vtkThresholdingHexahedronFactory: parallelize/speed up

[Janik Zikovsky]

Use multiple cores (if possible) for optimal speed in creating a visualization data set.

[Janik Zikovsky]

(In [12935]) Corrected Q in MakeDiffractionEventWorkspace (to not have 2*pi and to be q = ki-kf, it was kf-ki). Added a Paraview plugin, PeaksReader, to load .peaks file to visualize them. Needed to add a link to the Crystal library. Refs #3232. Added a vtkPeakMarkerFactory to render single-crystal peaks (as points). Not totally finished. Refs #3232. Made vtkDataSetFactory::initialise() take a Workspace_sptr instead of a IMDWorkspace_sptr. This allowed the visualization of a PeaksWorkspace. Still needs tests. Made the MultiDimensionalDbPresenter::getMesh() method return a vtkDataSet* (instead of vtkUnstructuredGrid*) to be more general. Attempted to speed up/parallelize vtkThresholdingHexahedronFactory without much improvement so far. Refs #3271. Made a MockObjects.hh file to reuse mock objects code in VatesAPITest. Fixes #3291. Sped up setUp of EventListTestPerformance; also InstrumentRayTracerTestPerformance. Refs #3217.

[Janik Zikovsky]

In [13450]:

Refs #3271: Sped up vtkThresholdingHexahedronFactory by about a factor of 2.

[Janik Zikovsky]

In [13451]:

Refs #3271: Fix test. The number of voxels created should be equal to binsX*binsY*binsZ, not (binsX-1)*(binsY-1)*(binsZ-1), hence the change in the test.

[Janik Zikovsky]

In [13483]:

Refs #3316 and refs #3420: Made the MDEW .nxs file format closer to the Design Document specification. Lots of bits (sample/run/instrument) are still missing. Adjusted the PV plugins to find the right kind of loader and fail (gracefully) if given a processedNexus or a SNS event Nexus.

Refs #3271: Sped up vtkThresholdingQuadFactory and vtkThresholdingHexahedronFactory furthere.
Time to create voxels sped up by 40x (19 seconds to 0.45 seconds for 100^{3 voxels).
Time to check the signal for thresholding of 500}3 voxels dropped from 19 seconds to 3.4 seconds.
Made vtkThresholdingUnstructuredGridFactory subclass vtkThresholdingHexahedronFactory so that they could share their common code.

[Owen Arnold]

In [13518]:

Refs #3271. Recover functionality that was lost. Need to be able to handle any type of IMDWorkspace. Identify MDHistoWorkspace, and run newer (faster) dataset generation methods, otherwise, rely on generic (slower) dataset generation.

[Owen Arnold]

In [13519]:

Refs #3271. Kill warnings.

[Owen Arnold]

In [13520]:

Refs #3271. Kill warnings.

[Janik Zikovsky]

In [13610]:

Refs #3271: It turns out that the speed up that is specific to MDHistoWorkspace is 2-3% at best! Therefore to simplify the code I've combined the 4 separate functions back into one with a flag for 3D or 4D and they work for all IMDWorkspace-types.

[Janik Zikovsky]

In [13714]:

Refs #3271: vtkMDEWHexahedronFactory is about twice as fast as before. More can be done.

[Janik Zikovsky]

I think most that can be done within the factory is now done. The larger slow downs are in other parts of paraview.

[Stuart Campbell]

Well trying with CNCS_7860 data set and setting some ridiculous number of bins... it seems quite quick (200x200x200 = 0.028s)

[Stuart Campbell]

This ticket has been transferred to github issue ​4118