To the comment above
You sir are a jackass!
5 posts • joined 25 Jun 2010
You sir are a jackass!
Wow...you win. As I stated in a prior post, you must be the poster child for the SMB crew. And Yes your DAS solution makes sense for the few thousand user shop or maybe the LH P4000 you mentioned...again small to medium at best.
Once you carry over to the big boys, at some point, you will see the light.
100K+ exchange users, multiple applications, integrated backup requirements with restore capabilities and leveraging the SAN investment is where it is at....and Yes this is where XIV plays and plays well.
You must agree at some point the Capex and Opex favor the SAN when you scale to the numbers I deal with above...if you don't, you sir don't understand the financial and business side of IT.
Sure the drives are going to fail, but what type of extra load and performance degradation will the rebuild cause on your Staples DAS solution?? And how long of a duration will the rebuild take? What effect will this bring to the business operations? On XIV, a rebuild takes under 30min for a 1TB drive under a fully utilized 79TB usable machine and ~50min for 2TB drive in a fully allocated 161TB machine.....So what is your rebuild times? What is the reponse time hit and user expaerience negative factor going to be?
No spare drives to allocate in XIV just spare capacity which is already built in.
"carefully provisioned luns"
Sorry, wrong! Not in XIV. Simply what is the size, how many do you need, and what is the naming convention and your done. All in one screen, in about 1-2 min (and that is if your using it for the first time) Check it out if you are a non-believer, same look and feel as the MAC. Even if you still don't like XIV for Exchange you must give some props for the management UI (which has no licensing fee to use nor a separate server requirement to implement)
And yes XIV is a predictable/consistent performer...reason being is the underlying architecture. Remember we are a grid consisting of off the shelf computing nodes with distributed cache and 84 core processors running Linux under the covers...much like how Google and Amazon handle their business...(pretty successful model don't you think) A node falls out from the grid, no worries the remaining pick up the slack..a new node is added in non-disruptively.. great you just added in addition cache, disk, and CPU power to the grid. No hot spots in XIV, we spread all the data in 1MB chunks across all the disk drives in the array...that's 180 spindles working for you at all times. All done within the array automatically with no management undertaking
And yes you can leverage that SAN because you don't just have to run Exchange...XIV is a powerful machine with customers hosting Exchange and other applications at the same time with no performance issues.
And to highlight, Snapshot technology, thin provisioning, copy services, replication, and data migration facility are all built into the XIV with no additional licensing fees like other vendors. So may I ask what type of Snapshot technologies, consistency groups for application integration does the Staples solution give the test/dev and back up teams?
Cheap and deep might look great on the surface but you need features, functionality, reliability and a tried and true support organization behind the solution for it to make sense for the business.
Sorry I must have accidentally been put into the SMB realm. XIV handles enterprise class customers...if you want your small to medium class at best...yes yes please do hit the Staples on your way into work.
Yeah, don't leverage you existing SAN infrastructure that procurement still has 2-3 years on the books before the swap for new gear....lets step back into time and build islands of DAS...That helps to CTO big time.
Reliability, management, and consistent predictable performance would just be a few benefits to XIV.
Not sure your Staples approach above would get the green light from the operations, support, and engineering teams at most corporations. Mom and Pop shops...maybe, just maybe. Folks that need dependable storage solutions that meet SLAs to the business or it is their ass....NOT
There is no layout discussions or planning with XIV, neither upfront or a year down the road. And we run on SATA behind the grid.
A single DAG solution comprising 5 mailbox servers was created that supported a total of 40,000 mailboxes with a mailbox size of 3.5GB and I/OPs target of .18