Hi,
Running a workspace with full inspection is a great tool, but it comes with a heavy price.
If - which is often the case - a huge dataset is among the inputs, the "inspection cache" is swamped with unnecessary data. This makes inspection of even minute datasets generated later in the workflow a tedious and slow task.
Would it not be possible to somehow index the cached inspection datasets, so retrieval becomes as fast as the individual dataset size determines, i.e. very fast for small amounts ?
The performance is currently just as bad in the 2017 beta as in 2016 release.
Cheers
Lars I.
