This paper outlines the technical and organizational measures implemented by the Italian supercomputing center, CINECA, to efficiently collect, process, and store sensitive-omics data in compliance with GDPR regulations. Indeed, the explosion of High Throughput Sequencing in medicine has raised tremendous opportunities for large-scale genomic data analysis. Cohort studies involving the processing of hundreds or thousands of input samples, combined with the integration of diverse diagnostic data, enable researchers to conduct integrative analyses at an unprecedented level of detail would have been impossible to achieve through single sample studies. To analyse such amount of data, centres that have access to High Performance Computing or extensive cloud resources have become crucial both for storage and efficient execution of data analysis pipelines. Nevertheless, since genomic data are considered sensitive personal data according to the EU General Data Protection Regulation, computational centres with high resource capabilities must prioritize data security and protection. This solution has been successfully applied to the Network for Italian Genomes use-case, demonstrating scalability to other hospitals and universities involved in research projects dealing with sensitive genomic data.

This entry is a part of a larger data set collected from the most recent Tier-0 supercomputer hosted at CINECA (Marconi100, https://www.hpc.cineca.it/hardware/marconi100). The data covers the entirety of the system, ranging from the computing nodes (980+ computing nodes) internal information such as core loads, temperatures, frequencies, memory write/read operations, CPU power consumption, fan speed, GPU usage details, etc., to the system-wide information, including the liquid cooling infrastructure, the air conditioning system, the power supply units, workload manager statistics, and job-related information, system status alerts, and weather forecast.It comprises hundreds of metrics measured on each computing node, in addition to hundreds of other metrics gathered from sensors monitored along all system components.The whole data set is stored as a collection of Zenodo entries; this particular entry corresponds to the period: 21-07, 21-09. The dataset is stored as a partitioned Parquet dataset, with this partitioning hierarchy: year_month ("YY-MM"), plugin, metric. The data is distributed as tarball files, each corresponding to one month of data (first-level partitioning, year_month).The collected data is generated by a monitoring infrastructure working on unstructured data (to improve efficiency and scalability); however, this data has been organized in a structured manner to facilitate its fruition. The simplest way to understand how the access the data is to refer to the companion software modules released together with the dataset itself, which can be found at: https://gitlab.com/ecs-lab/exadata.

This entry is a part of a larger data set collected from the most recent Tier-0 supercomputer hosted at CINECA (Marconi100, https://www.hpc.cineca.it/hardware/marconi100). The data covers the entirety of the system, ranging from the computing nodes (980+ computing nodes) internal information such as core loads, temperatures, frequencies, memory write/read operations, CPU power consumption, fan speed, GPU usage details, etc., to the system-wide information, including the liquid cooling infrastructure, the air conditioning system, the power supply units, workload manager statistics, and job-related information, system status alerts, and weather forecast.It comprises hundreds of metrics measured on each computing node, in addition to hundreds of other metrics gathered from sensors monitored along all system components. This particular dataset is made for anomaly detection purposes, it contains the same data as the main dataset but aggregated over time, with one Parquet file for each node. The data is distributed in tarballs, each one including all the files relative to the nodes contained in a given rack. For each file, the rows represent periods of 15 minutes, with the columns being aggregated values (average, standard deviation, min, max) over all the IPMI metrics that are available for the node; an additional column contains anomaly labels from Nagios. More details can be found in the companion repository: https://gitlab.com/ecs-lab/exadata, including the spatial distribution of the nodes in the room.

This entry is a part of a larger data set collected from the most recent Tier-0 supercomputer hosted at CINECA (Marconi100, https://www.hpc.cineca.it/hardware/marconi100). The data covers the entirety of the system, ranging from the computing nodes (980+ computing nodes) internal information such as core loads, temperatures, frequencies, memory write/read operations, CPU power consumption, fan speed, GPU usage details, etc., to the system-wide information, including the liquid cooling infrastructure, the air conditioning system, the power supply units, workload manager statistics, and job-related information, system status alerts, and weather forecast.It comprises hundreds of metrics measured on each computing node, in addition to hundreds of other metrics gathered from sensors monitored along all system components.The whole data set is stored as a collection of Zenodo entries; this particular entry corresponds to the period: 22-07. The dataset is stored as a partitioned Parquet dataset, with this partitioning hierarchy: year_month ("YY-MM"), plugin, metric. The data is distributed as tarball files, each corresponding to one month of data (first-level partitioning, year_month).The collected data is generated by a monitoring infrastructure working on unstructured data (to improve efficiency and scalability); however, this data has been organized in a structured manner to facilitate its fruition. The simplest way to understand how the access the data is to refer to the companion software modules released together with the dataset itself, which can be found at: https://gitlab.com/ecs-lab/exadata.

