Choose timezone
Your profile timezone:
ROOT Parallelisation, Performance and Programming Model Meeting
Description
CERN number: 71400
Extension: 109284483#
old Actions:
* Lorenzo: integrate PR2 of Xavi (#367) about vectorisation * Lorenzo, Danilo: merge VecCore PR once discussion about it is complete - Now the comments are all integrated - Need to discuss further involving Pere if needed
* Danilo: integrate new ctors taking a TEntryList in the TDF
- Make sure to throw an informative message if more than one tree is present in the file(s)
--> This is pending until the creation of TEntryLists is addressed with an Action.
* Gerri, Enric, Xavi, Danilo: identify a chunking (packetising) procedure for MT, MP, collections and trees (perhaps respecting clusters' granularity a la TTreeProcessor) * Xavi: Slide 7 2N debug nasty scaling obtained. Ideas: - With perf count the page faults (Pere's idea). - Artificially increase the work to see if the overhead is less. - Try only one worker. * Xavi: try fma with Ofast before ruling that out * Danilo: investigate fallback mechanism with backquotes for branch names which are not c++ names
New Actions:
Minutes:
o For the conversion from hdf5 to root we agreed that this flow is potentially useful: Pytables -- numpyarrays --> rootnumpy: hdf52root
o Name change: AddColumn --> WithColumn:
- Jim: if the user is not thinking in terms of tables, one could use "Define".
- Will be in femtocode
- Perhaps offering something analogous to Spark would be more coherency
o Parallel Merging
- Potential improvement: use intermediate TMemFiles if an option is being passed
- List of Cleanups: remove input and output files from the list of files, manage deletion by hand. Open a file, remove from the list of files, remember to delete it.
- For trees the merge will not be as faster but it is also the usecase is less common: one merges many files with histos rather than many files with trees as these can be used via a chain.
There are minutes attached to this event.
Show them.