Hi, I recently took over a half-well done vmware environment and need some direction.
There are 3 servers currently:
host 1 - single proc supermicro, runs vcsa (6.7)
host 2- dual proc supermicro, runs several vm's, locally (6.5)
host 3 - dual proc supermicro, runs several vm's, locally (6.7)
Total storage is about 1TB on all of the vm's combined.
All vm's running are in production and need to be up 100% of the time (dc's, puppet, dev, veeam, etc)
Budget = $15k give or take.
1 - Should I updated local storage on each ESX host and ram and call it a day?
2 - Should I purchase a shared DAS (iscsi) storage, if so, which one? vendor websites are annoying to get to the meat and potatoes (prices) without jumping through the sales team and days of waiting.
3 - Since total storage use is low (don't ever see it going above 5 TB), should I opt for all-flash DAS?
4 - What do you guys recommend?
Ideally, I'd like to get to a point where all hosts are on same esxi, vcenter can do vmotion and manage everything, and all vm's are running on fast, local, flash storage so esxi hosts can fail but the vm will get migrated to a different host. (this will also help me use update manager and get things updated when needed)
I'm open to all good ideas.
Load on each system is roughly 10% cpu, 70% ram. (64gb ram each, will be upgrading to 128gb each soon)
Goal is to be able to survive 1 host failure and have essential infra vm's continue to run.
A short time after I started, one of these vm hosts locked up / purple screened and that's when I started to dig into how it's set up. There is no real redundancy or failover right now. I'm just trying to get essential services to continue to run in the case that a host fails.
Hello RadiatedBirds My suggestion would be to purchase a DAS storage, my friend has lab with the same setup and he uses MD3200 Dell
The MD3200 has dual controller with 8 direct SAS 6GB ports that you can connect redundantly to 4 hosts. Best of all, you get the massive IO of direct attached storage
hope it helps