VMware Cloud Community
aseaz404
Enthusiast
Enthusiast
Jump to solution

IAAS/vRA 7.3 response seems slow

I'm coming from a vCD 5.x background and I'm used to an almost immediate entry/action in vSphere when for example I deploy, undeploy, start, delete, etc. VMs and vApps.  I have a test environment built on 4 vxrail units and one vSAN.  vCenter is intregrated in the vxrail cluster, IAAS is running on 2012Server (16GB RAM, 4 CPUs Xeon 2.6 GHz VM paravirtual SCSI) with SQL Server 2014.  Everything is running pretty smooth and deployed VMs are very fast.  The issue I've run into is that "Requests", whether getting a “service” from the catalog or destroying a running machine, so what I consider deploying/undeploying in vCD terminology, takes between 45 and 65 seconds from the moment I click OK on the vRA Web GUI to when tasks (clone virtual machine, etc.) start in vSphere.  Things move very fast in vSphere but there's always a significant delay between submitting a request and actions actually taking place.  Additionally when the action completes there's another 40 seconds or so delay for the Web GUI to show the action completed.  So, let's say I had a request for a blueprint to be deployed, even when I see it cloned, configured and running in vSphere it won't be available in the vRA Web GUI, so overall there's 1 1/2 to 2 minutes added to what is otherwise a very fast deployment.  Forgot to mention these are linked clones.  So I'm not sure if this is normal/expected because I never experienced anything like it in vCD and if it's not I could use some guidance in terms of how to reduce these delays.

Tags (1)
0 Kudos
1 Solution

Accepted Solutions
daphnissov
Immortal
Immortal
Jump to solution

Your experiences are pretty consistent with even the fastest most optimal deployments. vRA just isn't as speedy as vCD, but you've gotta understand the massive complexity of vRA and the power of its extensible systems as compared to vCD. There are tons of things going on in the background, and then there's the overhead of RabbitMQ and queue times. There's nothing wrong with your setup, things just take a bit longer with vRA while all the necessary checking/setting take place.

View solution in original post

0 Kudos
2 Replies
daphnissov
Immortal
Immortal
Jump to solution

Your experiences are pretty consistent with even the fastest most optimal deployments. vRA just isn't as speedy as vCD, but you've gotta understand the massive complexity of vRA and the power of its extensible systems as compared to vCD. There are tons of things going on in the background, and then there's the overhead of RabbitMQ and queue times. There's nothing wrong with your setup, things just take a bit longer with vRA while all the necessary checking/setting take place.

0 Kudos
aseaz404
Enthusiast
Enthusiast
Jump to solution

Thanks for the response.  I've been trying to identify potential bottlenecks but as you indicated in your reply there may not be any.  Besides all the checks that take place during the installation are you aware of any optimization documents for the VMs or SQL or something else?  Thanks.

0 Kudos