Re: 64 node Oracle RAC Cluster (The reality of...)

On 06/22/2005 04:34:04 PM, Kevin Closson wrote:
>  >
> >1) Just because a CFS is supported doesn't mean it is the most 
> >reliable service of an OS. If a given vintage of ASM or 
> >straight shared raw has fewer "moving parts" (shall we say 
> >less code path?) than a given CFS,
> 
>  can you tell me how having a separate instance specifically
> for ASM in addition to your production instances is considered
> less moving parts? As far as code path, raw versus direct IO
> CFS comparisons are old school. ASM has, um, quite a bit
> of overhead and comms when manipulating files (not to be
> confused with manipulating the contents of files).


ASM was advertised as "Veritas killer" (wink, wink) and is, according to what I 
read,
a database-aware version of LVM. I am slowly getting used to idea of Oracle 
devouring its
customers (PeopleSoft, BEA Systems, Veritas) but I still don't see why I 
wouldn't go with
EMC and clustered FS (Polyserve Matrix, VxFS, OCFS) instead of Oracle. Why 
would I want 
to add another layer of complexity to the database software and not go with 
proven volume
managers (I believe that both Veritas and EMC have something like that) which 
do not drop
a file here and there? My problem with Veritas killer is why would they want to 
kill Veritas
in the first place? Just because Geoff Squire used to be Ellison's no. 2? I 
believe that Veritas
have brought many customers to Oracle and that they have an excellent range of 
products 
(file system, backup, replication) which doesn't compete with Oracle. What is 
next? Oracle to 
buy Red Hat? IBM? HP? SUN? The company in Redmond WA? No, they cannot conquer 
Redmond, but everything
else is in the realm of possibility.

-- 
Mladen Gogala
Oracle DBA


--
http://www.freelists.org/webpage/oracle-l

Other related posts: