Back to Programs

Science and medicine

CERN-HSF

Umbrella for Particle Physics-related projects

aicc++container orchestrationdata analysismachine learningpython

Participation history

10 GSoC years

2026

26 projects

Official year page

ML Inference on heterogeneous architectures using SOFIE

SOFIE generates standalone C++ inference code from trained neural networks, letting HEP frameworks run ML models without heavy external dependencies....

Publication of OmniFold Weights A Standardized Format for ML Based Unfolding Results

OmniFold produces per-event weights that cannot be reused without the original analysis environment. The original OmniFold paper deposited input...

Enabling Differentiable Rendering via AD of Parallel C++ STL Primitives in Clad

This project proposes to enhance Clad, a Clang-based automatic differentiation (AD) tool, by leveraging its liveness analysis to automatically...

Fine grained storage for the DUNE experiment

DUNE’s fine-grained Phlex/FORM I/O model stores data products across many independent RNTuple containers, which can increase read-side overhead...

BOA Constrictor: Benchmarking Neural Backbones for HEP Data Compression

The LHC at CERN generates 30 petabytes of physics data every year. Current compression tools like LZMA treat this data like any other file and miss...

EKO Oxidation: Maturing the Rust Core of EKO

EKO (Evolution Kernel Operators) is a foundational Python library used in high-precision particle physics that currently faces performance...

HTCondor Live Tools with Solution Memory

This project enhances the Archi framework by enabling real-time interaction with HTCondor job scheduling infrastructure and introducing a persistent...

Apache Arrow Interface for PODIO

This project aims to implement a high-performance Apache Arrow backend for PODIO. Currently, PODIO data is primarily persistent in ROOT and SIO...

Characterizing I/O Performance for ML Data Loaders at Scale using Darshan

Modern machine learning workflows in High Energy Physics increasingly rely on large-scale datasets, where the data-loading stage can become a primary...

Automated Software Performance Monitoring for the ATLAS experiment

At present, software regression testing at ATLAS is predominantly performed manually via inspection of metrics on the ATLAS project management board....

Clad as a First-Class Gradient Engine in LibTorch

This proposal targets the HSF 2026 idea Clad as a first class gradient engine in libtorch, a 350-hour project intended to make compiler-generated...

Parallel Processing Improvements in Julia Jet Reconstruction

This project aims to improve the parallel scaling of JetReconstruction.jl, a Julia package for jet clustering in high-energy physics. While the...

S3 Backend for RNTuple

RNTuple is ROOT's next-generation columnar data format for High Energy Physics. Its storage layer is decoupled from its format layer, enabling new...

Implement CppInterOp API exposing memory ownership and thread safety

CppInterOp is a Clang-based library that provides an API to examine C/C++ code, enabling language bindings such as cppyy to interact with C++ code at...

Debugging CVMFS Processes in D-State

This project proposes to build diagnostic tooling for extracting debugging information from CVMFS processes stuck in uninterruptible sleep (D-state),...

CARTopiaX: Extending a Next-Generation Platform for Computational Cancer Biology

This project proposes a comprehensive strategy to extend the CARTopiaX platform within BioDynaMo to replicate a wider range of in vitro observations...

AI-Assisted PostgreSQL Optimization Using EXPLAIN Plans

Nopayloaddb is used to serve time-sensitive conditions data for high-energy physics (HEP) experiments, so performance is critical: slow IOV lookups...

Debuggable Installations for Spack Packages

Spack is a dominant HPC package manager that automates scientific software installation, widely used in the HEP community. Spack builds software in...

Data Representation Optimization for ML-based calorimeter simulation

This project aims to develop a pipeline for producing calorimeter shower datasets in point-cloud form for a given detector and various types of...

Integration of Combine with FCCAnalyses

The Future Circular Collider (FCC) is a flagship proposal for a next-generation post-LHC particle accelerator to be built at CERN. As the FCC project...

Enabling Sustainable ARM Support for the Belle II Software

ARM-based processors are becoming increasingly common in HPC environments, as in the AWS Graviton, Fujitsu A64FX (one of the world's fastest...

Generative-AI Assisted Testing of Complex Stacks of Spack Packages

The Problem: Spack’s combinatorial flexibility makes exhaustive CI testing computationally impossible. Current CI focuses almost exclusively on...

Constellation Web Interface for Controlling and Monitoring Experimental Setups

During late-night beam tests, physicists must leave detector halls to find workstations with Qt installed just to check satellite status. This...

Consolidate and advance the GPU infrastructure in Clad

Automatic Differentiation(AD) is a set of techniques to evaluate the derivative of functions specified by a computer program. Clad is a Clang-based...

Transformer based Reconstruction for LHCb PicoCal Calorimeter

Current LHCb calorimeter reconstruction uses rule-based clustering without exploiting timing or learned representations. The PicoCal upgrade will...

AI-Accelerated Signal Reconstruction for the ATLAS Tile Calorimeter at the HL-LHC

The ATLAS Tile Calorimeter (TileCal) currently reconstructs energy using a linear signal-processing method. For the Phase-II HL-LHC upgrade, readout...

2025

25 projects

Official year page

Implementing Debugging Support for xeus-cpp

This proposal outlines integrating debugging into the xeus-cpp kernel for Jupyter using LLDB and its Debug Adapter Protocol (lldb-dap). Modeled after...

