SAC R Visualization loading times

[#1]

Hi guys,

building stories and apps in SAC, we often rely on R Visualizations. We noticed that the loading time of R Visualizations is significantly slower than that of the other widgets.

When I watched this video :https://www.youtube.com/watch?v=od-SxJTRk8E this didn't seem to be case.

Can you influence the loading time by changing settings or anything?

[mfoeken]

Hi Valentin,

Are you using the embedded R-server or your own? The 'embedded' server is actually provided by a third party to SAP and obviously, it takes time to move the data and process it. To my knowledge, there is no setting that can influence loading time other than limiting the volume of the data and the complexity of the script.

Kind regards,

Martijn van Foeken | Interdobs

[Former Member]

Thank you Martijn,

we are indeed not using our own server. I will suggest doing that to the team.

Best

Valentin

[mfoeken]

Hi Valentin,

The tricky part of using your own R-server is that this has to be accessible from the internet. So please make sure you implement the necessary security measures while exposing the URL.

Kind regards,

Martijn van Foeken | Interdobs

[Former Member]

Hi Martin,

this is the current discussion that is going on. We are not fully understanding how the data from our on-premise BW system would be moving between the Client, SAC, BW and the R-Server. We have super strict compliance rules and therefore it would be nice to know if the connection can achieved via the Cloud-Connector.

Best

Valentin

[mfoeken]

Hi Valentin,

If you open a story with an R visualization based on a BW data set (is that live or import) and check the network tab in developer tools, what domains are displayed? Here you can see what requests are sent and that would give more clarity on the actual flow of data.

Kind regards,

Martijn van Foeken | Interdobs

[Former Member]

Hi Martin,

it would be a Live-connection ideally. Thank you for the network tab advice! Basically, what we don't want is for our data to be exposed to the internet. I am not a back-end expert but as I understand, the current live connection is procted by the reverse invoke proxy function of the cloud connector. However, the query data is nonetheless "exposed" to the internet isn't it? So even if we use our own R server the data protection of the cloud connector still applies, if I am not mistaken?

[kevin0071]

Push. Has this ever been answered (maybe via PM?). Would also be interested in the security/compliance implications of using RVisualizations. Can anyone help here, is this a non-issue?

Thank you for your help,

Kevin