Dual Processors = Automatic CPU Failover?

Dumb Question, I fear, but...in dual-processor servers--like Dell PowerEdge 1900, for example, with, let's say, two dual-core or quad-core Xeon processors--does the server fail operations over transparently to the remaining processor if one of the two suddenly dies?  Is this just implicit in all dual-processor machines?
mtn_lionAsked:
Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

☠ MASQ ☠Commented:
This "standby processing capability" for multi-core processors is part of Intel's development strategy.  

Yes, theoretically processing should fall-over to the remaining core but this in itself needs hardware programming so that the transfer of affinity is seamless.  You also need to bear in mind that, unless you are specifically transfering the affinity from one core to another and that the second core is pretty much unused, just waiting for a failiure so it can take over, that your actual processing power will be significantly reduced.

Currently if one core fails suddenly it is still likely you will be left with an inherently unstable system.
0
mtn_lionAuthor Commented:
Uh, thanks, masqueraid, for your prompt reply, but...I think I was hoping for a simpler answer.  Also, to be clear, I wasn't asking about fail-over between *cores* but between separate processors in a dual-processor system (regardless of processor type).  Surely there is a simple answer.  If I buy an "off the shelf" server with two processors in it, and one fails, will the server just keep on running as a one-cpu server?
0
☠ MASQ ☠Commented:
Sorry, misread your question as integrated processors.  Yes, multiple processor based servers already cope with this.

I'll save the other comment until someone asks the question I answered :)
0
Big Business Goals? Which KPIs Will Help You

The most successful MSPs rely on metrics – known as key performance indicators (KPIs) – for making informed decisions that help their businesses thrive, rather than just survive. This eBook provides an overview of the most important KPIs used by top MSPs.

CallandorCommented:
I don't think failover will occur.  When a system boots up, all the hardware is made aware of what is present, unless the device is designed for hot-swapping.  A failure of a processor in the middle of a calculation, for example, would be catastrophic because OS'es just send tasks to cpus and expect results back.  A cluster would provide failover from one server to another, but that is at a higher level than a processor.
0

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
mtn_lionAuthor Commented:
No worries, happens to me all the time.  So...seamless transition from two to one upon failure, then, right?
0
CallandorCommented:
Oops - MASQUERAID had it right - there are servers that do automatically detect processor failure and act on it.  For example: http://www.serverwatch.com/hreviews/article.php/3662151
0
mtn_lionAuthor Commented:
My last post was pre-Callandor.  Now things are getting interesting.  That makes sense, C.  So likely a server would crash on failure of one cpu?
0
☠ MASQ ☠Commented:
Callandor has the better answer.  When I'm talking about "multiple processor based servers" I'm talking about clusters.  Processors failing within a single machine will (just the same as in dual core) leave the system unstable.  You would be able to restart an individual machine and it could run with a processor "down"  but if it failed in use there is no mechanism to allow the other processor(s) to take over active tasks.
0
☠ MASQ ☠Commented:
This is what happens with multiple cross posts :)
0
mtn_lionAuthor Commented:
If I read that article right, that's only on the enterprise-class, $30K 4-way server, and it's a proprietary feature.  So I'm thinking you're right, C., for off-the-shelf small biz servers, e.g., Dell PowerEdge 2900.  And now I see M's last post, which agrees.  Likely crash, then likely successful functioning on remaining single cpu after restart (assuming no other damage done).

What say I split the points?
0
CallandorCommented:
Sounds good - the automatic failover is an option for expensive machines, but not for Dell Poweredge 1900's.
0
☠ MASQ ☠Commented:
No problem (unless Callandor has already posted!)

The "Real-Time Capacity" being offered on the Unisys platform only (AFAIK) falls over when processing power is reduced, a restart is still required following actual processor failure.
0
mtn_lionAuthor Commented:
Thanks, guys.  Good discussion, always nice to learn some stuff!
0
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Components

From novice to tech pro — start learning today.