Support usage of Thrust API in Clad

This project proposes to enhance Clad, a Clang-based automatic differentiation (AD) tool, with support for NVIDIA's Thrust library. By enabling...

Highly Granular Quantization for CICADA

The CICADA (Calorimeter Image Convolutional Anomaly Detection Algorithm) project aims to detect anomalous physics signatures without bias from...

Implement and improve an efficient, layered tape with prefetching capabilities

Automatic Differentiation (AD) is a computational technique that enables efficient and precise evaluation of derivatives for functions expressed in...

Using ROOT in the field of genome sequencing

This project will improve how we store and analyze DNA sequence data by using ROOT's newer RNTuple format. We'll work in two main steps: First, we'll...

Integrating Support for Google XLS in hls4ml

At the end of Moore’s law, building domain-specific computer architectures is considered the next step in improving program efficiency. Hls4ml is an...

Physics-Constrained Autoencoders: Intelligent Compression in High Energy Physics

The Large Hadron Collider (LHC) hosts high-energy physics (HEP) experiments, such as ATLAS, which generate over 1 petabyte of data every second....

Evaluating CVMFS for Machine Learning Model Distribution

Physicists at the LHC need to perform ML inference on massive amounts of data. Currently, the bookkeeping of ML model files is an unsolved problem....

Enable automatic differentiation of OpenMP programs with Clad

This project aims to extend Clad, a Clang-based automatic differentiation tool for C++, to support OpenMP programs. By enabling Clad to parse and...

Background Enrichment augmented Anomaly Detection (BEAD)– Contrastive VAE & Transformer Architecture

This project aims to enhance the BEAD framework for anomaly detection in high-energy physics by developing a new multi-stage architecture that...

Benchmarking Sustainability of Classical & Quantum Algorithms for particle trajectory reconstruction

The Large Hadron Collider (LHC) particle track reconstruction faces increasing computational demands, driving exploration into novel methods like...

Implementing a deprecation system for Ganga

This project aims to introduce a structured, automated deprecation system for the Ganga project. Ganga currently lacks a formal process for phasing...

RNTuple in JSROOT

This project addresses the lack of RNTuple support in JSROOT, a critical gap hindering browser-based visualization and analysis of high-energy...

Extending support on custom kernels with virtme-ng

The CernVM File System - CVMFS distributes LHC experiment software and conditions data to the LHC computing infrastructure. The client application is...

Implement activity analysis for reverse-mode differentiation of (CUDA) GPU kernels

GPUs and CUDA have revolutionized computing beyond graphics rendering, becoming essential in various fields due to their parallel processing...

Rucio WebUI Revamp

As a final-year Software Engineering student at Kyiv Polytechnic Institute, I propose transforming Rucio WebUI into a more powerful, intuitive, and...

Intelligent Log Analysis for the HSF Conditions Database

The nopayloaddb tool works as an implementation of the conditions database reference for the HSF. The project is based upon Nginx, Django, and the...

Improve automatic differentiation of object-oriented paradigms using Clad

The project addresses a limitation in Clad, a Clang plugin for C++ automatic differentiation, which currently cannot fully support non-trivially...

TMVA SOFIE - Enhancing Keras Parser and JAX/FLAX Integration

The SOFIE (System for Optimized Fast Inference Code Emit) project is an initiative within the TMVA (Toolkit for Multivariate Data Analysis) framework...

Enhancing LLM Training Efficiency with Clad for Automatic Differentiation in C++

Training Large Language Models (LLMs) is computationally expensive, often bottlenecked by the performance limitations of Python-based frameworks....

TMVA SOFIE - GPU Support for Machine Learning Inference

ROOT's TMVA( Toolkit for Multi-Variate Analysis ) has SOFIE (System for Optimized Fast Inference code Emit) which offers a parser capable of...

GreenML@CERN – A Comprehensive Framework for Energy-Efficient Scientific Machine Learning

The project addresses the rising energy consumption of ML workloads in high-energy physics (HEP) at CERN. While models like Baler and ATLAS top...

CARTopiaX: an Agent-Based Simulation of CAR T-Cell Therapy built on BioDynaMo

CAR T-cell therapy is a form of cancer immunotherapy that engineers a patient’s T cells to recognize and eliminate malignant cells. Although highly...

Data Representation Optimisation for Generative Model-based Fast Calorimeter Shower Simulation

This project addresses the computational cost of full Geant4 calorimeter shower simulation by constructing compact, readout-segmentation–independent...

Neural (De)compression for High Energy Physics

In high-energy physics experiments such as those conducted at CERN's ATLAS project, the volume of data generated is immense, reaching up to 60...

2024

17 projects

Official year page

Integrate a Large Language Model with the xeus-cpp Jupyter kernel

Xeus-cpp, a Jupyter kernel for C++, is built upon the native implementation of the Jupyter protocol, xeus. This setup empowers users to interactively...

Implement Differentiating of the Kokkos Framework in Clad

The goal is to implement the differentiation of the Kokkos framework including the support of Kokkos functors, lambdas, methods such as parallel_for,...

MCnet/Rivet+YODA CI-based Multi-Arch Docker and Code Coverage

This proposal aims to enhance the MCnet projects, Rivet and YODA, by automating the generation and deployment of Docker images across multiple...

