[OpenAFS] recommended hardware for new production environment
Jonathan Nilsson
jnilsson@uci.edu
Fri, 19 Mar 2010 16:25:02 -0700
--0016e6d7f07b4ccf0a04822fa934
Content-Type: text/plain; charset=UTF-8
> > Thoughts and other feedback is much appreciated. Thanks!
>
> In fact, probably the greatest concern is if you use any number of
> virtual machines make sure the virtual hosts with the db processes
> are on distinct physical machines.
>
That makes sense, and is definitely something I can enforce.
With even a large cell I can't remember the last time I thought
> "my db servers need more memory or cpu oomph".
>
Good to know... I guess, I just really didn't know what to expect as the
dbserver entries grew, or what sort of limits defined a "large" cell.
> I think it's more
> important to find the most rock-stable machines you can, really.
>
Perhaps our donated systems will not make great dbservers since their
stability is unknown, or at least questionable.
> The cell at work (umich.edu) has a giant number of entries in the
> vldb and pt databases, and probably the most load the db processes
> get is the nightly "dump the vldb so we can stash it away and
> do interesting things with it" job.
>
Thanks for your input everyone!
--
Jonathan Nilsson, jnilsson@uci.edu
Social Sciences Computing Services
949.824.1536, SSPA 4110, UC Irvine
--0016e6d7f07b4ccf0a04822fa934
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
<br><div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"m=
argin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); paddin=
g-left: 1ex;">> =C2=A0 =C2=A0Thoughts and other feedback is much appreci=
ated.=C2=A0 Thanks!<br>
<br>
In fact, probably the greatest concern is if you use any number of<br>
virtual machines make sure the virtual hosts with the db processes<br>
are on distinct physical machines.<br></blockquote><div><br>That makes sens=
e, and is definitely something I can enforce.=C2=A0 <br><br></div><blockquo=
te class=3D"gmail_quote" style=3D"margin: 0pt 0pt 0pt 0.8ex; border-left: 1=
px solid rgb(204, 204, 204); padding-left: 1ex;">
With even a large cell I can't remember the last time I thought<br>
"my db servers need more memory or cpu oomph". <br></blockquote><=
div><br>Good to know... I guess, I just really didn't know what to expe=
ct as the dbserver entries grew, or what sort of limits defined a "lar=
ge" cell.<br>
=C2=A0</div><blockquote class=3D"gmail_quote" style=3D"margin: 0pt 0pt 0pt =
0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">I thi=
nk it's more<br>
important to find the most rock-stable machines you can, really.<br></block=
quote><div><br>Perhaps our donated systems will not make great dbservers si=
nce their=20
stability is unknown, or at least questionable.<br>=C2=A0</div><blockquote =
class=3D"gmail_quote" style=3D"margin: 0pt 0pt 0pt 0.8ex; border-left: 1px =
solid rgb(204, 204, 204); padding-left: 1ex;">
The cell at work (<a href=3D"http://umich.edu" target=3D"_blank">umich.edu<=
/a>) has a giant number of entries in the<br>
vldb and pt databases, and probably the most load the db processes<br>
get is the nightly "dump the vldb so we can stash it away and<br>
do interesting things with it" job.<br></blockquote><div><br>Thanks fo=
r your input everyone!<br></div></div><br clear=3D"all">--<br>Jonathan Nils=
son, <a href=3D"mailto:jnilsson@uci.edu">jnilsson@uci.edu</a><br>Social=20
Sciences Computing Services<br>949.824.1536, SSPA 4110, UC Irvine<br>
<br>
--0016e6d7f07b4ccf0a04822fa934--