This entry is a part of a larger data set collected from the most recent Tier-0 supercomputer hosted at CINECA (Marconi100, https://www.hpc.cineca.it/hardware/marconi100). The data covers the entirety of the system, ranging from the computing nodes (980+ computing nodes) internal information such as core loads, temperatures, frequencies, memory write/read operations, CPU power consumption, fan speed, GPU usage details, etc., to the system-wide information, including the liquid cooling infrastructure, the air conditioning system, the power supply units, workload manager statistics, and job-related information, system status alerts, and weather forecast.It comprises hundreds of metrics measured on each computing node, in addition to hundreds of other metrics gathered from sensors monitored along all system components.The whole data set is stored as a collection of Zenodo entries; this particular entry corresponds to the period: 20-03, 20-12. The dataset is stored as a partitioned Parquet dataset, with this partitioning hierarchy: year_month ("YY-MM"), plugin, metric. The data is distributed as tarball files, each corresponding to one month of data (first-level partitioning, year_month).The collected data is generated by a monitoring infrastructure working on unstructured data (to improve efficiency and scalability); however, this data has been organized in a structured manner to facilitate its fruition. The simplest way to understand how the access the data is to refer to the companion software modules released together with the dataset itself, which can be found at: https://gitlab.com/ecs-lab/exadata.

This entry is a part of a larger data set collected from the most recent Tier-0 supercomputer hosted at CINECA (Marconi100, https://www.hpc.cineca.it/hardware/marconi100). The data covers the entirety of the system, ranging from the computing nodes (980+ computing nodes) internal information such as core loads, temperatures, frequencies, memory write/read operations, CPU power consumption, fan speed, GPU usage details, etc., to the system-wide information, including the liquid cooling infrastructure, the air conditioning system, the power supply units, workload manager statistics, and job-related information, system status alerts, and weather forecast.It comprises hundreds of metrics measured on each computing node, in addition to hundreds of other metrics gathered from sensors monitored along all system components.The whole data set is stored as a collection of Zenodo entries; this particular entry corresponds to the period: 22-05. The dataset is stored as a partitioned Parquet dataset, with this partitioning hierarchy: year_month ("YY-MM"), plugin, metric. The data is distributed as tarball files, each corresponding to one month of data (first-level partitioning, year_month).The collected data is generated by a monitoring infrastructure working on unstructured data (to improve efficiency and scalability); however, this data has been organized in a structured manner to facilitate its fruition. The simplest way to understand how the access the data is to refer to the companion software modules released together with the dataset itself, which can be found at: https://gitlab.com/ecs-lab/exadata.

Supercomputers are the most powerful computing machines available to society. They play a central role in economic, industrial, and societal development. While they are used by scientists, engineers, decision-makers, and data-analyst to computationally solve complex problems, supercomputers and their hosting datacenters are themselves complex power-hungry systems. Improving their efficiency, availability, and resiliency is vital and the subject of many research and engineering efforts. Still, a major roadblock hinders researchers: dearth of reliable data describing the behavior of production supercomputers. In this paper, we present the result of a ten-year-long project to design a monitoring framework (EXAMON) deployed at the Italian supercomputers at CINECA datacenter. We disclose the first holistic dataset of a tier-0 Top10 supercomputer. It includes the management, workload, facility, and infrastructure data of the Marconi100 supercomputer for two and half years of operation. The dataset (published via Zenodo) is the largest ever made public, with a size of 49.9TB before compression. We also provide open-source software modules to simplify access to the data and provide direct usage examples.

