[geeks] Sun to adopt newest Intel Xeon chips for upcoming servers (link)

[Patrick Giagnocavo]

On Jan 23, 2007, at 2:27 PM, Mark Benson wrote:

> On 23 Jan 2007, at 11:23, Bill Bradford wrote:
>
>> On Tue, Jan 23, 2007 at 02:15:23AM -0600, Jonathan C. Patschke wrote:
>>> At this point, why by a Sun if it's just going to be Yet Another
>>> Intel
>>> Reference Design In An Overpriced Chassis?
>>
>> The only "reference design" system they were selling that I saw was
>> the
>> original x2100 which was a Tyan-motherboard system.
>
> The fact the only 'reference' design came from Tyan, not AMD leads me
> to believe AMD possess few reference  designs for server hardware.

Actually the very first Sun/Opteron systems were the V20z and V40z 
systems, which were Celestica designs.  I know that for sure, am not 
sure about the x2100 being Tyan based.

> following if anyone can prove otherwise, I've not looked into Opteron
> servers in huge depth outside of Sun's (why bother, Sun's are the
> best anyway ;o) ) but I don't see a huge number of companies taking
> them up, which leads me to believe something Intel offers makes
> Xeon;s an easier platform to work with.
>

The HP DL45 is virtually an identical copy of the Celestica boards as 
well (but with a lame IPMI interface instead of the nicer ILOM).  So 
oddly, the Sun V20z and HP DL145 are brothers :-)
> o my
> knowledge at least). Much like Apple, Sun have a lot of the
> backbreaking work done already, all they have to do is shape it to
> suit, and that special Sun ingredient and you've got really good
> server hardware, relatively easily.
>
Amazingly however, I have found a challenge with non-Sun memory in the 
V20z, even though the same RAM works in HP and IBM systems.  I would 
have thought that the memory controller being integrated would mean 
that the same RAM would work with every Opteron.

--Patrick