Geant4-FastSim - Fast inference of Diffusion models

Fast and accurate simulation of high-energy physics experiments is crucial for advancing our understanding of fundamental particles and forces in...

SOFIE Developments - Inference Code Generation for Deep Learning models

SOFIE aims to streamline the integration of deep learning models into the ROOT scientific software framework through efficient inference code...

Improving performance of BioDynaMo using ROOT C++ Modules

BioDynaMo, a powerful agent-based simulation platform, utilizes ROOT for functionalities like statistical analysis, random number generation, and IO....

Any Collection in Data Model Explorer

EDM4hep offers a powerful set of objects to describe event data from simulation to analysis. At any step of this chain, one can inspect the event...

Superbuilds for ROOT

ROOT is a framework for data processing, born at CERN, at the heart of the research on high-energy, molecular and laser physics, as well as in...

Julia interoperating with HEP C++ libraries

ROOT is CERN's data analysis framework, which is widely used for HEP applications. Currently, the UnROOT package supports reading the ROOT files in...

Containerization of CBACK backup system

The project aims to enhance CERN Storage's backup orchestration system, CBACK, by introducing a Scheduler Agent to dynamically manage job agent...

Add support for consteval and constexpr functions in clad

Clad is a Clang library which allows automatic differentiation (AD) of functions. Clang provides the necessary facilities for code transformation....

CVMFS: Benchmarking of new FUSE features

The project focuses on integration of FUSE-T kextless module for CVMFS on macOS. This initiative is motivated by the computational demands of...

Lossless compression of raw data for the ATLAS experiment at CERN

The goal of this project is to study the performance and effectiveness of various compression algorithms, specifically on ATLAS RAW data. The ATLAS...

Reverse-mode automatic differentiation of GPU kernels using Clad

Nowadays, the rise of AI has shed light into the power of GPUs. The notion of General Purpose GPU Programming is becoming more and more popular and...

Incorporating a Large Language Model in Ganga to assist users

Ganga is a tool for composing, running, and tracking computing jobs across a variety of backends and application types. Ganga primarily runs as a...

(MCnet/LHAPDF) Online dashboard and data-visualisation for parton density functions

At the Large Hadron Collider (LHC), protons collide at the highest energies achieved by humanity, unravelling the particles within. To decode these...

Machine Learning in Julia for Calorimeter Showers

The calorimeter, a key detector in Large Hadron Collider (LHC) experiments, measures the energy of particles interacting with detector materials....

2023

16 projects

Official year page

Extend the Cppyy support in Numba

Numba is a JIT compiler that translates a subset of Python and NumPy code into fast machine code. Cppyy is an automatic, run-time, Python-C++...

Deep Autoencoders for Scientific Data Compression

Development and Deployment of a Lossy Compression Tool 'Baler' that uses deep autoencoders to flexibly compress scientific data. Use Baler to...

MCnet/LHAPDF - Test Suite and Coverage for Parton Density Calculations

The LHAPDF C++ library plays a crucial role in the Large Hadron Collider (LHC) programme by providing it with Parton Distribution Function (PDF) data...

Geant4-FastSim: Transformer-based architecture for fast shower simulation

Calorimeter is one of the most important components of the Large Hadron Collider (LHC) experiments, detecting energy loss of particles after...

Improve automatic differentiation of object-oriented paradigms using Clad

I would like to apply for the Improve automatic differentiation of object-oriented paradigms using Clad project. This project is mainly focused on...

Extend and improve the CERNBox Notifications platform

Reva is the core of the CERNBox service. It is a middleware framework, providing interoperability between storage and application providers using a...

Belle II Event Display with Phoenix

The proposed project aims to improve the accessibility and user experience of Belle II by developing a web application for event display. Currently,...

Graph Construction for Charged Particle Tracking using Point Cloud Networks

Graph neural networks (GNNs) have shown great potential in solving the complex problem of charged particle tracking, which is an essential task in...

Estimating the energy cost of scientific software

The Large Hadron Collider (LHC) experiments generate massive datasets composed of billions of proton-proton collisions. The analysis of this data...

Leverage Spark Connect for interactive data analysis in Jupyter Notebooks

CERN uses a service called Service for Web Analysis (SWAN) to perform analyses on scientific data, which is built on top of Jupyter notebook....

ROOT - TMVA SOFIE Developments - Inference Code Generation for Deep Learning models

Toolkit for Multivariate Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software framework, used in...

Enable cross-talk between Python and C++ kernels in xeus-clang-REPL by using Cppyy

xeus-clang-REPL is a C++ kernel for Jupyter notebooks using clang-REPL as its C++ Interpreter. Cppyy is an automatic, run-time, Python-C++ bindings...

Improving the sense of scale and navigation in high energy physics event visualization

This project aims to improve the scale and navigation inside the Phoenix application which is basically an Angular-based web application written in...

Real-time lossless data compression for the FASER experiment

This project aims to add support for real-time lossless compression libraries and algorithms in the data acquisition system that is being used for...

Implement vector mode in forward mode automatic differentiation in Clad

In mathematics and computer algebra, automatic differentiation (AD) is a set of techniques to numerically evaluate the derivative of a function...

RooFit - Pythonic interaction with the RooWorkspace

RooFit uses RooWorkSpace to manage its models. Currently, the syntax to interact with RooWorkspace is completely different from Python syntax, which...

