2

I use Chrome and Google Sheets with large spreadsheets which have IMPORTRANGEs, VLOOKUPs and pivot tables. I am looking to improve speed.

For this use which has the bigger impact CPU or RAM?

Perhaps drive speed too (SSD)?

Trying to understand better how much improvement can be made locally compared to what runs Google's side.

  • Related: webapps.stackexchange.com/questions/90765/… – Rubén Apr 4 '17 at 21:17
  • As IMPORTRANGE requires access to the server in order to be ran while VLOOKUP and PIVOT TABLES should be able to run offline but the question didn't specified if the spreadsheet is being used offline or online so I think that this question is unclear and too broad. – Rubén Apr 5 '17 at 21:32
  • Online spreadsheet. – user151646 Apr 7 '17 at 6:37
  • I have just seen this post. webapps.stackexchange.com/questions/75449/… however the current 2 answers are contradictory.. If calculations are mainly done locally, surely being offline would make little difference. – user151646 Apr 7 '17 at 7:00
  • Functions like IMPORTRANGE can't run offline. – Rubén Apr 8 '17 at 13:56
3

My lessons from working with slow Google Sheets - might be partial and very particular to the task:

  1. CPU on your local machine is important -- it is the CPU executing the majority of work. You can see it working during long spreadsheet recalculations. From what I can tell, there is not much impact on RAM, disk or network.
  2. Use ARRAYFORMULAs with VLOOKUPs where possible, instead of copying the VLOOKUPs - or any other formulas - from cell to cell
  3. Use MMULT instead of SUMPRODUCT (plays nicer with ARRAYFORMULA too)
  4. Delete unnecessary rows and columns, do not keep them if they are empty
  5. Avoid conditional formatting applied to large areas of the sheet
  6. Use Safari if you are on Mac, as of 2019-2020 it is about twice as fast as Chrome in executing Google Sheets
| improve this answer | |
1

I had the same question, didn't find an answer, so I ran the test by myself.

I opened the task manager and checked the values of cpu and ram Nothing opened, CPU 3% and RAM 43 % Opened a shortcut to a huge spreadsheets file, with lots of tabs, formulas, vlookups, importranges, conditional formatings... CPU 40-56% and RAM 49%

Looks like we need CPU

| improve this answer | |
0

I can confirm that CPU is most important, circa 8GB ram is enough. I work everyday on Google sheets at work, with big data files like 100k rows, query formulas with 20 import ranges and my cpu is 95%-100% used (i56300u). When I use my home pc same files use 20-25% cpu (ryzen 3900x). More cores, better performance in that case I suppose.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy