Who's Hyperconverged, who\what did you use?

[Jaan]

We're due a server refresh this year. We have two r610 acting as hyper-v hosts with local (replicated) data.

 

These were never specced as VM hosts and were not even running hyper-v when i started. We're having IO/RAM uses as we are over utilising them, the hardware has very limited warranty time left.

 

We've been looking at hyperconverged, with integrated interconnect or external physical switch based interconnect as well as Nvme\HDD storage tiering.

 

I guess my questions are as follows;

 

Are you hyper converged or planning to be?

If so how many nodes do you have?

How are they interlinked?

What hardware is it?

Who supplied and/or installed the servers?

 

Sorry for all the questions, and thinks in advanced.

[sluggster66]

Hi

 

We moved to two Starwind HCA nodes, about 4 months ago. They have been rock solid.

Support is amazing.

 

Linked with 10Gb direct.

 

The hardware is supplied and supported by starwind themselves.

 

The underlying hardware is Dell 730xds

 

Let me know if you want anymore info

[SpaceInvader83]

Hi,

 

Would it be possible to have a bit more info on your Starwind setup? Full spec and ball park costings would be great.

 

Thanks,

 

Simon

[sluggster66]

Drop me a PM

[dhicks]

Are you hyper converged or planning to be?

If so how many nodes do you have?

How are they interlinked?

What hardware is it?

Who supplied and/or installed the servers?

 

We're using a pretty much hyper-converged solution, yes, although it wasn't called that when we set it up (it's that old) - we use DRBD as an underlying block replication method to replicate storage accross three nodes running XenServer. They are interlinked via a dedicated switch - a couple of 7-year-old Dell servers and one slightly newer one, we also have a fourth 5-year-old HP server as part of the pool (but not sharing storage). The servers are all random devices bought for other purposes and at some point smooshed together into a virtualisation setup. We had some very useful help and advice at the time we were upgrading the hardware from VeryPC, who provided things like more RAM and new RAID cards and disks for the servers.

 

We have our eye on Stone's Equinox system - basically a nice all-in-one drop-in solution, I gather a two node solution should be somewhere under £20,000 (that's with VAT). Other companies provide similar systems based on Windows Server 2016 / Hyper-V with a tiered storage solution (Storage Spaces). We had a look at Starwind too, I think it came out about the same price as Stone's system in the end.

[Jaan]

Thread Zombie!

 

Who's now Hyperconverged, and what did you use?

 

This is my project for this year.

[sluggster66]

Starwind HCA

[pablo007]

For those who bought Starwind HCA, did you get a demo of it prior also which supplier did you use, i like the idea of this setup but want to know more about it, how long warranty, are they VMware compatible?

 

Thanks as always in advance.

[Jaan]

For those who bought Starwind HCA, did you get a demo of it prior also which supplier did you use, i like the idea of this setup but want to know more about it, how long warranty, are they VMware compatible?

 

Thanks as always in advance.

 

Yeah i'm not sure what the benefit of starwind is over the standard 2 node S2d

[sluggster66]

Yeah i'm not sure what the benefit of starwind is over the standard 2 node S2d

 

The support! The cluster is proactively monitored by them, so they know when you have an issue before you do. It allows for me to concentrate on the other 5 infrastructures we have to support.

[sluggster66]

For those who bought Starwind HCA, did you get a demo of it prior also which supplier did you use, i like the idea of this setup but want to know more about it, how long warranty, are they VMware compatible?

 

Thanks as always in advance.

 

I bought through softcat as they are a partner. I also had several Demos.

[mavhc]

The support! The cluster is proactively monitored by them, so they know when you have an issue before you do. It allows for me to concentrate on the other 5 infrastructures we have to support.

 

Do they send you a list of what they've done that week, or similar?

[sluggster66]

Only if an issue as been observed. They will first let you know what the issues is and if you need any help dealing with it. It’s a 24 hour service, so have had an email after an unexpected failover at 5AM on a Sunday morning.

 

It’s hands down the best company support I have used. Engineers know their product and Hyper-v/VMware inside out.