2022

21 projects

Official year page

[IRIS-HEP] Uproot + Dask

This project aims to create an API for users that can provide the data in ROOT files directly in a “delayed” form that is supported by Dask. It will...

ROOT - Automatic conversion of data stored in TTree form to RNTuple

TTree is the ROOT’s legacy columnar storage that has been used to store more than 1 exabyte of high-energy physics data during the last 25 years....

Acts GPU R&D - Optimization of GPU tracking pipeline

Traccc is a demonstrator for GPU tracking algorithms. This project focuses on optimizing GPU utilization of traccc using CUDA Multi Process Service...

Implementation of a Python library that generalizes the CLUE clustering algorithm

The CLUE algorithm is a clustering algorithm written by Rovere et al. in C++, which is used in High-Energy Physics to reconstruct particle showers by...

ROOT - TMVA SOFIE Developments - Inference Code Generation for Deep Learning models

Toolkit for Multivariate Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software framework, used in...

Add Initial Integration of Clad with Enzyme

Clad is an open source plugin to the Clang compiler that detects from the parsed Abstract syntax tree, calls to differentiate a defined function,...

ROOT - Machine Learning Developments - Batch Generator for training machine learning models

Toolkit for Multivariate Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software framework, used in...

Vectorized Linear Algebra Implementation for ACTS

This project is for implementing a vectorized linear algebra backend. The current linear algebra implementation uses scalar instructions, but we can...

Symplectic Integrators

The Geat4 toolkit lacks numerical integrators that preserve energy over the long course of simulations. Specifically, the g-2 physics experiment at...

A concurrency model for the monitoring in Ganga

Ganga routinely handles a tremendous amount of computational tasks simultaneously and keeping track of their status is prone to lagging behind with...

Geant4-FastSim - Memory footprint optimization for ML fast shower simulation

Geant4 is a highly accurate and detailed simulation toolkit used for simulating the passage of particles through matter. Due to its strict precision...

Project Proposal Geant4-FastSim - ML pipeline optimization using Kubeflow

This is the tentative proposal submission for the Geant4-FastSim - Building an ML pipeline for fast shower simulation. This proposal discusses a...

Interfacing PODIO to Julia

Currently, two major languages are used in high-energy physics (HEP): C++ for numerically intensive code, where execution speed is critical, and...

Geant4 - Performance Data Visualization using d3.js

The project aims at monitoring performance and improving the data visuals used to represent the performance report for Geant4 which is a toolkit for...

Optimize ROOT use of modules for large codebases

ROOT is a data analysis framework designed to handle large amounts of data with high performance. This proposal aims at optimizing the performance of...

Advanced Belle II Software Validation

The international Belle II collaboration handles huge amounts of data from the Belle II detector, relying on custom software for simulation,...

Rucio WebUI Revamp

Rucio has proven its potential to be used for providing functionality to scientific collaborations to organize, manage, monitor, and access their...

TMVA Developments - Improve Python interface for TMVA

Toolkit for Multivariate Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software framework. The...

Electromagnetic Cluster Finding on GPUs

The goal of this project is to re-implement the current clustering algorithms using SYCL for higher-throughput processing and handling future...

Revamped Testing Infrastructure for Phoenix

This project aims to introduce a new testing strategy so that we encourage Behaviour-driven Development here, at Phoenix — the official web event...

MadAnalysis 5 - Integration of theoretical uncertainty calculation with multi-weight integration

The purpose of this project is to enable multi-weight calculations in the MadAnalysis 5 platform. The current implementation only uses the nominal...

2021

25 projects

Official year page

OpenForBC

The goal of the project is to develop and test a suite of ready-to-run benchmarks to measure the performances of various partitioning options on...

Add support for in-browser interactive averaging of physics results

The Heavy Flavour AVeraging (HFLAV) group is responsible for collecting and combining measurements made at different High Energy Physics (HEP)...

Accuracy and parallel computation in parton density calculation

This project is about adding optimised and accurate high-order interpolators to the LHAPDF C++ library. This is required for adding support for...

Add numerical differentiation support in Clad

In mathematics and computer algebra, automatic differentiation (AD) is a set of techniques to numerically evaluate the derivative of a function...

Upgrading the Ganga Graphical User Interface

Ganga is a tool to run data analysis jobs along with managing associated data files easier. Ganga also allows for submission,...

Single precision support for GPU acceleration in VecGeom

The project aims to implement single-precision floating-point support for GPU acceleration in the VecGeom library. This will be achieved by first...

Implementing an application for visualizing the LHCb DAQ network

The supercomputer that is currently constructed at the LHCb will allegedly be the largest real-time data acquisition system in the world in 2021 and...

Portability for the Patatrack Pixel Track Reconstruction with Alpaka

The Compact Muon Solenoid (CMS) experiment is one of the largest experiments at Large Hadron Collider (LHC) that has been built to search for new...

Add support for functor objects in clad

This proposal aims to add support for directly differentiating functors and lambda expressions and increase clad coverage and overcome its...

Deep autoencoders for ATLAS data compression

Storage is one of the main limiting factors to the recording of information from proton-proton collision events at the Large Hadron Collider (LHC),...

Enhancing the Plotting Functionality in MCnet/Rivet

Rivet is a tool used by particle physicists to test and develop Monte Carlo event generators, i.e., simulation of particle accelerator collisions....

