VMware VSAN … and the great missed opportunity One of the most interesting announcements at VMworld was VSAN 6.1. The product is quickly maturing and new features are being added version after version (so here's what’s new). Even though some of the features could easily be disputable – like the 2-node cluster for ROBO, with its workarounds to make it work – VSAN is …
The last time I checked vSAN was so expensive it made us laugh.
I could buy an entire enterprise infrastructure 3 times over for the cost of licensing vSAN
They simply don't understand that enterprise storage costs have dramatically reduced in recent years thanks to all the new entrants in the market, Nimble, Pure, Tintri et al
Which is why EMC is losing ground. This could be their Kodak moment, do they hold back on vSAN so it doesn't eat the parents lunch, or do they let it fly and try and eat everyone else's?
Currently they are holding back by not giving it the features it could have and by making it expensive to use compared to alternatives.
EMC's play in this space is ScaleIO NOT VSAN.
ScaleIO is a true competitor in the hyper-converged space with actual client support beyond a VMware cluster, multi-hypervisor support and a flexible license model.
VSAN is not cost competitive and with a real limit on a minimum of 4 nodes will not fit into the Essentials+ bundle so you need full vSphere licenses and then a very expensive, socket based, license for your storage system. It just does not work in a SMB commercial model and enterprises want the possibility of more than 32 nodes.
Disclaimer: I work as an EMC Solution Architect in the distribution channel in APJ.
Just to be clear:
1) VSAN is fully supported with 3 nodes, and we have many customers running with a 3 node config.
2) VSAN is supported as of 6.1 in a ROBO configuration with a central witness and just 2 nodes.
3) VSAN supports 64 hosts with vSphere 6.0.
4) With regards to being a competitor, we currently have over 2000 VSAN customers, which to my knowledge makes VMware one of the biggest hyperconverged players.
5) We also have per CPU licensing and there is a ROBO pack available for those type of deployments. I hope that clarifies things. Thanks.
Disclaimer: I work for VMware in the Storage and Availability BU
Duncan I have a lot of respect for you but here you are being disingenuous at best.
Look at the first paragraph on Cormac's design guide that is on the VSAN product page....
The minimum configuration required for Virtual SAN is 3 ESXi hosts. However, this
smallest environment has important restrictions. In Virtual SAN, if there is a failure,
an attempt is made to rebuild any virtual machine components from the failed
device or host on the remaining cluster. In a 3-node cluster, if one node fails, there is
nowhere to rebuild the failed components. The same principle holds for a host that
is placed in maintenance mode. One of the maintenance mode options is to evacuate
all the data from the host. However this will only be possible if there are 4 or more
nodes in the cluster, and the cluster has enough spare capacity.
I've read Cormac's guide, and we have had extensive discussions on this. Any scale-out storage system has the same limitation. Even a traditional storage system with 2 controllers has the same problem... when you do maintenance on 1 controller than you are at risk. The statement here was that the minimum is 4, this is not the case. The minimum today is 2 with an external witness, or 3 full hosts. Yes if you do maintenance you will be at risk, if you want to mitigate that and can afford the associated cost I would always recommend to go with 4. I think I have been pretty clear about that on my blog in the past.
Same applies to HA / DRS by the way, more hosts is better.
PS: I seriously have no customer who has 4 nodes or more and does a full "data evac" during maintenance mode...
Yep I agree with you on a 2 controller SAN vs. Scale Out maintenance risks.
However on a dual controller array a firmware upgrade cycle is on the order of 12 months on average, assuming a tier 1 vendor with a mature product. If we look at an NDU on a EMC VNX for instance total exposure is 2 x 30 minute windows every 12 months (one per controller).
vSphere 5.5 had 15 patch release cycles in 25 months, an average of every 1.6 months. Assuming your 3 hosts and an average of 30 minutes to evacuate the VM's, apply the patches, reboot (if required) and re-enter the cluster we are talking 7-8 x 90 minute windows in those same 12 months.
Not quite the same risk profile.
As for any scale out solution has the same limits I agree and with the ScaleIO solution you get that 4th host for the cost of the hardware and hypervisor license, no per processor costs, just license the space you consume. In little old NZ that equates to between $15-20K in savings depending on support level.
Have I missed something, since when has vSAN been storage? vSAN on its own is nothing, it requires hardware & critically disk to deliver anything. It is simply software, in this case enabling Software Defined Storage. In other words, buy the cheapest possible disks and then spend a fortune turning it into an array with some software.
As technology has moved on many enterprise features are now trickling down into modest arrays making this type of product more questionable, particularly at the current costs. Software Defined Storage is where the likes of VMware are trying to obtain complete buy-in and control of the entire tech stack to protect their revenues.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
