GAE Availability

[Mauro Blanco]

Hi, I'm trying to understand GAE's availability.

By reading this I understand that GAE offers 99.95% availability for the “Serving Infrastructure” and “Datastore” components. But also, by reading this I understand that (currently) the actual availability for all components is 98.92%. As I understand, the actual availability is inferior than the SLA's because the SLA only covers two components. Am I right?

Thanks in advance.

[Mauro Blanco]

Just noticed I didn't add links to my post...

Here is againg with the missing links:

---------------------------

Hi, I'm trying to understand GAE's availability.

By reading this I understand that GAE offers 99.95% availability for the “Serving Infrastructure” and “Datastore” components. But also, by reading this I understand that (currently) the actual availability for all components is 98.92%. As I understand, the actual availability is inferior than the SLA's because the SLA only covers two components. Am I right?

Thanks in advance.

---------------------------

[Brandon Wirtz]

Published numbers include MS which is not part of the SLA.

 

In the past 90 days, I have experienced 0 minutes of time when I achieved the 10% of failed request numbers.  This is partly because of resiliency I built in to my app to continue to function even when DataStore and Memcache are read-only, and the fact that Edge Cache handled 40-70% of my requests.

 

I have seen some times when pages served more slowly, or instance count was higher than average.

 

So you your Service level my vary, but I’m at 100% over 90 days.

--
You received this message because you are subscribed to the Google Groups "Google App Engine" group.
To view this discussion on the web visit https://groups.google.com/d/msg/google-appengine/-/H5hrCnVE1VoJ.
To post to this group, send email to google-a...@googlegroups.com.
To unsubscribe from this group, send email to google-appengi...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/google-appengine?hl=en.

[Mauro Blanco]

Thanks for your answer! Just one thing I didn't understand, what is MS? Is the scheduled maintenance?

[Wilson MacGyver]

No, Master Slave datastore.

On Sat, Dec 17, 2011 at 6:12 AM, Mauro Blanco <blan...@gmail.com> wrote:
> Thanks for your answer! Just one thing I didn't understand, what is MS? Is
> the scheduled maintenance?

--
Omnem crede diem tibi diluxisse supremum.

[Mauro Blanco]

Ok, but the SLA also does not cover other components, like Images, Mail, Memcache, Taskqueue, Urlfetch and Users, right?

I'm comparing the System Status with the covered errors from the SLA, maybe it is not comparable info. I'm interested in finding wich components are not part of the SLA.

Thanks.

[Mauro Blanco]

Anyone from Google can explain if my assumption is right? Are other components (apart from MS) from System Status not covered in the SLA, or at least without an associated error rate?

Thanks