Logging, Unit Test Infrastructure and Precooked Sources for PRMON

PRocess MONitor (PRMON) is a utility programme used by the Worldwide LHC Computing Grid (WLCG) to monitor the resource consumption and performance of...

CernVM-FS preload capability

CernVM-FS (CVMFS) is a service for fast and reliable software distribution on a global scale. Data is organized in repositories. Files and metadata...

TMVA Deep Learning Developments - Inference Code Generation for Batch Normalization

Toolkit for Multivariate Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software framework, used in...

Utilize second order derivatives from Clad in ROOT

In mathematics and computer algebra, automatic differentiation (AD) is a set of techniques to numerically evaluate the derivative of a function...

RooUnfold - Efficient deconvolution using state of the art algorithms

This project aims to address obsolescence in the underlying RooUnfold implementation that would allow the RooUnfold package to act as a lightweight...

New protocols for exascale data management with Rucio

Rucio is an open-source software framework that provides functionality to scientific collaborations to organize, manage, monitor, and access their...

Upgrading the Ganga graphical user interface

Ganga is an open source job management tool for processing and bookkeeping of intensive computational tasks on a wide set of distributed resources....

Inference Code Generation for Recurrent Neural Networks

Toolkit for Multivariate Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software framework, used in...

ROOT Storage of Deep Learning models in TMVA

The Toolkit for Multivariate Data Analysis with ROOT (TMVA) provides a machine learning environment for the processing and evaluation of multivariate...

Implementation of Physical Shape Function

Zfit is a highly scalable and customizable model manipulation and fitting library. Using Tensorflow as its backend, it has been optimised for simple...

Rucio and CS3API to enable data management for the ScienceMesh cloud

In recent years Sync and Share Services have become more relevant than ever in our day to day lives. The scientists at CERN are not an exception to...

RooFit Development - Intuitive Python bindings for RooFit

RooFit is a C++ library for statistical data analysis and for modeling the expected distribution of observables measured in particle physics...

Modern Plotting Machinery for the Large Hadron Collider's Monte Carlo Event Analysis Tool

The Rivet toolkit is a software package for performing data analysis on simulated particle collision events like those in the Large Hadron Collider...

Runtime plugin ecosystem support for OCIS

This project aims to add runtime pluggability to the Reva framework to enhance developer experience at AARNet, ownCloud and CERN. This would involve:...

2020

34 projects

Official year page

Enable Modules on Windows

ROOT has several features that interact with libraries and require implicit header inclusion. This can be triggered by reading or writing data on...

Implementation of Gilbert-Johnson-Keerthi Algorithm for Convex Shapes in VecGeom

The goal of this project is to implement the Gilbert-Johnson-Keerthi algorithm on convex objects in VecGeom along with a ray casting algorithm that...

Manipulation of massive astronomical data using graphs

AstroLab Software has developed Fink, an apache-spark based broker infrastructure, which is able to analyze this large stream of alerts data from the...

Phoenix - Experiment Independent Event Display

Phoenix is a web-based, experiment independent event display framework in JavaScript for visualizing HEP experiment data from multiple experiments....

GeoModelExplorer: improving interactive 3D geometry visualization tool based on Qt3D

Interactive visualization can provide intuitive insights for experiment setups and results in many domains, including High Energy Physics. Therefore,...

Kubernetes operator for XRootD

This project aims to develop a Kubernetes Operator for XRootD, along with its related documentation, in order to ease and fully automate deployment...

Reduce boost dependence in CMSSW

This project has the goal to find and decrease boost dependencies in CMSSW. Modern C++ introduced a lot of new features that were only available...

Falcon: Fast Simulation using Deep Generative Models

Prior to the HL-LHC operation, detector simulations undergo further developments in order to adapt to the increasing amounts of events. Current...

Graph Neural Networks

Graph neural networks is a powerful generalization of a whole class of neural (and not only neural) architectures on the data, organized in a graph....

Fast Merging of RNTuple Data Sets

The RNTuple interface is an experimental I/O subsystem for the ROOT analysis framework. Preliminary benchmarks have demonstrated significant...

TMVA Interfaces - Pytorch

Toolkit for Multivariate Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software framework, used in...

(QMLHEP Proposal) Quple - Quantum Machine Learning Framework for High Energy Physics

This project serves to implement a common framework for applying quantum machine learning algorithms to high energy physics analysis. A major focus...

Extend SkyhookDM programmable object storage with statistics, sort/aggregate or data compaction functions.

SkyhookDM supports dynamic data management in the cloud by enabling data management tasks to be executed directly within the storage. It uses...

Framework for Automating the Retrieval and Analysis of System Information of Grid sites in WLCG

JAliEn introduces automatic containerization for jobs submitted to the WLCG. To fully utilize this feature, and to plan ahead for how new software...

Upgrading the Ganga user interface to use a relational database for persistent storage

This project will replace the old xml-based metadata system with a database, which will be assisted by a caching mechanism that will allow for a...

MCnet/Rivet - Speed and accuracy in the LHC's MC analysis tool

Rivet is a software package for performing data analysis on simulated particle collision events like those in the Large Hadron Collider. This project...

Cosmic-Ray Imaging Studies via Mission-Imagery from Space (CRISMIS)

CRISMIS is an open-source, AI-based tool for the identification and classification of cosmic-ray artifacts in imaging data. Its purpose is to cater...

