- The paper presents the design and performance of a two-tier trigger architecture, combining a hardware-based Level-1 trigger with a software-driven High-Level Trigger.
- The study evaluates trigger efficiencies and latency, ensuring optimal event selection even under high data rates and challenging LHC conditions.
- The paper emphasizes the system’s evolution and adaptability, laying a foundation for future improvements in high-energy physics data acquisition.
Overview of the CMS Trigger System
The paper provides a comprehensive description of the Compact Muon Solenoid (CMS) trigger system operational during Run 1 of the Large Hadron Collider (LHC). The CMS trigger system is divided into two primary levels: the Level-1 (L1) trigger and the High-Level Trigger (HLT). Each of these components is intricately designed to manage and filter a substantial amount of data to facilitate the detection of significant physics events from the noise.
Level-1 Trigger System
The L1 trigger is a hardware-based system tasked with reducing the interaction rate—approximately 1 GHz for proton-proton collisions—down to 100 kHz, the maximum that the CMS data acquisition system can handle. It operates on information from the calorimeter and muon detector subsystems to make decisions on interesting events that contain specific physics signatures like electrons, muons, jets, or missing transverse energy (\MET). The L1 system is further subdivided into specific subsystems that include the Regional Calorimeter Trigger (RCT), Global Calorimeter Trigger (GCT), and Global Muon Trigger (GMT), all of which perform specialized functions to identify candidate events for further investigation.
High-Level Trigger System
Following L1 processing, the HLT system uses software algorithms running on a computing farm to further refine event selection. The HLT system is capable of reducing the event rate to around 400 Hz suitable for data storage and offline analysis. The software at this stage benefits from full event reconstruction, providing higher-resolution information on physics objects such as leptons, jets, and \MET. The HLT algorithms are designed to closely match offline analysis tools, allowing the replica of complex physics analyses to be conducted online, facilitating a seamless transition from trigger selection to offline analysis.
The system's performance is evaluated through detailed measurements of trigger efficiencies and resolution with respect to various physics objects like muons and electrons. CMS has demonstrated robust trigger efficiency, essential for ensuring that major physics programs, such as the search for the Higgs boson and studies on top quark production, are comprehensively covered.
Additionally, the paper provides an in-depth look at the evolution and operation of the CMS trigger system, highlighting strategies such as prescaling and adaptive changes to the luminance and pileup conditions encountered during data-taking. It also elucidates the technical intricacies behind improving trigger efficiency and maintaining low latency and deadtime, crucial for acquiring high-quality data under increasingly challenging conditions at the LHC.
Implications and Future Developments
The CMS trigger system's ability to efficiently handle large volumes of data from complex collision environments with high reliability underscores its critical role in experimental high-energy physics. As LHC operations evolve, continued improvements to hardware and software components of the trigger will be necessary to accommodate higher interaction rates and more nuanced physics explorations, such as searches for new phenomena beyond the standard model. The experience and data from the CMS Run 1 provide a valuable foundation for these future developments, ensuring that the CMS experiment remains at the forefront of particle physics discoveries.