CMS High Level Trigger Timing Measurements

Clint Richardson

doi:10.1088/1742-6596/664/8/082045

Abstract

The two-level trigger system employed by CMS consists of the Level 1 (L1) Trigger, which is implemented using custom-built electronics, and the High Level Trigger (HLT), a farm of commercial CPUs running a streamlined version of the offline CMS reconstruction software. The operational L1 output rate of 100 kHz, together with the number of CPUs in the HLT farm, imposes a fundamental constraint on the amount of time available for the HLT to process events. Exceeding this limit impacts the experiment's ability to collect data efficiently. Hence, there is a critical need to characterize the performance of the HLT farm as well as the algorithms run prior to start up in order to ensure optimal data taking. Additional complications arise from the fact that the HLT farm consists of multiple generations of hardware and there can be subtleties in machine performance. We present our methods of measuring the timing performance of the CMS HLT, including the challenges of making such measurements. Results for the performance of various Intel Xeon architectures from 2009-2014 and different data taking scenarios are also presented.

Highlights

The CMS experiment [1] at CERN employs a non-traditional two-level triggering system for the selection of events to store for offline reconstruction
Tests Performed The effects of three main changes with respect to 2012 running were measured in order to derive an estimate of High Level Trigger (HLT) performance in 2015: changes in hardware, changes in instantaneous luminosity, and changes in online reconstruction software
HLT performance in 2015 Combining the information of the tests above, together with the number of cores of each generation of machine which will be present in the HLT farm in 2015 (Table 2), one can derive a timing budget for the HLT

Summary

Introduction

The CMS experiment [1] at CERN employs a non-traditional two-level triggering system for the selection of events to store for offline reconstruction. 3. Tests Performed The effects of three main changes with respect to 2012 running were measured in order to derive an estimate of HLT performance in 2015: changes in hardware, changes in instantaneous luminosity, and changes in online reconstruction software.

Results

Conclusion