Another glorious vSAN Chat is under our belts! This month we dove into VMware on AWS! We brought in experts John Nicholson (featuring his dog Auto) and Glenn Sizemore. Catch up on the hour below!
Q1: How is Elastic vSAN connected to #VMWonAWS and how do they complement one another?
A1: It helps handle well cases where storage is growing asymmetric to compute needs. Helps quite a bit when you can’t always predict storage growth.
A1: Elastic vSAN is a fusion of AWS EBS and VMware #vSAN which enables customers with high capacity workloads to control costs by reducing the number of #VMWonAWS hosts required for a given amount of capacity.
Q2: What happens if there is an Availability Zone failure inside AWS? How does #VMWonAWS and Elastic vSAN protect from that?
A2: vSAN stretched clusters are a true active/active design to protect from this. You can even have applications running on both sides.
A2: Deploy a Stretched Cluster, and it’s an HA event! vSphere Availability simply restarts any failed VMs in the surviving AZ. -> link
Q3: Can you explain how Elastic vSAN solved the data gravity challenge within #VMWonAWS?
A3: For internal network gravity: Host failures don’t require hosts having full rebuilds of data. they can just reattach the EBS volumes to a new server with no human intervention.
A3: for external gravity (WAN gravity) there is the fact that it’s a lot cheaper and easier to process those mountains of data you’ve stuffed into S3 and other AWS PaaS services than exfiltrate that out over the WAN. The amount of data #VMWonAWS can support on a single host is limited by our ability to recover from failure. Elastic vSAN uses an innovative recovery process that leverages the EBS durability; allowing us to support high-density hosts.
Q4: I heard that #VMWonAWS now supports native SQL Server Clusters on vSAN. Can you provide more details on this? #vSANChat
A4: While vSAN has supported SCSI3PR clustering on iSCSI for a while, this is a native VMDK supported option. Going to be able to play nice with snapshots, VADP backups etc.
Also nothing stopping you from AlwaysOn shared nothing mirroring, but some DBAs reallllllllly love their shared volume clustering for FCI and wanted something more native than layering on iSCSI.
A4: One of the more exciting aspects of running the #VMWonAWS service is being able to ship advancements as we develop them continuously. SCSI3-PR support is another example of that process in work. Native SQL Server Cluster support on vSAN
Q5: What Applications are a good fit for Elastic vSAN? #vSANChat
A5: Anything that is high capacity moderate performance would be ideal. Fortunately, there’s no need to guess. Just go plug-in the workload, and the sizer will tell you how many hosts are required.
(link: https://vmcsizer.vmware.com/)
A5: Any application which has data growth that grows unpredictably, or out of the normal ratio to compute that regular @vmwarecloudaws offers.
Q6: What class of EBS media is used with Elastic vSAN and the R5.metal instances?
A6: Today Elastic vSAN is all GP2. We continue to evaluate that decision and if it ever makes sense to support more or other flavors, we will.
A6: It’s nice that you don’t have to learn how to manage EBS, as all of this underlying hardware, connectivity and lifecycle is automatic.
Q7: When would I want to use a stretched cluster in #VMWonAWS?
A7: Stretched Clusters provide an infrastructure level solution to the AWS Availability challenge. They empower customers to move to the cloud regardless of their current applications and or availability requirements.
No two vSAN chats are the same! Bring your Qs to our next chat on the last Thursday of every month. If there are topics you’d love to discuss, send us a DM with your questions and ideas. We’d love to hear from you. We’ll catch you on the interwebs.
