In the last decade or so, the main attraction at CERN has been the Large Hadron Collider (LHC), which was indispensable to the discovery of the Higgs boson, as well as that of many other particles.[1] The LINAC3 is an integral part of the CERN accelerator complex, as it provides the initial ion b

WinCC-Open Architecture is a toolkit for creating Supervisory Control and Data Acquisition (SCADA) applications. There are about 650 instances of WinCC OA deployed across the accelerator control systems and experiments at CERN.

Greater luminosities of future Large Hadron Collider runs will demand an unprecedented number of event simulations. Computationally that would be an extremely demanding task. Hence new approaches for such undertakes are required.

We have developed a software platform, ABAQUS (Automated Benchmarking of Algorithms for Quantum Systems), that can be used to benchmark the performance of both software frameworks and hardware devices used in the simulation of quantum computers.

HPC systems are becoming ever more important as a data processing resource for the LHC experiments. HPC sites typically use storage systems different from the well-understood HEP storage systems.

On Unix systems (like Linux), the cron daemon allows users to periodically execute specific tasks.

The number of functionalities covered by the Oracle database increases from release to release.

Thus also the possibilities to communicate with it from outside. More and more systems use HTTP requests in the form of REST API calls to interact with other systems.

In this report, we present the Deep Learning generative model GAN for the Higgs bosont 𝒕𝒕̅𝑯(𝒃𝒃̅) process data generation. Initially, a classical GAN model is considered, with Convolutional layers, Batch Normalization layers, and a Leaky ReLU activation function.

BioDynaMo is an open-source, high-performance, and modular agent-based simulation platform. The main goal of this project is to make BioDynaMo more accessible to the public.

At the beginning of my internship, I was introduced to a few tools used at CERN. These tools include the HPCBatch cluster, where I was going to develop the project.

We analyse how the training and performance of VQC models is affected by noise inherent to NISQ devices. In particular, we study the influence of three different types of quantum hardware noise: measurement errors, single qubit gate errors, and two-qubit gate errors (e.g., CNOT gate).

The CERN Search as a Service project provides a platform for providing indexing and search functionality to a wide range of CERN information sources such as Indico, EDMS, SharePoint, Drupal and CERN’s Web pages. The content of these indexes is available via the CERN Search Portal. [1]

This project was completed in the scope of NextGen Archiver (NGA) for WinCC OA SCADA systems. The NGA is a new archiver for WinCC OA that uses a pluggable architecture to support multiple database technologies.

REANA is an open-source reusable research data analysis platform, that allows researchers to run their analyses in remote compute clouds. The analyses use containerised environments and rely on declarative computational workflow specifications.

CERN produces a huge amount of data, such as experimental results, digital documents and multimedia content, which needs to be stored, archived and preserved for many years to come.

The Travelling Salesman Problem is a well-known problem where it is required to figure out the shortest route to visit all the stops. Even though it was formulated almost over a hundred years ago, nowadays it still has relevance and interest.

DUNE - Deep Underground Neutrino Experiment - will be an experiment based in the USA whose main goal will be to study long-baseline neutrino oscillations from an accelerator beam.

Deploying a new open-source based service is an extensive and challenging process, from both technical and user-support perspective.

The Level 1 (L1) trigger at CMS uses coarse-grained information to search for signatures of interesting physics. L1 scouting is a new paradigm for data collection at CMS which could help in the early identification of promising potential signals, independently of any trigger selection bias.

This report investigates how to create reproducible analyses by providing a systemic approach for general users to follow best research practises. It bases on REANA, a dedicated tool for analysis reproducibility developed by CERN.

Nowadays production applications of the Database Applications and Reporting services section (IT-DB-DAR) are running in Kubernetes, taking advantage of the benefits of this technology.

The purpose of this project was to test and analyse the BioDynaMo project’s performance. In order to achieve this, BioDynaMo uses its own benchmark tools. Benchmarking is highly important because it allows us to test the code performance and monitor the effect of the changes.

This report describes how the EOS Storage Resource Monitor was created. It explains which tools were used and the reasoning behind these choices. It describes the deployment model and the visualisations that are displayed.

In the last years the amount of resources that are needed to conclude experiments at CERN grew rapidly, and some computations began to be performed in clouds. To control them we need some tools for cloud costing.

The project is aimed at deploying the ESCAPE DataLake-as-a-Service over the course of the summer. The DataLake-as-a-Service is a service that allows end-users to interact with the ESCAPE Data Lake in an easily-understandable and user-friendly way.

