× The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.



Not directly, no.

While i'm nowhere near an expert on System i hardware, but PC server
hardware also uses buffering, lot's of it.

On a quality PC server, like a xSeries 346 (Yes, there is successor, but
haven't gotten my hands on one of those yet), you have a hardware raid
controller for the locally attached disk drives. In the base setup, this
controller can do RAID0 and RAID1 without any caching - If you purchase
the additional ServeRaid 7k Card (which consists of a memory bank and a
battery), the onboard controller will also allow RAID5, and also offer
write caching using the memory on board of the controller.

There are different setups from different vendors, HP sold controllers
which could do read caching only, until you inserted a battery card,
which enabled you to do write caching. The memory was mounted fixed on
the controller.

But these are all U320/SAS Controllers, not IDE. IDE is not used for
disk attachments from the mid-end upwards. Low-End server usually use
uncached, RAID1 SATA-Harddrives, with a non-protected Cache of 8-16MB
located on the Disk Assembly itself. This cache is used for read and
writes, but cleared when commiting Data to the disk - degrading the
performance and usefulness of the cache, but this behavior is important
for data security.

This cache also exists on U320 and SATA disks, but it is not important
as it's size is weak when compared to the 256MB/512MB battery backed
cache on the raid controller.

Of course, the operating System (be it Windows Server, Linux, BSD,
Solaris, or whatever) of course does it's own caching in RAM - but this
write cache is also cleared when the application requests a commited
write, which must be either on Disk or in battery Backed RAID-Cache.

PC-Servers usually don't have such fancy Disk-Expansion technologies
like the System i does - Storage Extension is usually done through the
use of a SAN, be it in a Budget fashion with iSCSI, and more enterprisey
technologies with Fibre Channel and other stuff.

As I only work for SMBs, I have next to no knowledge of enterprisey
technology, so someone else on this list might offer a better opinion.

In the SMB World, when it comes to the hardware side, the System i
doesn't offer any advantages over PC servers - it offers comparable
performance for a MUCH higher price.

-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of albartell
Sent: Friday, December 29, 2006 4:16 PM
To: 'Midrange Systems Technical Discussion'
Subject: RE: i5 Cache Battery

The reason it effected performance was due to all writes to disk were
direct instead of buffered, which apparently makes a pretty big
difference.

Forgive my ignorance on the subject, but would that be one of the
benefits
to iSeries I/O vs. regular PC IDE drives?  That being they have buffered
mechanisms built into both the hardware and OS/400 that provide
significant
performance increases.  When the cache battery dies is one experiencing
I/O
performance similar to what a PC IDE HD would give?

I am always looking for additional "one ups" that the iSeries has over
other
platforms :-)

Aaron Bartell
http://mowyourlawn.com

-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Ron Adams
Sent: Friday, December 29, 2006 9:01 AM
To: Midrange Systems Technical Discussion
Subject: Re: i5 Cache Battery

Incidentally, if the cache battery fails, or for some strange reason you
miss the warning message, system performance will be seriously degraded.
We
had this happen a couple of weeks ago and it messed everything up. Jobs
that
previously ran in 1 or 2 minutes were taking 20-30 minutes. I struggled
with
the issue for almost a week until I stumbled upon it accidentally when
doing
a WRKDSKSTS and noticed the protection status was "DEGRADED". The reason
it
effected performance was due to all writes to disk were direct instead
of
buffered, which apparently makes a pretty big difference.
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe,
unsubscribe,
or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx Before posting, please take a
moment to review the archives at http://archive.midrange.com/midrange-l.


As an Amazon Associate we earn from qualifying purchases.

This thread ...

Replies:

Follow On AppleNews
Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].

Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.