[OpenAFS] recommended UFS parameters for cache partition?

Ted Creedon tcreedon@easystreet.net
Tue, 27 Dec 2011 18:55:18 -0600


--20cf307f3b7217dead04b51c7424
Content-Type: text/plain; charset=ISO-8859-1

assuming that the heads aren't off on another track...

On Tuesday, December 27, 2011, chas williams - CONTRACTOR <
chas@cmf.nrl.navy.mil> wrote:
> information is generally denser on the outer tracks because there is
> simply more space for data which requires less seeks.
>
> On Tue, 27 Dec 2011 12:51:15 -0600
> Ted Creedon <tcreedon@easystreet.net> wrote:
>
>> Not necessarily true with on disk caching
>>
>> Disk heads dont recal after every seek
>>
>> On Tuesday, December 27, 2011, chas williams - CONTRACTOR <
>> chas@cmf.nrl.navy.mil> wrote:
>> > On Mon, 26 Dec 2011 15:16:16 -0500 (EST)
>> > Benjamin Kaduk <kaduk@MIT.EDU> wrote:
>> >
>> >
>> >> (Or should I just not worry about it and assume that any potential
>> >> improvements would be lost in the noise?)
>> >
>> > other than i believe you still cannot use journaling on the caching
>> > partition, there isnt much to do.
>> >
>> > however, as with all disks, the closer to the beginning of the disk
>> > the faster the filesystem will perform.  this is why people now suggest
>> > putting swap at the beginning of the disk (ignoring any tiny boot
>> > partitions you might need).
>> > _______________________________________________
>> > OpenAFS-info mailing list
>> > OpenAFS-info@openafs.org
>> > https://lists.openafs.org/mailman/listinfo/openafs-info
>> >
>
>

--20cf307f3b7217dead04b51c7424
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

assuming that the heads aren&#39;t off on another track...<br><br>On Tuesda=
y, December 27, 2011, chas williams - CONTRACTOR &lt;<a href=3D"mailto:chas=
@cmf.nrl.navy.mil">chas@cmf.nrl.navy.mil</a>&gt; wrote:<br>&gt; information=
 is generally denser on the outer tracks because there is<br>
&gt; simply more space for data which requires less seeks.<br>&gt;<br>&gt; =
On Tue, 27 Dec 2011 12:51:15 -0600<br>&gt; Ted Creedon &lt;<a href=3D"mailt=
o:tcreedon@easystreet.net">tcreedon@easystreet.net</a>&gt; wrote:<br>&gt;<b=
r>
&gt;&gt; Not necessarily true with on disk caching<br>&gt;&gt;<br>&gt;&gt; =
Disk heads dont recal after every seek<br>&gt;&gt;<br>&gt;&gt; On Tuesday, =
December 27, 2011, chas williams - CONTRACTOR &lt;<br>&gt;&gt; <a href=3D"m=
ailto:chas@cmf.nrl.navy.mil">chas@cmf.nrl.navy.mil</a>&gt; wrote:<br>
&gt;&gt; &gt; On Mon, 26 Dec 2011 15:16:16 -0500 (EST)<br>&gt;&gt; &gt; Ben=
jamin Kaduk &lt;<a href=3D"mailto:kaduk@MIT.EDU">kaduk@MIT.EDU</a>&gt; wrot=
e:<br>&gt;&gt; &gt;<br>&gt;&gt; &gt;<br>&gt;&gt; &gt;&gt; (Or should I just=
 not worry about it and assume that any potential<br>
&gt;&gt; &gt;&gt; improvements would be lost in the noise?)<br>&gt;&gt; &gt=
;<br>&gt;&gt; &gt; other than i believe you still cannot use journaling on =
the caching<br>&gt;&gt; &gt; partition, there isnt much to do.<br>&gt;&gt; =
&gt;<br>
&gt;&gt; &gt; however, as with all disks, the closer to the beginning of th=
e disk<br>&gt;&gt; &gt; the faster the filesystem will perform. =A0this is =
why people now suggest<br>&gt;&gt; &gt; putting swap at the beginning of th=
e disk (ignoring any tiny boot<br>
&gt;&gt; &gt; partitions you might need).<br>&gt;&gt; &gt; ________________=
_______________________________<br>&gt;&gt; &gt; OpenAFS-info mailing list<=
br>&gt;&gt; &gt; <a href=3D"mailto:OpenAFS-info@openafs.org">OpenAFS-info@o=
penafs.org</a><br>
&gt;&gt; &gt; <a href=3D"https://lists.openafs.org/mailman/listinfo/openafs=
-info">https://lists.openafs.org/mailman/listinfo/openafs-info</a><br>&gt;&=
gt; &gt;<br>&gt;<br>&gt;

--20cf307f3b7217dead04b51c7424--