Skip to content

[ntuple] Address some RNTupleProcessor performance bottlenecks #22593

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
enirolf wants to merge 3 commits into
base: master
Choose a base branch
from
+14 −4
Open

[ntuple] Address some RNTupleProcessor performance bottlenecks #22593

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5e31171
[ntuple] Prevent redundant processor (re-)connection in chains
enirolf Jun 11, 2026
0c7ea38
[ntuple] Remove unnecessary init and connect of single processors
enirolf Jun 12, 2026
d7d839b
[ntuple] Use cached entry counts before calling GetNEntries
enirolf Jun 12, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 1 addition & 2 deletions tree/ntuple/inc/ROOT/RNTupleProcessor.hxx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -615,9 +615,8 @@ private:
/// \brief Get the total number of entries in this processor.
ROOT::NTupleSize_t GetNEntries() final
{
Initialize();
if (fNEntries == ROOT::kInvalidNTupleIndex)
Connect(fFieldIdxs);
Initialize();
return fNEntries;
}

Expand Down
15 changes: 13 additions & 2 deletions tree/ntuple/src/RNTupleProcessor.cxx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -347,13 +347,24 @@ ROOT::Experimental::RNTupleChainProcessor::AddFieldToEntry(const std::string &fi

ROOT::NTupleSize_t ROOT::Experimental::RNTupleChainProcessor::LoadEntry(ROOT::NTupleSize_t entryNumber)
{
ROOT::NTupleSize_t localEntryNumber = entryNumber;
std::size_t currProcessorNumber = 0;
// If the requested entry number is lower than the current entry number, we have to again localise the correct local
// entry number starting from the first processor in the chain. Otherwise, we can continue looking from the inner
// processor that is currently connected, which is much faster when the chain consists of many inner processors.
if (entryNumber < fCurrentEntryNumber) {
Comment on lines +350 to 353

@pcanal pcanal Jun 12, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can't this be speed up in case the entryNumber is less than the fCurrentEntryNumber but more than the starting entry number of the current file? (and/or is the set of lengths cached and thus fast to go through again?)

@enirolf enirolf Jun 15, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, the set of lengths is cached:

root/tree/ntuple/src/RNTupleProcessor.cxx

Lines 360 to 365 in d7d839b

for (unsigned i = 0; i < currProcessorNumber; ++i) {
if (fInnerNEntries[i] == kInvalidNTupleIndex) {
fInnerNEntries[i] = fInnerProcessors[i]->GetNEntries();
}
entriesSeen += fInnerNEntries[i];
}

fCurrentProcessorNumber = 0;
ConnectInnerProcessor(fCurrentProcessorNumber);
}

std::size_t currProcessorNumber = fCurrentProcessorNumber;
ROOT::NTupleSize_t entriesSeen = 0;
for (unsigned i = 0; i < currProcessorNumber; ++i) {
if (fInnerNEntries[i] == kInvalidNTupleIndex) {
fInnerNEntries[i] = fInnerProcessors[i]->GetNEntries();
}
entriesSeen += fInnerNEntries[i];
}
Comment on lines +358 to +365

@vepadulano vepadulano Jun 12, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not for this PR, but in principle we could have a cache vector of number of entries per processor which is filled lazily at discovery time whenever a processor needs to connect to file(s)

@enirolf enirolf Jun 12, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actually this is exactly what is done a few lines down and somehow didn't think to do it here, so thanks for pointing this out :D. Let me quickly add it here as well.

ROOT::NTupleSize_t localEntryNumber = entryNumber - entriesSeen;

// As long as the entry fails to load from the current processor, we decrement the local entry number with the number
// of entries in this processor and try with the next processor until we find the correct local entry number.
while (fInnerProcessors[currProcessorNumber]->LoadEntry(localEntryNumber) == kInvalidNTupleIndex) {
Expand Down
Loading