<s>
In	O
computer	B-General_Concept
science	I-General_Concept
,	O
resource	B-Operating_System
starvation	I-Operating_System
is	O
a	O
problem	O
encountered	O
in	O
concurrent	B-Architecture
computing	I-Architecture
where	O
a	O
process	B-Operating_System
is	O
perpetually	O
denied	O
necessary	O
resources	B-General_Concept
to	O
process	B-Operating_System
its	O
work	O
.	O
</s>
<s>
Starvation	B-Operating_System
may	O
be	O
caused	O
by	O
errors	O
in	O
a	O
scheduling	O
or	O
mutual	B-Operating_System
exclusion	I-Operating_System
algorithm	O
,	O
but	O
can	O
also	O
be	O
caused	O
by	O
resource	O
leaks	O
,	O
and	O
can	O
be	O
intentionally	O
caused	O
via	O
a	O
denial-of-service	O
attack	O
such	O
as	O
a	O
fork	B-Operating_System
bomb	I-Operating_System
.	O
</s>
<s>
When	O
starvation	B-Operating_System
is	O
impossible	O
in	O
a	O
concurrent	B-Architecture
algorithm	I-Architecture
,	O
the	O
algorithm	O
is	O
called	O
starvation-free	O
,	O
lockout-freed	O
or	O
said	O
to	O
have	O
finite	B-Operating_System
bypass	I-Operating_System
.	O
</s>
<s>
This	O
property	O
is	O
an	O
instance	O
of	O
liveness	B-Application
,	O
and	O
is	O
one	O
of	O
the	O
two	O
requirements	O
for	O
any	O
mutual	B-Operating_System
exclusion	I-Operating_System
algorithm	O
;	O
the	O
other	O
being	O
correctness	O
.	O
</s>
<s>
The	O
name	O
"	O
finite	B-Operating_System
bypass	I-Operating_System
"	O
means	O
that	O
any	O
process	B-Operating_System
(	O
concurrent	O
part	O
)	O
of	O
the	O
algorithm	O
is	O
bypassed	O
at	O
most	O
a	O
finite	O
number	O
times	O
before	O
being	O
allowed	O
access	O
to	O
the	O
shared	B-General_Concept
resource	I-General_Concept
.	O
</s>
<s>
Starvation	B-Operating_System
is	O
usually	O
caused	O
by	O
an	O
overly	O
simplistic	O
scheduling	O
algorithm	O
.	O
</s>
<s>
The	O
scheduling	O
algorithm	O
,	O
which	O
is	O
part	O
of	O
the	O
kernel	B-Operating_System
,	O
is	O
supposed	O
to	O
allocate	O
resources	B-General_Concept
equitably	O
;	O
that	O
is	O
,	O
the	O
algorithm	O
should	O
allocate	O
resources	B-General_Concept
so	O
that	O
no	O
process	B-Operating_System
perpetually	O
lacks	O
necessary	O
resources	B-General_Concept
.	O
</s>
<s>
Many	O
operating	O
system	O
schedulers	O
employ	O
the	O
concept	O
of	O
process	B-Operating_System
priority	O
.	O
</s>
<s>
A	O
high	O
priority	O
process	B-Operating_System
A	O
will	O
run	O
before	O
a	O
low	O
priority	O
process	B-Operating_System
B	O
.	O
</s>
<s>
If	O
the	O
high	O
priority	O
process	B-Operating_System
(	O
process	B-Operating_System
A	O
)	O
blocks	O
and	O
never	O
yields	O
,	O
the	O
low	O
priority	O
process	B-Operating_System
(	O
B	O
)	O
will	O
(	O
in	O
some	O
systems	O
)	O
never	O
be	O
scheduled	O
—	O
it	O
will	O
experience	O
starvation	B-Operating_System
.	O
</s>
<s>
If	O
there	O
is	O
an	O
even	O
higher	O
priority	O
process	B-Operating_System
X	O
,	O
which	O
is	O
dependent	O
on	O
a	O
result	O
from	O
process	B-Operating_System
B	O
,	O
then	O
process	B-Operating_System
X	O
might	O
never	O
finish	O
,	O
even	O
though	O
it	O
is	O
the	O
most	O
important	O
process	B-Operating_System
in	O
the	O
system	O
.	O
</s>
<s>
This	O
condition	O
is	O
called	O
a	O
priority	B-Operating_System
inversion	I-Operating_System
.	O
</s>
<s>
Modern	O
scheduling	O
algorithms	O
normally	O
contain	O
code	O
to	O
guarantee	O
that	O
all	O
processes	O
will	O
receive	O
a	O
minimum	O
amount	O
of	O
each	O
important	O
resource	O
(	O
most	O
often	O
CPU	O
time	O
)	O
in	O
order	O
to	O
prevent	O
any	O
process	B-Operating_System
from	O
being	O
subjected	O
to	O
starvation	B-Operating_System
.	O
</s>
<s>
In	O
computer	O
networks	O
,	O
especially	O
wireless	O
networks	O
,	O
scheduling	O
algorithms	O
may	O
suffer	O
from	O
scheduling	B-Operating_System
starvation	I-Operating_System
.	O
</s>
<s>
An	O
example	O
is	O
maximum	B-Algorithm
throughput	I-Algorithm
scheduling	I-Algorithm
.	O
</s>
<s>
Starvation	B-Operating_System
is	O
normally	O
caused	O
by	O
deadlock	B-Operating_System
in	O
that	O
it	O
causes	O
a	O
process	B-Operating_System
to	O
freeze	O
.	O
</s>
<s>
On	O
the	O
other	O
hand	O
,	O
a	O
process	B-Operating_System
is	O
in	O
starvation	B-Operating_System
when	O
it	O
is	O
waiting	O
for	O
a	O
resource	O
that	O
is	O
continuously	O
given	O
to	O
other	O
processes	O
.	O
</s>
<s>
Starvation-freedom	B-Operating_System
is	O
a	O
stronger	O
guarantee	O
than	O
the	O
absence	O
of	O
deadlock	B-Operating_System
:	O
a	O
mutual	B-Operating_System
exclusion	I-Operating_System
algorithm	O
that	O
must	O
choose	O
to	O
allow	O
one	O
of	O
two	O
processes	O
into	O
a	O
critical	B-Operating_System
section	I-Operating_System
and	O
picks	O
one	O
arbitrarily	O
is	O
deadlock-free	O
,	O
but	O
not	O
starvation-free	O
.	O
</s>
<s>
A	O
possible	O
solution	O
to	O
starvation	B-Operating_System
is	O
to	O
use	O
a	O
scheduling	O
algorithm	O
with	O
priority	O
queue	O
that	O
also	O
uses	O
the	O
aging	B-Operating_System
technique	O
.	O
</s>
<s>
Aging	B-Operating_System
is	O
a	O
technique	O
of	O
gradually	O
increasing	O
the	O
priority	O
of	O
processes	O
that	O
wait	O
in	O
the	O
system	O
for	O
a	O
long	O
time	O
.	O
</s>
