FastJet

FastJet
Name	FastJet
Developer	CERN; original authors: Matthias Cacciari and Gavin P. Salam
Released	2006
Programming language	C++
Operating system	Linux, macOS, Windows
Genre	Computer program (particle physics), jet clustering
License	GNU General Public License

FastJet

FastJet is a software package for jet clustering and analysis widely used in ATLAS, CMS, ALICE, and LHCb collaborations at the Large Hadron Collider. It provides implementations of modern jet algorithms, tools for jet area subtraction, and utilities for collider phenomenology employed by researchers from institutions such as CERN, DESY, SLAC, and Fermilab. FastJet has been cited in publications spanning experimental measurements, theoretical calculations, and Monte Carlo generator validation involving PYTHIA, HERWIG, and SHERPA.

Overview

FastJet implements a set of clustering algorithms derived from theoretical work by researchers associated with Les Houches workshops and conferences such as Rencontres de Moriond and ICHEP. It supports sequential recombination algorithms like those motivated by studies from Sterman–Weinberg-style jet definitions and formal developments linked to the QCD community including contributions from Stefan Catani, Gavin Salam, and Matthias Cacciari. The package integrates with event records produced by HEPMC, ROOT, and generator frameworks used by ATLAS, CMS, and LHCb analyses. FastJet's design reflects comparisons made in work by groups at Brookhaven National Laboratory, Lawrence Berkeley National Laboratory, and university groups at Oxford University, University of Cambridge, and MIT.

Algorithms and Features

FastJet provides implementations of algorithms such as the kT algorithm, anti-kT algorithm, and Cambridge/Aachen algorithm originating from literature by authors including Seymour, M.H.-style analyses and later refinements by Cacciari and Salam. It includes area-based pileup subtraction methods influenced by techniques employed in ATLAS and CMS pileup mitigation studies, and grooming tools comparable to methods introduced in work by Butterworth, J.M. and Dasgupta, M.. FastJet supports jet quality measures used in searches reported by collaborations including Tevatron experiments at Fermilab and heavy-ion studies performed by ALICE. Plugins extend functionality to algorithms from JADE (algorithm), SISCone, and other novel proposals presented at Snowmass (physics conference).

Software Architecture and Implementation

FastJet is implemented in C++ with a modular API. Its architecture separates core clustering engines, area calculators, and user-facing wrappers to facilitate integration with analysis frameworks like ROOT and event formats such as HEPMC. The codebase uses computational geometry techniques akin to approaches discussed in literature from Computational Geometry research groups at ETH Zurich and algorithmic efficiency optimizations similar to work at Stanford University and University of California, Berkeley. Extensions are packaged as plugins; examples of plugin use appear in software stacks maintained by CERN experiments and research groups at Imperial College London and University of Oxford.

Usage and Interfaces

Analysts employ FastJet via a C++ API, Python bindings used in notebooks hosted on platforms like Jupyter Notebook, and through integrations with analysis ecosystems at CERN and university clusters. Workflows combine FastJet with event generators such as PYTHIA, HERWIG, and SHERPA, detector simulators like GEANT4, and data analysis tools including ROOT histograms and Matplotlib for visualization. Tutorials and training sessions have been run at schools organized by CERN Summer Student Programme, Les Houches Summer School, and CMS Open Data outreach events.

Performance and Benchmarks

FastJet emphasizes algorithmic scaling and has benchmark comparisons against implementations used in experiments like ATLAS and CMS and by theory groups at IPPP (Institute for Particle Physics Phenomenology), CERN Theory Department, and Perimeter Institute. Benchmarks consider jet multiplicity scenarios typical of LHC collisions at 7, 8, 13, and 14 TeV, and heavy-ion conditions explored by ALICE. Performance metrics reported by collaborations compare clustering time, memory usage, and pileup correction throughput, with studies published in journals such as Journal of High Energy Physics and presented at conferences like ICHEP and EPS-HEP.

Development, Licensing, and Community

FastJet is distributed under the GNU General Public License and developed collaboratively by maintainers associated with CERN, university groups at University College London, McGill University, and contributors from laboratories like Fermilab and DESY. Development activities occur on version control platforms used by scientific software communities, and contributions have been discussed at meetings including LHCP and CHEP. The user community comprises analysts from ATLAS, CMS, ALICE, LHCb, theory researchers at Institute for Advanced Study, and educators running training at schools such as Les Houches Summer School.

Category:Particle physics software