The purpose of this report is to explain in detail the improvements to the Alpaka version of the Patatrack pixel track and vertex reconstruction repository in the form of caching allocators for allocating and reusing device and host memory as well as smart pointers as an interface for memory mana

Quantum technologies have the potential to revolutionise science and society as early as the next five to ten years but require resources that are not mainstream today.

We have developed two quantum classifier models for the ttH¯ (bb¯) classification problem, both of which fall into the category of hybrid quantumclassical algorithms for Noisy Intermediate Scale Quantum devices (NISQ).

Two of the most well-known quantum algorithms, those introduced by Deutsch–Jozsa and Bernstein–Vazirani, can solve promise problems with just one function query, showing an oracular separation with deterministic classical algorithms.

Generative models, and Generative Adversarial Networks (GAN) in particular, are being studied as possible alternatives to Monte Carlo simulations. It has been proposed that, in certain circumstances, simulation using GANs can be sped-up by using quantum GANs (qGANs).

Deep Neural Networks (DNNs) come into the limelight in High Energy Physics (HEP) in order to manipulate the increasing amount of data encountered in the next generation of accelerators.

The High Luminosity Large Hadron Collider (HL-LHC) at CERN will involve a significant increase in complexity and sheer size of data with respect to the current LHC experimental complex.

The Large Hadron Collider (LHC) at the European Organisation for Nuclear Research (CERN) will be upgraded to further increase the instantaneous rate of particle collisions (luminosity) and become the High Luminosity LHC (HL-LHC).

Starting from the idea of Quantum Computing which is a concept that dates back to 80s, we come to the present day where we can perform calculations on real quantum computers.

Network utilisation efficiency can, at least in principle, often be improved by dynamically re-configuring routing policies to better distribute ongoing large data transfers.

Generative Adversarial Networks (GANs) have gained notoriety by generating highly realistic images. The present work explores GAN for simulating High Energy Physics detectors, interpreting detector output as three-dimensional images.

We present the 3DGAN for the simulation of a future high granularity calorimeter output as three-dimensional images.

The race to meet the challenges of the global pandemic has served as a reminder that the existing drug discovery process is expensive, inefficient and slow.

We present MadFlow, a first general multi-purpose framework for Monte Carlo (MC) event simulation of particle physics processes designed to take full advantage of hardware accelerators, in particular, graphics processing units (GPUs).

Network utilisation efficiency can, at least in principle, often be improved by dynamically re-configuring routing policies to better distribute ongoing large data transfers.

The HEPiX Benchmarking Working Group has developed a framework to benchmark the performance of a computational server using the software applications of the High Energy Physics (HEP) community. This framework consists of two main components, named HEP-Workloads and HEPscore.

As part of CERN-GEANT-PRACE-SKA collaboration and in the context of EGI-ACE (Advanced Computing for the European Open Science Cloud ) collaborators are working towards enabling efficient HPC use for Big Data sciences.

The advantage of quantum computers over classical devices lies in the possibility of using quantum superposition effects of n qubits to perform exponential computations in parallel.

In classical deep learning, a number of studies have proven that noise plays a crucial role in the training of neural networks. Artificial noises are often injected in order to make the model more robust, faster converging, and stable.

In an earlier work [1], we introduced dual-Parameterized Quantum Circuit (PQC) Generative Adversarial Networks (GAN), an advanced prototype of quantum GAN, which consists of a classical discriminator and two quantum generators that take the form of PQCs.

Foreseen increasing demand for simulations of particle transport through detectors in High Energy Physics motivated the search for faster alternatives to Monte Carlo based simulations.

With the increasing number of Machine and Deep Learning applications in High Energy Physics, easy access to dedicated infrastructure represents a requirement for fast and efficient R&D.

In this work, we investigate diferent machine learning-based strategies for denoising raw simulation data from the ProtoDUNE experiment. The ProtoDUNE detector is hosted by CERN and it aims to test and calibrate the technologies for DUNE, a forthcoming experiment in neutrino physics.

Quantum Machine Learning (qML) is one of the most promising and very intuitive applications on near-term quantum devices which possess the potential to combat computing resource challenges faster than traditional computers.

Highly precise simulations of elementary particles interaction and processes are fundamental to accurately reproduce and interpret the experimental results in High Energy Physics (HEP) detectors and to correctly reconstruct the particle flows.

