Hello,
I am using SciDB to hold 2dimensional matrix representing a bunch of spectras (xcoordinate: spectra seq num, ycoordinate: wave length).
I have two questions:

I want the wave length coordinate to be like numpy.arange(1,1000, 0.5) => y= [1, 1.5, 2, 2.5…999.5, 1000].
I read about the float type in dimensions but then I loose all the benefits of integer dimension (like loading by order and not specifying the index of ycoordinate) . Can you think of a way I can do it comfortably? 
If I have two arrays with different wave length coordinate, for example: y1:[3, 3.5, 4…1000] y2:[15, 15.1, 15.2…30].
Is there any way to do interpolation between them? Or to do interpolation based on a similar to y1 given array?
Thank you very much!
Dalya.