I had similar questions and spent enormous effort dredging answers out of Dell, 
so let me share to the best of my memory and/or correctness of the answers I 
was given.


1.        The reason it's hard to get answers out of your Dell reps is that 
this is a productization of semi-custom work they've done for large customers 
who buy 100 racks worth at a time.  They're trying to make it available to 
smaller shops than that, but the C-series is still sold on a rather different 
basis than the R-series.  They will want you to buy at least a dozen of them in 
one go, and they will want you to already be fasttrack certified and buy your 
own spare parts.  They are warrantied, but it's more like they replace your 
parts usage than that they really do troubleshooting and 
Unisys-parts-replacement.  They also have a relatively long lead time (over a 
month).  But they are available now.

2.       In terms of the hardware, there's no shared backplane or anything like 
that.  The only shared element is the power supplies.  Each set of six drives 
is physically cabled to a RAID card in one of the four servers.  Each server 
has its own Ethernet ports and each server also has a management Ethernet port 
to talk to the BMC via IPMI.  This is *not* a DRAC, and they do not run OMSA.  
You can run whatever RAID types the LSI cards support, and you can get options 
ranging from JBOD to an LSI card that seems to have specs similar to what is 
sold as H700 on the R-series.

3.       Pricing for us ended up being basically the same as 4x similarly 
configured R610s.  We are part of the Expedia family and have a negotiated rate 
on R-series hardware, so this may or may not be true for you.  Obviously 
there's a space savings.  There's some power savings, but the numbers we ran 
made it seem not that enormous.  Beware that depending on how power-hungry your 
configuration is you may not have true 1+1 power supply redundancy.  Supposedly 
if you lose a PSU and you're over the capacity of the other one it does some 
kind of throttling rather than just crashing one or more of the four nodes, but 
I was unsuccessful at getting good information on how that works.

Hope that helps.  I'll try to dredge my memory if there's any other information 
that could be helpful.

Ryan
_______________________________________________
Linux-PowerEdge mailing list
[email protected]
https://lists.us.dell.com/mailman/listinfo/linux-poweredge
Please read the FAQ at http://lists.us.dell.com/faq

Reply via email to