The Human Brain Project (HBP) ( https://humanbrainproject.eu/ ) is a large-scale flagship project funded by the European Commission with the goal of establishing a research infrastructure for brain science. This research infrastructure is currently being realised and will be called EBRAINS ( https://ebrains.eu/ ). The wide ranging EBRAINS services for the brain research communities require diverse access, processing and storage capabilities. As a result, it will strongly rely on e-infrastructure services. The HBP led to the creation of Fenix ( https://fenix-ri.eu/ ), a collaboration of five European supercomputing centres, who are providing a set of federated e-infrastructure services to EBRAINS. The Fenix architecture has been designed to uniquely address the need for a wide spectrum of services, from high performance computing (HPC) to on-demand cloud technologies to identity and access federation, for facilitating ease of access and usage of distributed e-infrastructure resources. In this article we describe the underlying concepts for an audience of computational science end-users and developers of domain-specific applications, workflows and platforms services. To exemplify the use of Fenix, we will discuss selected use cases demonstrating how brain researchers can use the offered infrastructure services and describe how access to these resources can be obtained.

Predicting the outcome of jet-milling based on the knowledge of process parameters and starting material properties is a task still far from being accomplished. Given the technical difficulties in measuring thermodynamics, flow properties and particle statistics directly in the mills, modelling and simulations constitute alternative tools to gain insight in the process physics and many papers have been recently published on the subject. An ideal predictive simulation tool should combine the correct description of non-isothermal, compressible, high Mach number fluid flow, the correct particle-fluid and particle-particle interactions and the correct fracture mechanics of particle upon collisions but it is not currently available. In this paper we present our coupled CFD-DEM simulation results; while comparing them with the recent modelling and experimental works we will review the current understating of the jet-mill physics and particle classification. Subsequently we analyze the missing elements and the bottlenecks currently limiting the simulation technique as well as the possible ways to circumvent them towards a quantitative, predictive simulation of jet-milling.

Background The advent of Next Generation Sequencing (NGS) technologies and the concomitant reduction in sequencing costs allows unprecedented high throughput profiling of biological systems in a cost-efficient manner. Modern biological experiments are increasingly becoming both data and computationally intensive and the wealth of publicly available biological data is introducing bioinformatics into the “Big Data” era. For these reasons, the effective application of High Performance Computing (HPC) architectures is becoming progressively more recognized also by bioinformaticians. Here we describe HPC resources provisioning pilot programs dedicated to bioinformaticians, run by the Italian Node of ELIXIR (ELIXIR-IT) in collaboration with CINECA, the main Italian supercomputing center. Results Starting from April 2016, CINECA and ELIXIR-IT launched the pilot Call “ELIXIR-IT HPC@CINECA”, offering streamlined access to HPC resources for bioinformatics. Resources are made available either through web front-ends to dedicated workflows developed at CINECA or by providing direct access to the High Performance Computing systems through a standard command-line interface tailored for bioinformatics data analysis. This allows to offer to the biomedical research community a production scale environment, continuously updated with the latest available versions of publicly available reference datasets and bioinformatic tools. Currently, 63 research projects have gained access to the HPC@CINECA program, for a total handout of ~ 8 Millions of CPU/hours and, for data storage, ~ 100 TB of permanent and ~ 300 TB of temporary space. Conclusions Three years after the beginning of the ELIXIR-IT HPC@CINECA program, we can appreciate its impact over the Italian bioinformatics community and draw some considerations. Several Italian researchers who applied to the program have gained access to one of the top-ranking public scientific supercomputing facilities in Europe. Those investigators had the opportunity to sensibly reduce computational turnaround times in their research projects and to process massive amounts of data, pursuing research approaches that would have been otherwise difficult or impossible to undertake. Moreover, by taking advantage of the wealth of documentation and training material provided by CINECA, participants had the opportunity to improve their skills in the usage of HPC systems and be better positioned to apply to similar EU programs of greater scale, such as PRACE. To illustrate the effective usage and impact of the resources awarded by the program - in different research applications - we report five successful use cases, which have already published their findings in peer-reviewed journals.

Five European supercomputing centres, namely BSC (Spain), CEA (France), CINECA (Italy), CSCS (Switzerland), and JSC (Germany), agreed to align their high-end computing and storage services to facilitate the creation of the Fenix Research Infrastructure. In addition to the traditional extreme-scale computing and data services, Fenix provides a set of Cloud-type services as well as services needed for federation. In this paper, we describe the architecture of the Fenix infrastructure and how it can be used for representative workflows from the Human Brain Project (HBP). The concept of the Active Data Repository (ACD) is chosen to highlight demarcation between HPC and Cloud access models.

