Virtualization Build - Punch holes in this build please
-
@DustinB3403 said:
I am also recommending the XOA appliance for our backup, but it has to be mentioned in the quote.
It was not in the post. You explicitly mentioned using XO not XOA. Or at least that is how I read it.
I did account for a rough amount of time 21-30 hours at $200/hour for labor.
I did see that time listed, but are you saying that there is not going to be any of your time also?
I did forget to mention that we currently have Shadow Protect for the next 3 years which will likely be used in someway here.
That is an agent based backup solution and, assuming you are P2V/V2V existing servers already setup in Shadow Protect, has nothing to do with this project unless you are stealing its storage.
-
@DustinB3403 this is a kinda "what if" question..
Let's say the client asked to pay for this project on a monthly basis to save cash (vs spending a lump sum)
What/how would you charge 'em? -
Sorry about the XO vs XOA, I am recommending for the XOA variant. Just so it's clear.
Our time is always here, so there is no added cost for us to be here to work on the migrate.
ShadowProtect can be run on the VMs (which we have already on the physical servers) in lue of XOA or along with it to create a separate backup of just the Network shares.
-
@DustinB3403 said:
Our time is always here, so there is no added cost for us to be here to work on the migrate.
tl;dr Employee time is not free.
This is a fallacy. Your time is not always there. Your time is dedicated to other tasks (or should be). It is true it is not an additional up front cost to the company because you are salary or something. But you have normal duties that do not involve a server upgrade that have to be put off to do a server upgrade.
-
@JaredBusch said:
@DustinB3403 said:
Our time is always here, so there is no added cost for us to be here to work on the migrate.
tl;dr Employee time is not free.
This is a fallacy. Your time is not always there. Your time is dedicated to other tasks (or should be). It is true it is not an additional up front cost to the company because you are salary or something. But you have normal duties that do not involve a server upgrade that have to be put off to do a server upgrade.
To summarize you agree with the above build, but use XOA.
-
@DustinB3403 said:
@JaredBusch said:
@DustinB3403 said:
Our time is always here, so there is no added cost for us to be here to work on the migrate.
tl;dr Employee time is not free.
This is a fallacy. Your time is not always there. Your time is dedicated to other tasks (or should be). It is true it is not an additional up front cost to the company because you are salary or something. But you have normal duties that do not involve a server upgrade that have to be put off to do a server upgrade.
To summarize you agree with the above build, but use XOA.
And you need to add time. but yes. it is not a bad design. I would need to know a lot more about your systems to say if the specs are too much or too lite. But they are solid.
-
Looks pretty solid. Good devices, good setup.
Only thing I would add is maybe throwing in something like a Scale setup as well so that you would have a point of comparison. It's a very different design, three nodes instead of two, all inclusive instead of parts. It shifts the pricing around to different things in different places. At that cost, though, you are able to consider one. The big advantage of the Scale is inclusive support pricing and easy future expansion through node addition... and the big one is providing something that your company is prepared to support themselves without needing your expertise should you leave, get hit by the bus, etc.
-
Thank you all for the advice.
-
-
@MattSpeller Yeah no fooling around with this build.
-
@DustinB3403 said:
@MattSpeller Yeah no fooling around with this build.
You should post iops numbers if you go this way, just so we can drool some more.
-
@travisdh1 said:
@DustinB3403 said:
@MattSpeller Yeah no fooling around with this build.
You should post iops numbers if you go this way, just so we can drool some more.
Don't do this. Please. I'll have to add pages to my diary to account for the emotional baggage this will create.
-
@travisdh1 said:
@DustinB3403 said:
@MattSpeller Yeah no fooling around with this build.
You should post iops numbers if you go this way, just so we can drool some more.
I'll be head of the lynch mob if he fails to post them
------[
------[
------[
------[ -
Doesn't using a Scale device change his risk solution?
As I understood it, Dustin plans to have two nearly synced servers over two diverse locations. In addition to that he'll have offsite backup at Rackspace or someone else.
This allows for the company to completely loose a location and stay up and running completely (though networking will have to be considered to make sure things flow correctly).
With a Scale solution, while it is possibly simpler to support also can't survive the main site being lost.
Now maybe that doesn't matter, i.e. if the main site is down, the whole company would be down anyway.
Just a thought.
-
12x 960GB 2.5" SATA MLC Solid State Drive (Edge Server Enterprise Class) ($399.99/Each)
Damn, that's what I paid for my 1 TB drives from HP a few years ago...
-
@Dashrender said:
12x 960GB 2.5" SATA MLC Solid State Drive (Edge Server Enterprise Class) ($399.99/Each)
Damn, that's what I paid for my 1 TB drives from HP a few years ago...
1TB SSD from a few years ago?
Or Spinning Rust from a few years ago?
-
@DustinB3403 said:
@Dashrender said:
12x 960GB 2.5" SATA MLC Solid State Drive (Edge Server Enterprise Class) ($399.99/Each)
Damn, that's what I paid for my 1 TB drives from HP a few years ago...
1TB SSD from a few years ago?
Or Spinning Rust from a few years ago?
Spinning rust.
-
I have a nearly 5 year old IBM server that if I replaced it with SSDs would probably last me another 5+ years.
-
@Dashrender said:
Doesn't using a Scale device change his risk solution?
As I understood it, Dustin plans to have two nearly synced servers over two diverse locations. In addition to that he'll have offsite backup at Rackspace or someone else.
This allows for the company to completely loose a location and stay up and running completely (though networking will have to be considered to make sure things flow correctly).
With a Scale solution, while it is possibly simpler to support also can't survive the main site being lost.
Now maybe that doesn't matter, i.e. if the main site is down, the whole company would be down anyway.
Just a thought.
I thought that his two main servers were at a single site.
-
@scottalanmiller said:
@Dashrender said:
Doesn't using a Scale device change his risk solution?
As I understood it, Dustin plans to have two nearly synced servers over two diverse locations. In addition to that he'll have offsite backup at Rackspace or someone else.
This allows for the company to completely loose a location and stay up and running completely (though networking will have to be considered to make sure things flow correctly).
With a Scale solution, while it is possibly simpler to support also can't survive the main site being lost.
Now maybe that doesn't matter, i.e. if the main site is down, the whole company would be down anyway.
Just a thought.
I thought that his two main servers were at a single site.
You're probably right, I must have misread it.