The precise simulation of particle transport through detectors remains a key element for the successful interpretation of high energy physics results. However, Monte Carlo based simulation is extremely demanding in terms of computing resources.

A paper that develops a three-dimensional in silico hybrid model of cancer, which describes the multi-variate phenotypic behaviour of tumour and host cells.

Researchers in biology are faced with the tough challenge of developing high-performance computer simulations of their increasingly complex agent-based models.

Motivation: Agent-based modeling is an indispensable tool for studying complex biological systems. However, existing simulation platforms do not always take full advantage of modern hardware and often have a field-specific software design.

This report investigates how to create reproducible analyses by providing a systemic approach for general users to follow best research practises. It bases on REANA, a dedicated tool for analysis reproducibility developed by CERN.

The High Luminosity Large Hadron Collider (HL-LHC) at CERN will involve a significant increase in the complexity and sheer size of data with respect to the current LHC experimental complex.

The Large Hadron Collider (LHC) at the European Organisation for Nuclear Research (CERN) will be upgraded to further increase the instantaneous rate of particle collisions (luminosity) and become the High Luminosity LHC.

The expected increase in simultaneous collisions creates a challenge for accurate particle trackreconstruction in High Luminosity LHC experiments. Similar challenges can be seen in non-HEPtrajectory reconstruction use-cases, where tracking and track evaluation algorithms are used.

The project HIOS aims to investigate ways of providing efficient interfaces for I/O intensive applications.

There is no unique way to encode a quantum algorithm into a quantum circuit. With limited qubit counts, connectivities, and coherence times, circuit optimization is essential to make the best use of near-term quantum devices.

We present studies of quantum algorithms exploiting machine learning to classify events of interest from background events, one of the most representative machine learning applications in high-energy physics.

In particle physics the simulation of particle transport through detectors requires an enormous amount of computational resources, utilizing more than 50% of the resources of the CERN Worldwide Large Hadron Collider Grid.

Deep learning is finding its way into high energy physics by replacing traditional Monte Carlo simulations. However, deep learning still requires an excessive amount of computational resources.

True Random Privacy (TRP) project, developed during the Random Power hackathon 2020,  aims to create a new differential privacy solution for images, embedding a state-of-the-art features description technique.

A massive amount of data is generated by the Openstack cloud services in the format of service logs. Besides timestamps and log level fields, these logs contain additional information useful for pattern analysis.

Invenio is an open source framework, initially developed at CERN, but with many external users and contributors at this moment and prospects of growing even more in the future. Its nature as a digital

Graph Neural Network possess prospect in track reconstruction for the Large Hadron Collider use-case due to high dimensional and sparse data.

Invenio is an open source framework, initially developed at CERN, but with many external users and contributors at this moment and prospects of growing even more in the future. Its nature as a digital

At particle colliders, more data are produced than what the experiments can store for further analysis.  This is why the incoming collisions are processed in real time by a so-called trigger system. At the 

With the pervasiveness of high-speed computers and processors, computer companies are  looking for new technologies to incorporate into their products and use as a competitive advantage  in the market. Two modern and rapidly growing techniques are quantum computing and the use 

The HPC service at CERN provides linux batch infrastructure to run high performance computing  applications that require MPI clusters.The HPC cluster  is therefore dedicated to run MPI programs. 

This report describes EOS Winston, an event driven alerting and mitigation automation platform.  Through the use of expert rules and online anomaly detection algorithms, it catches events which 

Sepsis is a life-threatening condition where microbes present in the blood stream cause an  unregulated immune response from the body which can result in tissue damage, multi-organ failure 

This report describes performance analysis of the CernVM-FS FUSE which is a software distribution  service used in high-energy physics research. The performance analysis was conducted in both kernel 

The purpose of this report is to provide a brief overview of what OpenStack is, focusing on the  advantages of the integration of its Manila component at CERN. Furthermore, this document briefly 

On this project, we decided to implement two solutions that integrate REANA and GitLab. They vary  on two main points. The first one is the amount of configuration necessary to set up the integration, 

In 2005, the first installation of the Oracle HTML DB came out in production. Very soon the CERN developer community adopted the technology, using it in all the areas of the organization, from administrative applications to accelerators control system.

The IoT security framework is a computer security platform designed to assess the risks of various heterogeneous IoT devices. The framework is currently being developed at CERN and analyses different IoT devices connected to CERN’s General Purpose Network (GPN). The GPN mostly

Using detailed simulations of calorimeter showers as training data, we investigate the use of deep learning algorithms for the simulation and reconstruction of particles produced in high-energy physics collisions.