Using a molecular field theory with atomistic modelling, we provide a complete description of the elastic and flexoelectric properties of the nematic phase formed by liquid crystal dimers which, depending on the parity of the number of atoms in the spacer, have either a bent (odd) or a straight (even) average shape. We can then estimate the flexoelastic ratio and make a direct comparison with the outcome of flexoelectro-optic measurements. Our results demonstrate the extreme sensitivity of the bend elasticity and flexoelectricity to the molecular structure, with dramatic differences between even and odd dimers. An unusually low bend elastic constant is predicted for the latter; we discuss the implications of this result for the high flexoelastic response and the existence of Blue Phases stable over a wide temperature range, which were both recently claimed for odd liquid crystal dimers.

The liquid-crystal dimer 1'',7''-bis(4-cyanobiphenyl-4'-yl)heptane (CB7CB) exhibits two liquid-crystalline mesophases on cooling from the isotropic phase. The high-temperature phase is nematic; the identification and characterization of the other liquid-crystal phase is reported in this paper. It is concluded that the low-temperature mesophase of CB7CB is a new type of uniaxial nematic phase having a nonuniform director distribution composed of twist-bend deformations. The techniques of small-angle x-ray scattering, modulated differential scanning calorimetry, and dielectric spectroscopy have been applied to establish the nature of the nematic-nematic phase transition and the structural features of the twist-bend nematic phase. In addition, magnetic resonance studies (electron-spin resonance and (2)H nuclear magnetic resonance) have been used to investigate the orientational order and director distribution in the liquid-crystalline phases of CB7CB. The synthesis of a specifically deuterated sample of CB7CB is reported, and measurements showed a bifurcation of the quadrupolar splitting on entering the low-temperature mesophase from the high-temperature nematic phase. This splitting could be interpreted in terms of the chirality of the twist-bend structure of the director. Calculations using an atomistic model and the surface interaction potential with Monte Carlo sampling have been carried out to determine the conformational distribution and predict dielectric and elastic properties in the nematic phase. The former are in agreement with experimental measurements, while the latter are consistent with the formation of a twist-bend nematic phase.

The elastic moduli of low-molar-mass thermotropic liquid crystals (LCs) exhibit an intriguing dependence on the molecular structure of the constituents, which can be very important for applications. We have recently developed a molecular field theory, wherein the elastic constants of nematics are expressed in terms of integrals over the molecular surface. This theory, combined with molecular geometry optimization, allows us to connect mesoscale deformations in liquid crystals to atomic-scale details. Here we investigate typical mesogenic systems, i.e.para-azoxyanisole (PAA) and 4-n-alkyl,-4′-cyanobiphenyls (nCBs), whose elastic properties exhibit clear differences. We show that these can be traced back to differences in molecular shape. Our calculations also highlight the importance of the flexibility of mesogens, which was generally ignored by previous theories: in view of their different shape, conformers are shown to give different contributions to the elastic constants. The key role of deviations from a rod-like shape, which is generally assumed by models of mesogens, emerges from our calculations. The bend elastic constant is shown to be particularly sensitive to this feature; for a given compound, rod-like conformers give a high contribution to the bending stiffness, whereas the contribution of bent conformers is low or even negative. The possible implications of these findings are discussed, with special reference to the behavior of bent-core mesogens. Finally, we predict the temperature dependence of the surface-like elastic constants, whose experimental determination is still controversial; we find that these are generally smaller than the bulk moduli and even more sensitive to changes in the molecular shape.