Intelligent Alert Management System For HEP experiments

This project aims to develop an intelligent and reliable monitoring system for large distributed services to monitor their status and reduce...

MCnet/YODA

Enhance a statistical toolkit for physics analysis interpretation.

Implementation of Graphical User Interface for Ganga

The idea of this project is to create an additional interface for users to interact with Ganga. The new responsive Graphical User Interface (GUI)...

CernVM-FS: Podman cvmfs integration

It has been shown that only a small portion of all the files in a container image is necessary to run the image itself. This is even more accentuated...

Pre-conditioners applied to ROOT compression algorithms

This project aims to validate the possibility of using pre-conditioners in compressing ROOT file formats (TTree and RNTuple); to understand how we...

Native Desktop Application for Rucio

Rucio is a data management framework which provides a multilevel organization, management, and access to large scale data generated during scientific...

Awkward Array GPU Kernels

The goal of this project is to add a GPU backend to Awkward Array 1.x. With such a backend, the library would have a unique capability of processing...

Development of 3D CNN in TMVA

This project is about the development of 3D CNN functionality in TMVA. I shall be developing both 3D convolution as well as 3D pooling layers. 3D...

Building a Python-based framework for Unsupervised Deep Learning applications in strong lensing cosmology

Gravitational lensing has been a cornerstone in many cosmology experiments, and studies since it was discussed in Einstein’s calculations back in...

Deep Learning Algorithms for Muon Momentum Estimation in the CMS Trigger System

CMS is a general-purpose detector at LHC. During a run, it generates about 40 TB data per second. Since It is not feasible to readout and store such...

Support for Rucio Users with Natural Language Processing

The aim of the project is to use Natural Language Processing (NLP) to develop an intelligent bot able to provide satisfying answers to users and...

Local replica of JAliEn central services for research and development

In preparation of LHC's next run, CERN has been developing JAliEn, a successor to their current grid computing middleware, AliEn. To ease development...

Workflow configuration import and validation for AliECS

The data-driven workflow dumps generated by the DPL (O2 Data Processing Layer) cannot be directly imported into AliECS. These workflow templates are...

Integration of Rucio in JupyterLab for SWAN

CERN and the high energy physics community stores their data in various storage, with varying locations and medium (from object storage to magnetic...

End-to-End Deep Learning Reconstruction for CMS Experiment

Developing End-to-End Deep Learning Models and optimizing them for the Reconstruction of single particles, jets and event topologies of interest in...

Hist: histogramming for analysis powered by boost-histogram

The Scikit-HEP project is a collection of several dozen packages intended to facilitate the use of Python in High Energy Physics. One of the major...

Deep-compression for HEP data

Storage is one of the main limiting factors to the recording of information from proton-proton collision events at the Large Hadron Collider at CERN....

2019

29 projects

Official year page

CERN Awkward Array Project

At CERN, the data from LHC collisions requires complex data types and functions to be processed. As a solution, the awkward-array library makes...

Optimisation of the Ganga toolkit in terms of memory consumption and persistent storage.

GANGA (Gaudi/Athena and Grid Alliance) is an interface used by scientists to interface with huge amount of computing power and storage available to...

Proposal for Atlas Experiment - Hasan Öztürk

Athena framework is being upgraded to run in multithreaded environment and the aim of this project is to create a new Atlas performance monitoring...

Novel Applications of Zstandard (ZSTD) compression algorithm to ROOT

This project aims to investigate application of ZSTD within the ROOT framework; benchmark it in comparison to the other algorithms; test it against...

Package manager for Jupyter Notebook / SWAN

This proposal promises to develop a Jupyter notebook extension, that will allow the users to specify python modules (and their respective versions)...

VectorFlow

A C++ adapter API for integrating vectorized components in a scalar workflow. Many FLOP-intensive algorithms may profit from the vector pipelines of...

Tools for Understanding CMS Data Access

Over the course of Run 2, from 2016 to 2018, the CMS detector produced an unparalleled amount of data, resulting in an intricate optimization problem...

Implement Event based Seeding and Multi-Threading

Pursuing the goal of running Allpix-Squared simulation’s events -independent by nature- in parallel, have led to the identification of performance...

Building a Python-based Analysis tool for AWAKE experiment

Building a library that reads a large number of HDF files and builds a database. Add support for searching and loading multiple datasets,...

Kalman Filter in Rust

The Kalman Filter is a method of iteratively predicting the future state of a system based on previous information. Not only is a Kalman Filter more...

Implementation and Optimisation in ACTS of algorithms exposed in TrackML challenge

Porting and Analysis of top solution algorithms from the TrackML challenge to ACTS framework. The algorithms include the combinatorial Mikado...

SIMT to SPMD Translation

High Level Trigger 1(HLT1) is the first and critical stage in software reconstruction of collisions at the LHCb experiment in the Large Hadron...

Experiment independent display framework and data format

Developing a new data format to represent event data, unifying the needs of the different experiments that will make use of it. Improvement on the...

Implement a GlobalModuleIndex in ROOT and Cling

ROOT has several features which interact with libraries and require implicit header inclusion. These headers are often immutable, and reparsing is...

