While many parts of the Esri ArcGIS Platform are able to run natively on Macs, such as ArcGIS Python API, AppStudio, Workforce for ArcGIS, and the Web GIS tools, Esri's ArcGIS Desktop products are not supported for native use on Mac OS. Many of my sciences customers ask me how to optimize the performance of their ArcGIS inside their Parallels instances, so I wanted to publish a post with some best-practice recommendations.
MacBooks and MapBook Airs are probably not going to have the power inside needed to run ArcGIS in a performant manner if you're running intensive processes. Starting off on the right foot with a MacBook Pro will help to ensure the best performance possible.
Power down your Mac, and then turn it back on (you can't do the next steps if you're already running/have run Parallels since your laptop has been turned on for some reason).
Before launching Parallels, inside your Mac OS launch your Parallels Desktop Control window.
Click on the Gear/Cog to open the settings
Set Performance to Faster Virtual Machine
Check the box to Enable Adaptive Hypervisor
Check box to Tune Windows for Speed
Set it for Better Performance (and try to always be plugged in when using Parallels)
Use Parallels in Fullscreen instead of Shaddow Mode
Make sure sufficient cores and RAM have been allotted to your Parallels instance
Running Parallels via Bootcamp will also enhance performance when you're working heavily in ArcGIS, allowing Windows to use as many system resources as possible. If you're gearing up for a significant amount of work in ArcGIS, rather than just quickly jumping in and doing things, consider Bootcamp.
If you're using ArcMap, make sure you have the free 64-bit Background Geoprocessing Add-On installed from MyEsri to eliminate out-of-memory errors when running certain GP tools. Remember ArcMap is native x32 bit.
While ArcGIS Pro is native x64 bit, it won't necessarily run faster, it just means it can handle bigger, more complex data. ArcGIS Pro more heavily leverages the GPU onboard, so make sure it's spec'd appropriately. Also, if you're switching back and forth between using Pro in your Parallels and your Mac OS again and again, it can impact the acceleration, having an impact on performance (see recommendation about using in fullscreen )
For those of you running ArcGIS Enterprise (previously ArcGIS Server), consider using the free Python API, which you can use natively on your Mac without Parallels to do the processing Server-side.
Are you a Mac owner who runs Parallels to leverage ArcGIS Desktop? If so, please post any additional tips you've found for enhancing performance in the comments. Let's turn this into a resource for the Community!