Hadoop ecosystem is distributed computing platform for Big Data solutions by comprising  autonomous components such as HDFS, Spark, YARN etc. HDFS is a Hadoop Distributed File  System for data storage. Current HDFS supports 3x replication for data fault-tolerance. When a 

This work has successfully deployed two different use cases of interest for High Energy Physics  using cloud resources:   CMS Big data reduction: This use case consists in running a data reduction workloads for 

T​he CERN Cloud Infrastructure team provides compute resources as a service to teams  across CERN. Users can provision resources to process experiment data, host web  applications, and accomplish other computing tasks. 

The main idea of this project is to carry out performance analysis on the RDataFrame class within the  ROOT operational framework. For this purpose, scalability analysis are performed on the execution 

The Elasticsearch Service is a distributed search and analytics engine widely used across CERN. Currently,  issues in the service are resolved manually after being detected through internal monitoring by service 

On this project we focused on benchmarking Influx against Oracle database. One of the  primary reason is ETM/Seimens were already working on Influx database backend. To  perform benchmarking using the Query Benchmark Tool we needed to have same data 

This report describes the work carried out to study and evaluate the performance and  footprint of different parquet compression codecs on data retrival and  analytics scenarios Parquet is a standard-de-facto and the data format used to persist 

When developing new programs or scientific libraries most of the efforts are focused on providing  efficient algorithms, the state-of-the-art techniques and maximum flexibility. However, in order for a 

The Database Services Group (IT-DB) is responsible for providing database and middleware services to  the laboratory. For these services, it is necessary to provide proper monitoring solutions to different user 

Neutrinos are subatomic particles, very similar to an electron, but without any electrical charge and  a very negligible rest mass. They are the most abundant and perhaps the most mysterious matter  particles in the universe!  

In run 4 of the LHC, the extreme high luminosity is expected to generate an enormous pileup of up to 200  proton-proton collisions for each bunch crossing. This has to be read out at 750 kHz with a maximum 

Big Data Technologies popularity continues to increase each year. The vast amount of data produced at the LHC experiments, which will increase further after the upgrade to HL-LHC, makes the exploration of new ways to perform physics

This project uses Wasserstain Generative Adversatial Networks (WGANs) to supply the demand for large simulation samples in the event of the CMS Phase II Upgrade. The distributions of real

At CERN we have many systems which provide critical services and scheduling downtime for them is quite difficult. Live kernel patching is a technique which aims to update the system without

The High Luminosity upgrade scheduled for 2026 will greatly increase the number of events per collision. Moore’s law will optimistically get a factor 4 performance gain, not enough to handle the

This reports summarises the project I worked on during my internship with the IT-DB-IMS team. This  report will detail my efforts to configure various technologies to work with Java Mission Control, the 

In recent times a tool for efficient unpacking of software work-flows from CernVM File System (CVMFS) into standalone images has become necessary. There are two types of use cases for such images: On the one hand they can be used to deliver

The interest on machine learning workloads in the HEP community has increased exponentially in the last years, making more and more important the need of a thorough benchmarking of the most relevant/significant workloads that are going to run on the experiments. The purpose

On this project, we chose to work with Clair, the tool developed by CoreOS, which uses static analysis to find vulnerabilities in container images. To use Clair, we had to build a Python client,

WinCC Open Architecture is a toolkit for creating Supervisory Control and Data  Acquisition (SCADA) applications, which is widely used at CERN. Hundreds of controls  applications, both in the accelerator complex and the experiments are based on it, 

The purpose of this report is to document the project I was working on for nine weeks during the summer of 2018. As part of the CERN openlab Summer Student Program 2018 I had the opportunity to work with the Digital Repositories (IT-CDA-DR) section at CERN on developing a

The interest in using containers to package applications is constantly growing in the software  development community, especially with new technologies such as Kubernetes, Open-shift  being adopted more frequently as well. This project also based on modularising the currently 

In June 2018, more than 68.5 Million people across the globe were reported to be fleeing war or persecution. Within the United Nations, UNOSAT is the organ in charge of collecting demo-

There are two very popular concepts that we hear in the world of technology​, Big  Data and Internet of Things. Big data is referring to a data which size, complexity and  velocity is really high and is difficult to capture, pre-process and analyze it with 

Computer simulations have become a very powerful tool for scientific research. In order to fa- cilitate research in computational biology, the BioDynaMo project aims at a general platform for

