Skip to main content

Hello,


Hope this is tagged correctly…


Recently we noticed the response times has increased considerably.


I believe we are using an old approach to get the data and just trying to work out if that is the problem and if so justify spending the time on fixing it or something else…


here is an example of a request
https://services.sentinel-hub.com/ogc/wms/?service=WMS&request=GetMap&layers=TRUE_COLOR_IMAGE&format=image%2Fpng&transparent=true&version=1.1.1&showlogo=false&name=Sentinel-2%20L1C&resX=10m&resY=10m&maxcc=100&time=2023-10-13/2023-10-13%2F2023-10-13/2023-10-13&bbox=12905645.43501498,-3736785.059857274,12878084.550255071,-3764592.4112180974&crs=EPSG:3857


The wait for server response for this on first run was over 25 seconds and second try nearly 19 seconds…


Any thoughts/suggestions would be appreciated.


Regards,

David

Hi @David ,


I checked the logs in our system for the past 10 days and the response time in general was fast, i.e., 88% of the wms requests made by the account linked to amir@MASKED responded within a second and only 0.6% of the requests responded in over 10 seconds (Fig 1). The service response time looks good to me at this point.


Did most of your wms requests take longer than 20 seconds? Which account was used to make those requests? When exactly did you experience the long response time? We can investigate further if we have more info from you.




Fig 1 (time in ms)

Hi,

Thankyou for the quick response. That does look reasonable, I had another look and am unsure if it is the size of the area or not, this area (did use different dates for the results) yielded the results below which are better than I saw yesterday but keen to try and improve if possible as it only seems to have changed recently.

request

service=WMS&request=GetMap&layers=TRUE_COLOR_PREVIEW&format=image%2Fpng&transparent=true&version=1.1.1&showlogo=false&name=Sentinel-2%20L1C&resX=10m&resY=10m&maxcc=100&time=2023-10-13/2023-10-13%2F2023-10-13/2023-10-13&bbox=12910367.674658988,-3754544.48107943,12883909.658194616,-3783925.287597976&crs=EPSG:3857

This is an example, not always… but when it does start to take a while it continues for a period of time

This is what I am seeing

Please understand in many cases it is good and 7 seconds <> is usually fine but of late it has just seemed to take longer

this last one was using the request directly in the browser

Regards


Hi,

Could you please share the first 4 digits of your instance id?


Hi This is it 28b4


Hi,

The layer itself seems ok. I’d believe the long response time comes from the large AOI you’re trying to request. Also, the size of image is 13.6 MB which could require some time to be downloaded depending on the band width.

I’d suggest requesting a smaller area to avoid long response time in this case.


Hi,

My concern is it seems fairly recent that the long waiting for server response has started to occur.

Based on how we are requesting the aoi can it be improved or changed? if so could this offer some level of improvement?

Many thanks,


Hi,

I communicated with our developers and they will investigate a bit further on the issue. I will let you know if they find anything.


Reply