Liquid crystals oppose a restoring force to distortions of the main alignment axis, the so-called director. For nematics this behavior is characterized by the three elastic moduli associated with the splay (K(11)), twist (K(22)), and bend (K(33)) modes; in addition, two moduli for mixed splay-bend (k(13)) and saddle-splay (k(24)) can be defined. The elastic constants are material properties which depend on the mesogen structure, but the relation between molecular features and deformations on a much longer scale has not been fully elucidated. The prediction of elastic properties is a challenge for theoretical and computational methods: atomistic simulations require large samples and must be integrated by statistical thermodynamics models to connect intermolecular correlations and elastic response. Here we present a molecular field theory, wherein expressions for the elastic constants of nematics are derived starting from a simple form of the single molecule orientational distribution function; this is parametrized according to the amount of molecular surface aligned to the nematic director. Such a model allows a detailed account of the chemical structure; moreover the conformational freedom, which is a common feature of mesogens, can be easily included. Given the atomic coordinates, the elastic constants can be calculated without any adjustable parameter at a low computational cost. The example of 4-n-pentyl,4(')-cyanobiphenyl (5CB) is used to illustrate the capability of the developed methodology; even for this mesogen, which is usually taken as a prototypal rodlike system, we predict a significant dependence of the elastic moduli on the molecular conformation. We show that good estimates of magnitude and temperature dependence of the elastic constants are obtained, provided that the molecular geometry is correctly taken into account.

This paper addresses the study of the electronic spectrum of the s-tetrazine molecule in the ab initio frame making use of the multireference n-electron Valence State Perturbation Theory (NEVPT2). The theoretical description of the excited states of this molecule is complex, because of the different computational requirements of the low-lying excited states which must be treated on an equal footing. More than forty electronic excited states of various nature (n → π*, nn → π*π*, π → π*, nπ → π*π*, and Rydberg) are considered here. Various active spaces are used to reach a good quality zero order description, needed to avoid unrealistic results in the perturbation treatment. The quasi-degenerate perturbation theory must be used in most of the cases, given the presence of a marked mixing among various states. The results here presented have been used to add new information to the interpretation of the experimental spectra. While in many cases previous assignments of the experimental features are confirmed, in various cases they are questioned and new assignments are proposed. With the new assignments, a good agreement is found between experiments and NEVPT2. The comparison with other high level ab initio methods shows that NEVPT2 performs well, being in general in close agreement with Extended-STEOM-CCSD and CC3, while the agreement with CASPT2 and GVVPT2 is less satisfactory (the CASPT2 excitation energies being lower than the NEVPT2 ones by ≃0.4 eV). Finally, comparing NEVPT2 with TD-DFT, a reasonable accordance with the values obtained with the PBE0 functional is observed, while the agreement with those computed with the HCTH functional is lower.