Simulations of beam dynamics vastly profit from parallelisation with high performance computing tech- niques. The two simulation libraries SixTrackLib and PyHEADTAIL are GPGPU accelerated. The former

In 2019 the LHCb experiment will go through an important upgrade, that will improve performance in many fields. One oh these fields is the DAQ system: it consists of a big flow of data that comes

One of the most time consuming algorithms that is currently employed for the reconstruction of High Energy Physics (HEP) workflows is the local energy reconstruction. The time spent to execute this algorithm constitutes 24% of the total processing time, thus achieving substantial

Improving the performance of an application is an important objective carried out from the application  conception until its deprecation. Developers are constantly trying to improve the performance of their 

The project described in this report is related to XRootD framework development. It was divided into two parts. First part was about publishing XRootD python bindings called PyXRootD to Python Package Index. This makes PyXRootD installation much easier and resolves problem

Puppet is a great tool for making changes on systems, and ensuring that those changes happen. But Puppet is not intended to make this happen on many systems at the same time. Puppet is intended for eventual compliance over time. Each agent checks in over a period of time, al-

In this report thin, single elements were compared between MAD-X and SixTrack. A testing framework for efficient comparisons between the two tracking codes was developed. A few dif- ferences between the tracking codes were found then documented and two bugs, one in the

CERN operates an OpenStack based private cloud to provide its users with resources on demand. It is one  of the largest OpenStack deployments in the world, with more than 300,000 cores over 9,000 hypervisors  [1].  

MPI Learn is a framework for the distributed training of neural networks. This platform is aimed at machine learning users, who can use it to train models faster, without dealing with the com-

Function as a service (FaaS) is a category of cloud computing services that  provides a platform allowing customers to develop, run, and manage application  functionalities without the complexity of building and maintaining the infrastructure 

The goal of this Openlab project is to create a Smart Data Analytics Platform for Science that will host analytical tools, publish data, share resources, interact with bots, collaborate and build communities of researchers with various backgrounds in a single ecosystem. With

We propose a novel neural network architecture called Hierarchical Latent Autoencoder to exploit the underlying hierarchical nature of the CMS Trigger System for data quality monitoring.

Some of the main challenges in scientific computing today deal with performance-preserving portability of software and reproducibility of the final results; likewise, with the advent of modern

Squid-Frontier system [1] is currently used to manage access to the COOL database [2].  This system includes many widely distributed computing sites and applications. Clients  presented by PanDA (Production ANd Distributed Analysis system, the ATLAS’ 

Malware Analysis Management (M.A.M.) or the automated sandbox analysis of  quarantined malware samples focuses on a detailed analysis of malware samples  reaching CERN through email traffic. M.A.M. is a side process of the main email pipeline 

REANA is a reusable analysis platform which offers physicists the ability to structure their research  data analysis and run their computational workflows in a containerized computing cloud. 

Computer simulations have become a very powerful tool for scientific research. In order to facilitate research in computational biology, the BioDynaMo project aims at a general platform for biological computer simulations, which should be executable on hybrid cloud computing systems.

In the competitive market for large-scale storage solutions the current main disk storage system at CERN EOS has been showing its excellence in the multi-Petabyte high-concurrency regime.

Exploring RapidIO RapidIO (http://rapidio.org/) technology is a packet-switched high-performance fabric, which has been under active development since 1997. The technology is used in all 4G/LTE basestations worldwide.

RapidIO (http://rapidio.org/) technology is a packet-switched high-performance fabric, which has been under active development since 1997. Originally meant to be a front side bus, it developed into a system level interconnect which is today used in all 4G/LTE base stations world wide.

Machine Learning techniques have been used in different applications by the HEP community: in this talk, we discuss the case of detector simulation.

Current research in the field of computational biology often involves simulations on high-performance computer clusters. It is crucial that the code of such simulations is efficient and correctly reflects the model specifications.

LHC Run3 and Run4 represent an unprecedented challenge for HEP computing in terms of both data volume and complexity. New approaches are needed for how data is collected and filtered, processed, moved, stored and analysed if these challenges are to be met with a realistic budget.

As computing power and I/O performance is increasing at an aggressive rate several RDMA enabled interconnect technologies have been entering the market, promising low latency and high throughput.

RapidIO technology is a packet-switched high-performance fabric, which has been under active development since 1997. The technology is used in all 4G/LTE base stations worldwide.