Work on creation time of Redis Cache
The time of creating a Redis Cache is to long. It should be improved

Thank you for your feedback. This is on our backlog
45 comments
-
Sandro commented
Come on Microsoft, 5 years?
-
Armaan Dhaliwal-McLeod commented
We need to prioritize this. Many customers are complaining about long deployment times for this particular service(much longer than any other service available). I've tried this myself and even just deploying a simple Standard C1 size takes about 30 min to deploy.
-
Suj commented
Any updates Azure?
-
Patrick Ecker commented
Our whole ARM deployment would take 5min, but then there is Redis Cache: 20-30min @ west europe
-
mumtaz khan commented
its better to create redis in container :) - Microsoft please do something .
-
Josiah Berkebile commented
Have to echo this concern. In our Terraform deployment of our job orchestrator which uses Redis, all of the Azure resources get created in 5 minutes or less, and then we are left waiting for another 15-20 minutes for Redis Cache to finish coming online. It's the same for the destruction of the infrastrucutre, as well - we can destroy all of our Azure resources in under 5 minutes, except Redis Cache takes again 15-20 minutes. This really slows down development on our Terraform infrastructure scripting.
-
Alexandre commented
About 20 minutes to create a redis in France Central ! Please do something !
You competitor GCP in Belgium is able to start a redis instance in less than 5 minutes. -
Paul Vrugt commented
this is ridiculous. Creating a redis cache as part of ARM deployment brings our deployement to a screeching halt. How can this still be an issue after more that 5 years?
-
Anonymous commented
Redis Cache in 2020 create in West Europe it will take approximate 35 to 40 minutes
-
Farukh Sharipov commented
West US still 5 years latter and I am waiting on my 27th minute for Azure Redis to be created..
-
Ryan Byrne (Europe) commented
France Central, standard redis cache (C1) - Just under 40 mins to deploy via Terraform.
To echo everyone else this is really hampering deployment patterns.
-
Alexander Trauzzi commented
These start times are unreasonable. Move this service to linux and pass down some savings. I do not understand why you would run redis on Windows?!
-
Oskar Kirmis commented
Europe West, same. Even the smallest instance takes at least 30-45 minutes to spin up.
Ye Gu wrote:
"Azure Cache currently runs on Windows VM's and the creation time is largely determined by how fast the underlying VM's become available on-demand."At least for Redis, it might be worth to consider alternatives for the underlying OS. And I don't see why this is "under review" 5 years later :(
-
Anonymous commented
UK south, it took 65 minutes to deploy redis......
-
Ye Gu commented
We understand that the Azure Cache creation time is long and that's been frustrating to you and us. Azure Cache currently runs on Windows VM's and the creation time is largely determined by how fast the underlying VM's become available on-demand. We hope that Windows server 2019 will help improve this situation. Otherwise we'll have to look at alternative VM types.
-
Anonymous commented
Has this even been investigated?
Even a workaround (e.g. if there are setting config values that makes a difference, or any API call for checking status beforehand) or explanation would be better than knowing it's on the backlog.
I would be surprised if you don't know more about why this is the case at this point, and it would be helpful if that information was the top MSFT post here. It's hard to go to internal clients with 'Microsoft are working on it for 4 years', even if that's a valid reason.
EDIT: 4 years, not 5. I can't count.
-
Darren Kattan commented
This is still a problem. Any updates?
-
Hal commented
+1, Australia East is slow as well. Affecting productivity here.
-
MikeN commented
Agreed that this is ridiculously slow. This is my first time deploying this and i've been waiting over 55 minutes in US East.
-
Fraser commented
Is this still on the backlog? Can't see any MS updates on this issue. Seems like everyone has the same issue that this is the longest deployment time of a pipeline. Have tried powershell, ARM template, manual and all of these are too long.