I, like many of you, have been working remotely for many months, and it looks like that will continue for some time. I was surprised not to see a single "catch all" topic that helped share tips for improving the speed of work when working remotely.
Of course, the ideal is that we could do everything in a wholly online environment and have all the latest server and software solutions. But of course, many of us don't have the ideal environment, and still do our day-to-day work over networks that have speed issues, especially on VPNs.
I'm working over a fiber connection from my home, so all the context and solutions I'm posting already assume pretty high end internet speeds - that has never seemed a significant limitation.
So I thought I'd ask: what are your tips for improving speed during remote work, especially over VPN? I'll add several answers to the question that I've found as separate entries, and encourage others to do the same!
Probably not a popular answer, including with me, but I've found that when connecting back to an ArcSDE geodatabase on our network over VPN, ArcMap does geoprocessing tasks considerably faster than ArcGIS Pro over a VPN. (Not sure if it's part of the speed limitation, but our ArcSDE is hosted in a SQL server environment).
We have much of our enterprise data on ArcSDE and published to services. I've found its far faster in our configuration to do geoprocessing against services published to an internal intranet as feature services, than it is to do those same geoprocessing functions against the actual ArcSDE-hosted geodatabase. The SDE is hosted on SQL server in our case. Not sure what the bottleneck is, but happy to have a solution.
This is the fastest solution, but it only really works in ArcGIS Pro. If I'm doing geoprocessing functions against the SDE-hosted geodatabase, that's considerably faster in ArcMap (curiously). But the Pro web-enabled environment shines overall, if you can do editing against feature services.
But for full benefit, that does require services that are set up as feature services (instead of just map services), and requires most editing functionality to be allowed when setting up the services ("Extract" and other functionality needs to be checked in the ArcMap feature service creation window, for example).
Have you encountered an internet disconnection in the middle of editing features through the services? If so, did you find this disruption to lead to any data corruption as the edit might not have been completed?
I have not experienced this sort of disruption. I've kind of assumed that if I saved/committed edits in Pro using the "Save" feature in the edit menu, then I'm good, and if I wasn't successful then I'd be subject to losing the edits. That feels like a kind of binary outcome that doesn't allow for much in the way of data corruption. But I'd certainly be interested in hearing if others had a different experience.
Ideally, we can use a VPN connection focused on remote connecting to another desktop computer in my workplace. In this situation, the data isn't transferred back and forth between my work and home, only the video feed from the remote computer comes to my home, and that makes everything far, far faster. Unfortunately, our organization's IT configuration doesn't generally allow for that as a practical matter.
This is what I'm currently doing as I've found it to be the fastest. Unfortunately, not everyone in my organization has the permissions to do so. Putting ArcMap in Citrix has improved performance for some of those users.
Great idea starting this thread Bob. This recently published Esri support article covers a number of common scenarios encountered when working remotely and reviews some of the methods for isolating problems and identifying appropriate actions. Problem: ArcGIS and VPN: Understanding connection errors, timeouts, and performance issues when work...
I'm working thru a VPN and running a geoprocessing tool. My wifi died for a few minutes, and so did my VPN.
I didn't get any errors, BUT my process has been at 6% for a while (an hour), but it's a large data set, so I'm not sure if it's still compiling or stuck. Let me know what you think.