This thesis focuses on the development and the application of a computational methodology, based on a molecular field theory and atomistic modelling, to connect dielectric and elastic properties of nematic liquid crystals to the structure of the constituent molecules. Chapter 1 is a general introduction on the subject of the thesis. Firstly, the problem of the connection between materials properties and structure of the molecular constituents is introduced, with special reference to the case of liquid crystals, and the object of this work is presented. The main features of liquid crystals are then recalled, considering in particular the elastic and dielectric properties, investigated in this thesis, which are directly involved in the electro-optical behaviour. We also show the molecular systems to which the theoretical- computational methodology developed here has been applied. These have the common structure of two mesogenic, rather rigid units, connected by a flexible spacer. For these reasons they are called ‘dimers’. These mesogens have several reasons of interest: their liquid crystal properties are very sensitive to changes in the molecular structure and exhibit some unusual and unexplained features. Therefore they can been devised as a benchmark for molecular modelling of liquid crystals. In chapter 2 the theoretical framework is presented. After a review of the state of the art of the computational methods for the study of liquid crystals, we present the molecular field approach used in this thesis, which is based on the ‘Surface Interaction’ (SI) model. Herein, the relation between molecular and mesoscale level is introduced through the assumption that each element of the molecular surface tends to align to the nematic director. A realistic account of the molecular structure is made possible by the use of a surface generated from atomic coordinates. We report the molecular expressions obtained in this framework for the ordering, thermodynamic, flexoelectric and dielectric properties of nematic liquid crystals. Given the role played by the molecular flexibility, special attention is devoted to the conformational degrees of freedom. Two different ways are proposed for its inclusion in the model: the Rotational Isomeric State (RIS) approximation, in which only the molecular geometries corresponding to the minima of the torsional potential are considered, or the Monte Carlo (MC) sampling of torsional angles. In chapter 3 we derive molecular expressions for the bulk and surfacelike elastic constants of nematics, within the framework of the SI model. This requires extensive use of tensor calculus; after some lengthy algebra, simple expressions are obtained, by exploiting the symmetry of the undeformed nematic phase. From the point of view of the theoret- ical development, this is the main result of the present thesis. The elastic constants can be calculated as a function of the orientational order, without any free parameter, at low computational cost. It enables us to investigate the role of molecular features and to explore how changes at the atomic level can be conveyed into changes in elastic behaviour, on a quite different length-scale. Therefore it can shed light on the origin, still poorly understood, of the different elasticity of mesogens with different structure. The predictive ability of this method makes it potentially useful for the synthetic design of tailored mesogens: the elastic constants can be easily calculated, if the molecular structure is known. We also derive molecular expressions for the surfacelike elastic constants of nematics. The surface elasticity of nematics has been a subject of intense theoretical and experimental investigation and no consensus has been reached; our analysis can be seen as a preliminary exploration of this problem, which deserves further investigation in the future, and we hope that our atomistic level approach can provide some new insight. In chapter 5 the elastic behaviour of three typical liquid crystals mesogens (PAA, 5CB, 8CB) is investigated, using the molecular field theory presented in Chapter 3. These have been chosen as representative cases because of their different elasticity, despite the structural similarity. The availability of experimental data allows us to assess the quality of the theoretical predictions. We show that the observed temperature dependence of the splay, twist and bend elastic moduli can be traced back to differences, even not dramatic, in molecular shape. Our calculations also highlight the importance of the flexibility of mesogens, which was generally ignored by previous theories: in view of their different shape, conformers are shown to give different contributions to the elastic moduli. The key role of deviations from a rod-like shape, which is generally assumed by models of mesogens, emerges from the calculations. The bend elastic constant is shown to be particularly sensitive to molecular bending; it can range from high values for rod-like conformers, to low and even negative values for bent conformers of a given compound. These findings could have important implications for bent-core mesogens, which are presently the object of intense investigation because of their unusual and attractive properties. We also report the surfacelike elastic constants of PAA, 5CB, 8CB, whose experimental determination is controversial; we have found that these are generally smaller than the bulk elastic moduli and even more sensitive to changes in molecular shape. The results obtained for the LC dimers, taking into account the conformational freedom at the RIS level, are reported in chapter 6. A full overview is provided, comprising order parameters, properties at the nematic-isotropic transition, dielectric permittivity, elastic and flexoelastic moduli. The molecular model enables us to reach an unprecedented insight into the origin of not yet explained experimental findings, and to predict behaviours not yet probed by experiment. Particularly interesting are the results obtained for the flexoelectric and elastic properties of the LC dimers. The common view, which gives electric and steric dipoles the main responsibility for the flexoelectric properties, cannot explain recent experimental findings for LC dimers; our results single out the importance of taking into account the whole distribution of charges and the real molecular shape. Experimental data are available for the splay elastic constants of dimers [Tsvetkov et al, Mol. Cryst. Liq. Cryst.: 331:1901, 1999]: we correctly predict not only magnitude of the elastic constants, but also their dependence on the length of the flexible spacer. No comparison with experiment is possible for the twist and bend elastic moduli; however our results appear very promising, in relation to some intriguing phenomena which have been recently reported for LC dimers [Coles et al, Nature, 436:997, 2005] and bent-core LCs [G¨rtz et al, Soft Matter, 5:463, 2009]. In chapter 7 we investigate whether the small amplitude fluctuations around the minima of the torsional potential, which are neglected by the RIS approximation, can affect the elastic and dielectric properties of LC dimers. To this purpose, we have performed calculations with MC sampling of the torsional angles. We show that small amplitude fluctuations do play a role for those properties which are particularly sensitive to the balance between elongated and bent conformations; these comprise the bend elasticity and flexoelectricity. Significant, though less subtle effects of torsional oscillations are also found for the dielectric permittivity, when some of the torsional angles are characterised by relatively low barriers between the minima. In this final chapter, collecting all the results obtained for LC dimers, we are able to provide a complete explanation for the experiments performed by Coles and colleagues [Coles et al, J. Mater. Chem.,11:2709, 2001; Morris et al, Phys. Rev. E, 75:041701,2007], which simultaneously involve elastic and flexoelectric properties.