n
t
he
hay
s
t
ac
k
s
y
s
t
e
m
.
M
a
s
s
a
c
hus
e
t
t
s
I
ns
t
i
t
ut
e
of
T
e
c
hnol
ogy
,
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
i
c
a
l
E
ngi
ne
e
r
i
ng
a
nd
C
om
put
e
r
S
c
i
e
nc
e
,
2001.
[
O
nl
i
ne
]
.
A
va
i
l
a
bl
e
:
ht
t
ps
:
/
/
books
.googl
e
.c
o.i
d/
books
?
i
d=0bhL
O
A
A
A
C
A
A
J
[
19]
A
.
S
i
m
e
unovi
c
a
nd
U
.
T
r
i
punovi
c
,
“
M
a
na
gi
ng
m
i
c
r
o
f
r
ont
e
nds
a
c
r
os
s
m
ul
t
i
pl
e
t
e
c
h
s
t
a
c
k
s
-
s
h
a
r
i
ng,
f
i
ndi
ng
a
nd
publ
i
s
hi
ng,”
L
und U
ni
ve
r
s
i
t
y, 2023. [
O
nl
i
ne
]
. A
va
i
l
a
bl
e
:
ht
t
p
s
:
/
/
l
up.l
ub.l
u.s
e
/
s
t
ude
nt
-
pa
pe
r
s
/
s
e
a
r
c
h/
publ
i
c
a
t
i
on/
9128849
Evaluation Warning : The document was created with Spire.PDF for Python.