Predicting/Regression Calculation For Server Utilization Power/CPU/Mem

I'm not quite sure I understand exactly what you are trying to do. Your title mentioned regression (without mentioning what kind of regression equation you want to use), so I tried a regression approach to the problem:

1) You did not give a regression equation (I'm not sure if regression is even what you are trying to do), so I assumed power load would be a simple linear regression against %CPUutil and %Memutil load=m2*%CPUutil+m1*%MEMutil+b
2) I normally perform linear regression in the spreadsheet using the LINEST() function (https://support.office.com/en-us/art...a-fa7abf772b6d ), so I entered =LINEST(B4:B877,E4:F877,TRUE) across H1:J1 (note that LINEST() is an array formula and must be entered as such by confirming with ctrl-shift-enter and not just enter).
3) In N6 I entered =$J$1+$I$1*N$5+$H$1*$M6. Note the mix of relative and absolute references.
4) Copy/paste/fill N6 into the rest of the table.

Is that close to what you are trying to do?

I must not be understanding, because I thought that was what my regression analysis would give you -- a way to calculated load at any given %CPU and %MEM.

Using row 421 as an example:
Measured values: E421=13.44, F421=13.25 B421=4.28 kW.
Regression equation using E421 and F421 as inputs: Estimated load of 4.26 kW.
Estimated load at 70% for each input value: Estimated load=6.14 kW.

Can you help me understand why that analysis is not correct and what you expect to get out of the analysis?

I put the =linest formula into H1:J1. All 3 cells returned the same number,

That does not sound right, I get three different numbers from the regression. How are you entering the LINEST() function? I would expect
1) Clear H1:J1
2) Select H1:J1
3) Type in the LINEST() formula as given above. Confirm entry with ctrl-shift-enter (not just enter).
I get 0.00308 in H1, 0.0302 in I1, and 3.813 in J1.

If you get a value error, make sure none of the entries has been cleared/deleted or changed to text.