my costumer plan to migrate from Managed to SaaS. I’m reply to saas-readiness-assessment survey, and one question is if i have cloud provider preference (AWS, Azure, GCP).
What are differences between these providers? Billing, performance, region to choose…
PS: my costumer wants to SaaS in EU, for GDPR privacy,
All three cloud providers are GDPR compliant, so there are no concerns in that area. Regarding billing, performance, etc., there won’t be any issues either, as you don’t have to manage these aspects. Dynatrace takes care of availability, scaling of the platform, etc.
The most significant differences among the three lie in the integration of new technologies. AWS leads in incorporating new tech, followed by Azure and then GCP.
For instance, AWS was the first to adopt Grail, followed by Azure. However, if you’re looking to use Grail on GCP, you’ll need to wait a few months before it becomes available.
Therefore, if Grail is a prerequisite for your client, GCP shouldn’t be your first choice, unless waiting is an option.
I would add one point: the latency between your sites and cloud locations. Although using ActiveGates solves much of these issues from the data ingestion point of view, you will also benefit from lower latency when accessing the Dynatrace UI.
It really depends on where you are. We are based in Portugal, and it’s probably one of the worst case scenarios in Europe.
Also, AWS/Azure/GCP do have several regions in Europe, but Dynatrace is only available in a subset of them. I know there is a map somewhere of where they are, but couldn’t find them. But I do know of these ones:
AWS Ireland
AWS Frankfurt
AWS London
Azure Netherlands
GCP Frankfurt
All of Ours are in AWS Ireland. It’s a 40 ms RTT, so it’s quite noticeable from a UX perspective. It’s quite different than having a 5/10 ms RTT, which is common for the central part of Europe.
Of course, we do have in Portugal all sorts of RTT problems when deploying applications from our clients to the Cloud, so we are quite aware of this impact…
The above site measures it from your browsers perspective, so for the results to be good, they have to be made from your client’s network.
Don’t forget that this works for all AWS regions, and Dynatrace is only present in some. From Italy, Frankfurt is probably the best, but check with your Sales rep.
I would imagine it to be a little bit better, but it might depend on your ISP and location in Italy (Milan? Rome? ?)
Important point: the values given by the https ping site are for a HTTPs latency test, the time for querying an HTTPS URL. It is not a network latency test, which has much lower values. For instance, for the tenants I use, I have a 52 ms network RTT from where I am at the moment, while the https ping is in at 105 ms. So https is double the network RTT for me, but it might be different in your case.