Generative Adversarial Networks for Particle Physics Applications

The project aims implementation of GANs in the Machine Learning toolkit, TMVA of the ROOT framework would be immensely useful because of the advent,...

Proposal for the Implementation of an HDF5 IO Layer for PODIO

PODIO is a C++ library that allows the creation of event data models and efficient I/O code for HEP experiments. It does so by avoiding deep-object...

Python Components for the SIMPLE Grid Framework

The SIMPLE Grid project is an extension of the SIMPLE Framework that combines popular configuration management technologies such as Puppet/Ansible...

Molr - Operational

In the view of LHC Run 3, we want to extend the functionalities of Molr so that it will be ready to use in production to control various operational...

Rucio - Exascale Data Management

Rucio is a data management system that provides the functionality to organize, manage and access a large amount of scientific data (in the order of...

Development of LSTM and GRU layers in TMVA

This project is about development of Long Short Term Memory(LSTM) and Gated Recurrent Unit(GRU) layers in TMVA, both of which belong to a general...

Monitoring DIRAC Components

DIRAC is a highly-scalable software used for accessing distributed resources from various distributed systems. DIRAC’s main contributor is LHCb and...

Real-time conditions data distribution for the Online data processing of the ALICE experiment

ALICE (A Large Ion Collider Experiment) is a heavy-ion detector on the Large Hadron Collider (LHC) ring. It is designed to study the physics of...

CERNBox: Bring Your Own Application

CERNBox provides cloud data storage to all CERN users to store, share and synchronize their data across all devices. It is integrated with variety of...

Creation and usage of disposable Spark on Kubernetes cluster from SWAN notebook

This project aims to develop a Jupyter notebook plugin which deploys Spark required services to a kubernetes cluster on OpenStack cloud at CERN....

Testing framework for Jupyter notebooks

SWAN (Service for Web-based ANalysis) is a cloud data analysis service developed and powered by CERN that provides Jupyter notebooks on demand. It is...

Create a user interface for Ganga that allows for the execution of tasks inside user specified virtual machines.

Ganga is used to execute a user defined computational task on a distributed back-end. Through this project we let the users define the environment in...

Distributed Computing Resources: aggregation, usage, monitoring

DIRAC is an open source interware platform whose roles are submission of jobs, the management of the data produced, to the orchestration of the...

Alert Redistribution System for Fink : an Apache Spark based Broker for Astronomy

A huge volume of data is generated every night by large astronomical telescopes around the world. A robust and scalable software infrastructure is...

Generating Hessians and Jacobians via CLAD

Clad is a C++ Clang compiler plugin that employs automatic differentiation to derive user-defined functions, performing source code transformations...

2018

26 projects

Official year page

FALCON

googledoc Timeline: Before 5/22/2018: Get familiar with ROOT. Get in touch with community. Have a preliminary to-do list of possible optimizing...

Recurrent Neural Networks and LSTMs on GPUs for Particle Physics Applications

Toolkit for Multivariate Analysis (TMVA) is a machine learning toolkit for the ROOT scientific software framework used in many particle physics data...

CernVM-FS powered WebAssembly I/O

The goal of this project is to build a JavaScript client for the CernVM filesystem implemented as a pluggable backend library for the Emscripten...

Generative Adversarial Networks for Particle Physics Applications

Deep Learning Networks have proven to perform well for a wide range of problems, especially those requiring large labelled dataset to learn patterns....

CPU Race for Particle Hunting

Investigate the viability of xtensor, a multi-dimensional array package which aims to bring the API feeling of NumPy to C++, in the ACTS...

Spark3D: Extend Apache Spark to support 3D Spatial Datasets

A large amount of 3D data is generated in High Energy Physics & Astrophysics experiments. To process this data efficiently, one would need...

DIANA-HEP: Analysis Functions:: Implementation and optimization of common particle physics analysis algorithms in a functional paradigm

High Energy Particle physics experiments typically generate a large amount of data. To analyze them requires significant computing resources. This is...

Modular YAMPL

Modular YAMPL The project fundamentally consists in developing a flexible plugin system that allows YAMPL to selectively load essential components at...

Improve multi-threading support for CERN’s Allpix-Squared project using dependency graphs

Allpix-Squared is a free and open-source simulation framework for silicon tracker and vertex detectors written in modern C++. Its goal is to...

Development of the Deep Learning Optimization Algorithms in TMVA.

The existing TMVA submodule has always used gradient descent to update the parameters and minimize the cost of the neural networks. More advanced...

Petabyte-Scale Cloud Storage File Manager

CERNBox is a cloud storage synchronisation service for CERN users: it allows syncing and sharing files on all major mobile and desktop platforms...

Extend clad - The Automatic Differentiation

CLAD is a Clang plugin that implements automatic differentiation. Automatic differentiation avoids usual disadvantages of symbolic and numerical...

ROOT package manager

ROOT is the data processing framework created at CERN - at the heart of the research on high-energy physics. Every day, thousands of physicists use...

Large-scale computing backend for Jupyter notebooks - HTCondor batch job submission and monitoring using the Ganga toolkit

Jupyter Notebook is an interactive computing environment that creates notebooks which contains computer code as well as rich text elements like...

Rucio : Billion-row scalable and flexible metadata

Rucio produces large amounts of metadata for it’s files and datasets which is stored in a central Rucio server. However there is a fixed set of...

