From 0c96a40355e8fdc7bfb66227810331f47c9e0b24 Mon Sep 17 00:00:00 2001 From: Matthew Feickert Date: Mon, 21 Oct 2024 07:20:56 +0200 Subject: [PATCH] Add note on per-event overhead in xAOD and per-batch overhead in columnar --- talk.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/talk.md b/talk.md index fc9fea8..f0fabbf 100644 --- a/talk.md +++ b/talk.md @@ -283,7 +283,7 @@ from atlascp import EgammaTools * While direct comparison not possible, tests are as close as possible - Only involves `C++` CP tool code (no Python involved) - Uses same version of CP tool - - xAOD includes event store access + - xAOD includes event store access (per-event overhead, paid per-batch in columnar) * Show .bold[substantial speedups] for migrated tools: .bold[columnar is 2-4x faster] than xAOD interface (EDM access dependent) - Time for I/O and connecting columns not included in the performance comparisons (not optimized in the tests, so removed from benchmark) ]