<s>
842	B-Algorithm
,	O
8-4-2	O
,	O
or	O
EFT	O
is	O
a	O
data	B-General_Concept
compression	I-General_Concept
algorithm	I-General_Concept
.	O
</s>
<s>
It	O
is	O
a	O
variation	O
on	O
Lempel	B-Algorithm
–	I-Algorithm
Ziv	I-Algorithm
compression	O
with	O
a	O
limited	O
dictionary	O
length	O
.	O
</s>
<s>
With	O
typical	O
data	O
,	O
842	B-Algorithm
gives	O
80	O
to	O
90	O
percent	O
of	O
the	O
compression	O
of	O
LZ77	O
with	O
much	O
faster	O
throughput	O
and	O
less	O
memory	O
use	O
.	O
</s>
<s>
842	B-Algorithm
compression	O
can	O
be	O
used	O
for	O
virtual	B-General_Concept
memory	I-General_Concept
compression	I-General_Concept
,	O
for	O
databases	O
—	O
especially	O
column-oriented	B-General_Concept
stores	I-General_Concept
,	O
and	O
when	O
streaming	O
input-output	O
—	O
for	O
example	O
to	O
do	O
backups	B-Protocol
or	O
to	O
write	O
to	O
log	B-Application
files	I-Application
.	O
</s>
<s>
IBM	O
added	O
hardware	O
accelerators	O
and	O
instructions	O
for	O
842	B-Algorithm
compression	O
to	O
their	O
Power	B-Device
processors	O
from	O
POWER7+	B-Device
onward	O
.	O
</s>
<s>
In	O
addition	O
,	O
POWER9	B-Device
and	O
Power10	B-Operating_System
added	O
hardware	O
acceleration	O
for	O
the	O
Deflate	B-Algorithm
algorithm	O
,	O
which	O
is	O
used	O
by	O
zlib	B-Language
and	O
gzip	B-Application
.	O
</s>
<s>
A	O
device	B-Application
driver	I-Application
for	O
hardware-assisted	O
842	B-Algorithm
compression	O
on	O
a	O
POWER	B-Device
processor	O
was	O
added	O
to	O
the	O
Linux	B-Operating_System
kernel	I-Operating_System
in	O
2011	O
.	O
</s>
<s>
More	O
recently	O
,	O
Linux	B-Application
can	O
fallback	O
to	O
a	O
software	O
implementation	O
,	O
which	O
of	O
course	O
is	O
much	O
slower	O
.	O
</s>
<s>
zram	B-Application
,	O
a	O
Linux	B-Operating_System
kernel	I-Operating_System
module	O
for	O
compressed	B-General_Concept
RAM	I-General_Concept
drives	O
,	O
can	O
be	O
configured	O
to	O
use	O
842	B-Algorithm
.	O
</s>
<s>
Researchers	O
have	O
implemented	O
842	B-Algorithm
using	O
graphics	B-Architecture
processing	I-Architecture
units	I-Architecture
and	O
found	O
about	O
30x	O
faster	O
decompression	O
using	O
dedicated	O
GPUs	B-Architecture
.	O
</s>
<s>
An	O
open	O
source	O
library	O
provides	O
842	B-Algorithm
for	O
CUDA	B-Architecture
and	O
OpenCL	B-Application
.	O
</s>
<s>
An	O
FPGA	B-Architecture
implementation	O
of	O
842	B-Algorithm
demonstrated	O
13	O
times	O
better	O
throughput	O
than	O
a	O
software	O
implementation	O
.	O
</s>
