Re: [google-appengine] Re: ~20ms latency between two servlets, same project, is that about right?

[Eric Hubbard]

Thanks for the great info!

I'm running standard and java8.  My latency times were much lower 24 hours
later.  I went from a chaotic..20-50ms...   down to a very stable ~12ms 24
hours later.  Maybe some  environmental things were going at at the time.
These numbers were reported by my code, and also seen by the log traces.




-Eric
http://www.google.com/profiles/eric.hubbard

On Tue, Aug 29, 2017 at 11:03 AM, 'Jordan (Cloud Platform Support)' via
Google App Engine  wrote:

> It really depends on if you are making calls within App Engine Standard
>  or App Engine Flexible
>  as they both
> send/accept requests using different infrastructures.
>
> - The App Engine Standard URLFetch infrastructure typically adds less than
> 1 ms latency overhead to HTTP requests. For URLFetch requests to other App
> Engine Standard services, I am seeing worst case ~8ms of latency in the
> 95th percentile (meaning 95 percent of URLFetch requests are under 8ms).
> Note that we do not guarantee any specific SLA on latency.
>
> - The App Engine Flexible environment uses Compute Engine
>  virtual machines that are behind
> their own gateway. I am seeing ~20ms serving latency for Flex apps in the
> US at the 99th percentile (meaning 99 percent of requests are under 20ms).
>
> There are many factors in App Engine that can be slowing down your ping,
> such as instance startup time (based on your code and runtime), scaling
> settings
> ,
> and instance class
> .
> Also your actual coding language adds to the ping time while it
> accepts/processes/and returns the request. Therefore, to get a more
> accurate ping for only the service-to-service communication time, you
> should do it by looking at the actual time your App Engine service accepted
> the request in the Logs Viewer
> .
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "Google App Engine" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/
> topic/google-appengine/2vx4_sdC8v0/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> google-appengine+unsubscr...@googlegroups.com.
> To post to this group, send email to google-appengine@googlegroups.com.
> Visit this group at https://groups.google.com/group/google-appengine.
> To view this discussion on the web visit https://groups.google.com/d/
> msgid/google-appengine/b0335619-c20c-4938-9994-
> 76549c75f009%40googlegroups.com
> 
> .
>
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-appengine+unsubscr...@googlegroups.com.
To post to this group, send email to google-appengine@googlegroups.com.
Visit this group at https://groups.google.com/group/google-appengine.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/google-appengine/CABOXiu2Sx0P4qzyVLcRYtVvuWbF%2B%2BTs8xVCo3vPsoLKfQ_pQYg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[google-appengine] Re: ~20ms latency between two servlets, same project, is that about right?

['Jordan (Cloud Platform Support)' via Google App Engine]

It really depends on if you are making calls within App Engine Standard 
 or App Engine Flexible 
 as they both 
send/accept requests using different infrastructures. 

- The App Engine Standard URLFetch infrastructure typically adds less than 
1 ms latency overhead to HTTP requests. For URLFetch requests to other App 
Engine Standard services, I am seeing worst case ~8ms of latency in the 
95th percentile (meaning 95 percent of URLFetch requests are under 8ms). 
Note that we do not guarantee any specific SLA on latency.

- The App Engine Flexible environment uses Compute Engine 
 virtual machines that are behind their 
own gateway. I am seeing ~20ms serving latency for Flex apps in the US at 
the 99th percentile (meaning 99 percent of requests are under 20ms).

There are many factors in App Engine that can be slowing down your ping, 
such as instance startup time (based on your code and runtime), scaling 
settings 
,
 
and instance class 
. Also 
your actual coding language adds to the ping time while it 
accepts/processes/and returns the request. Therefore, to get a more 
accurate ping for only the service-to-service communication time, you 
should do it by looking at the actual time your App Engine service accepted 
the request in the Logs Viewer 
. 

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-appengine+unsubscr...@googlegroups.com.
To post to this group, send email to google-appengine@googlegroups.com.
Visit this group at https://groups.google.com/group/google-appengine.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/google-appengine/b0335619-c20c-4938-9994-76549c75f009%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.