Variational Autoencoders on GPUs for particle physics applications

Deep Learning relies heavily on a large number of linear operations. Data parallelism is a consequence of this property of Deep Learning.GPUs are...

Pure-Go XRootD client implementation

This project aims at implementing a pure-Go BSD-3 client library for XRootD using Go builtin features for concurrency and scalability.

HAhRD: DeepReconstruction

One of the challenges faced in Particle Physics Experiment after the collision of particles in LHC is the reconstruction of the events.This includes...

Convolutional Deep Neural Networks on GPUs for Particle Physics Applications

The project's ultimate goal is to provide a GPU implementation for the existing Convolutional Neural Network package within root/tmva. During my...

Monitoring and traceability of jobs using ElasticSearch - DIRAC

The DIRAC Interware can handle hundreds of thousands of “jobs” daily. LHCb (the “beauty” experiment at the LHC) is DIRAC’s main user and developer....

Open-Source Simulations for Gas Detector on Python

Magboltz solves the Boltzmann transport equations with numerical integration in order to simulate the interactions of electrons in gas mixtures under...

GDML I/O for VecGeom Geometry Package

VecGeom is the new geometry library developed within the high-energy physics community, aiming to replace the legacy geometry navigation...

Faster Matrix Algebra for ATLAS

Eigen is a C++ template library for linear algebra that aims for high performance in combination of high reliablity and good compiler support. A lot...

Optimize and Integrate Standalone Tracking Library (SixTrackLib)

The project targets optimizing the data structures and source code of a standalone tracking library, SixTrackLib, written in C. The aim is to...

Distributed Big Data Analysis with TDataFrame

The main objective of this project is to make it easier for researchers/developers to submit distributed jobs for analyzing datasets using TDataFrame...

Configuration of Lightweight Sites Components using Ansible

Lightweight Sites is a project at CERN that enables the quick setup of new sites on the grid. Often, a lot of time is spent in setting up new sites...

2017

20 projects

Official year page

Using Pseudo-random number repeatably in a fine-grain multithreaded simulation

Particle transport Monte Carlo simulations are a key tool for High Energy Physics experiments, including the LHC experiments at CERN. All Monte Carlo...

Jet Clustering Optimizations in Fads

Fads is a fast detector simulation toolkit in Go used for High Energy Physics analyses. The current version is not very scalable and it takes up too...

Convolutional Deep Neural Networks on GPUs for Particle Physics Applications

Toolkit for Multivariate Analysis (TMVA) is a machine learning toolkit for the ROOT scientific software framework. It contains a variety of...

Optimisation of GeantV HPC workload balancing

This is proposal for implementing suggested HPC aware infrastructure with workload balancer.

Big Data Tools for Physics Analysis

Jupyter Notebook is an interactive computing environment that is used to create notebooks which contain code, output, plots, widgets and theory....

New error control methods for integration of trajectories

Track of a charged particle in a non-uniform electromagnetic field can be described by the first-order initial value problem (IVP). Robustness and...

Assess ROOT's asynchronous data prefetching

Assess the performance of ROOT's asynchronous data prefetching compared to standard prefetching mechanism for reading remote ROOT files.

Deep Anomaly

This project is aimed towards building a framework which monitors incoming ATLAS Computing Operations data for anomalies and then autonomously, acts...

Improvements in vectorization and parallelization of ROOT Math libraries

ROOT is a data analysis software used by scientists all over the world; therefore, its efficiency is critical for the scientific community in general...

Updating gopy to support Python3 and PyPy

gopy is an excellent tool which generates (and compiles) a CPython extension module from a go package. And I hope more developers could make full use...

Efficient Python routines for analysis on massively multi-threaded platforms-Python bindings for the Hydra C++ library

‘Effective Python routines for analysis on massively multi-threaded platforms-Python bindings for the Hydra C++ library'. The title of this project...

Machine Learning Project: Multi-target Regression for Particle Physics

Toolkit for Multivariate Data Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software network, used...

Next generation of CMS data replication system

The CMS (Compact Muon Solenoid) is a high-energy physics experiment at the LHC. The CMS PhEDEx(Physics Experiment Data Export) project is responsible...

Automatic code generation for fast inference in the CMS tracking software

Extracting shape information from simulated CMS events Using Tensorflow to train a DNN based on the shape of the hits, detector information and...

Convolutional Deep Neural Networks on GPUs for Particle Physics Applications

The Convolutional Neural Networks (CNNs) are one special type of a deep learning neural networks with an enormous discriminative power for image...

Smart Data Structures in CUDA

The CUDA language, used for programming NVIDIA GPUs, is not dynamic-data-structure friendly since it was designed for working with data structures of...

Integration of TMVA and OpenML platform

OpenML is an open source project that aims to create a novel ecosystem for machine learning experimentation. It is possible to integrate different...

Electron-ion collision in Pythia8

PYTHIA is used for generation of events in high-energy collisions. Originally developed in fortran until Pythia6.4, it was rewritten in C++ starting...

Convolutional Deep Neural Networks on GPUs for Particle Physics Applications

In the last few years, deep learning has led to very good performance on a variety of problems, such as visual recognition, speech recognition and...

Launching Wagon, a WebAssembly interpreter in Go

WebAssembly (wasm) is a low level bytecode binary file format made for execution on the Web, designed as a compilation target for various programming...