arXiv subjects

Francesco Pace

Publications and source records attributed to Francesco Pace.

At least 55 records · Page 3Linked to original sources

arXiv2017

Experimental Performance Evaluation of Cloud-Based Analytics-as-a-Service

An increasing number of Analytics-as-a-Service solutions has recently seen the light, in the landscape of cloud-based services. These services allow flexible composition of compute and storage components, that create powerful data ingestion and processing pipelines. This work is a first attempt at an experimental evaluation of analytic application performance executed using a wide range of storage service configurations. We present an intuitive notion of data locality, that we use as a proxy to rank different service compositions in terms of expected performance. Through an empirical analysis, we dissect the performance achieved by analytic workloads and unveil problems due to the impedance mismatch that arise in some configurations. Our work paves the way to a better understanding of modern cloud-based analytic services and their performance, both for its end-users and their providers.

cs.DC↗

arXiv2016

Effects of tidal gravitational fields in clustering dark energy models

We extend a previous work by Reischke et al., 2016 by studying the effects of tidal shear on clustering dark energy models within the framework of the extended spherical collapse model and using the Zel'dovich approximation. As in previous works on clustering dark energy, we assumed a vanishing effective sound speed describing the perturbations in dark energy models. To be self-consistent, our treatment is valid only on linear scales since we do not intend to introduce any heuristic models. This approach makes the linear overdensity $δ_{\rm c}$ mass dependent and similarly to the case of smooth dark energy, its effects are predominant at small masses and redshifts. Tidal shear has effects of the order of percent or less, regardless of the model and preserves a well known feature of clustering dark energy: When dark energy perturbations are included, the models resemble better the $Λ$CDM evolution of perturbations. We also showed that effects on the comoving number density of halos are small and qualitatively and quantitatively in agreement with what previously found for smooth dark energy models.

astro-ph.CO↗

arXiv2016

Flexible Scheduling of Distributed Analytic Applications

This work addresses the problem of scheduling user-defined analytic applications, which we define as high-level compositions of frameworks, their components, and the logic necessary to carry out work. The key idea in our application definition, is to distinguish classes of components, including rigid and elastic types: the first being required for an application to make progress, the latter contributing to reduced execution times. We show that the problem of scheduling such applications poses new challenges, which existing approaches address inefficiently. Thus, we present the design and evaluation of a novel, flexible heuristic to schedule analytic applications, that aims at high system responsiveness, by allocating resources efficiently. Our algorithm is evaluated using trace-driven simulations, with large-scale real system traces: our flexible scheduler outperforms a baseline approach across a variety of metrics, including application turnaround times, and resource allocation efficiency. We also present the design and evaluation of a full-fledged system, which we have called Zoe, that incorporates the ideas presented in this paper, and report concrete improvements in terms of efficiency and performance, with respect to prior generations of our system.

cs.DC↗

arXiv2016

Constraints on shear and rotation with massive galaxy clusters

A precise determination of the mass function is an important tool to verify cosmological predictions of the $Λ$CDM model and to infer more precisely the better model describing the evolution of the Universe. Galaxy clusters have been currently used to infer cosmological parameters, in particular the matter density parameter $Ω_{\rm m}$, the matter power spectrum normalization $σ_8$ and the equation of state parameter $w_{\rm de}$ of the dark energy fluid. In this work, using data on massive galaxy clusters ($M>8\times 10^{14}~h^{-1}~M_{\odot}$) in the redshift range $0.05\lesssim z\lesssim 0.83$ we put constraints on the parameter $α$ introduced within the formalism of the extended spherical collapse model to quantify deviations from sphericity due to shear and rotation. Since at the moment there is no physical model describing its functional shape, we assume it to be a logarithmic function of the cluster mass. By holding $σ_8$ fixed and restricting our analysis to a $Λ$CDM model, we find, at $1-σ$ confidence level, $Ω_{\rm m}=0.284\pm0.0064$, $h=0.678\pm0.017$ and $β=0.0019^{+0.0008}_{-0.0015}$, where $β$ represents the slope of the parameter $α$. This results translates into a $9\%$ decrement of the number of massive clusters with respect to a standard $Λ$CDM mass function, but better data are required to better constrain this quantity, since at the $2-σ$ and $3-σ$ confidence level we are only able to infer upper limits.

astro-ph.CO↗

arXiv2016

Approximation of the potential in scalar field dark energy models

We study the nature of potentials in scalar field based models for dark energy - with both canonical and noncanonical kinetic terms. We calculate numerically, and using an analytic approximation around $a\approx 1$, potentials for models with constant equation-of-state parameter, $w_ϕ$. We find that for a wide range of models with canonical and noncanonical kinetic terms there is a simple approximation for the potential that holds when the scale factor is in the range $0.6\lesssim a\lesssim 1.4$. We discuss how this form of the potential can also be used to represent models with non-constant $w_ϕ$ and, hence, how it could be used in reconstruction from cosmological data.

astro-ph.CO↗

arXiv2016

Spherical collapse of dark matter haloes in tidal gravitational fields

We study the spherical collapse model in the presence of external gravitational tidal shear fields for different dark energy scenarios and investigate the impact on the mass function and cluster number counts. While previous studies of the influence of shear and rotation on $δ_\mathrm{c}$ have been performed with heuristically motivated models, we try to avoid this model dependence and sample the external tidal shear values directly from the statistics of the underlying linearly evolved density field based on first order Lagrangian perturbation theory. Within this self-consistent approach, in the sense that we restrict our treatment to scales where linear theory is still applicable, only fluctuations larger than the scale of the considered objects are included into the sampling process which naturally introduces a mass dependence of $δ_\mathrm{c}$. We find that shear effects are predominant for smaller objects and at lower redshifts, i. e. the effect on $δ_\mathrm{c}$ is at or below the percent level for the $Λ$CDM model. For dark energy models we also find small but noticeable differences, similar to $Λ$CDM. The virial overdensity $Δ_\mathrm{V}$ is nearly unaffected by the external shear. The now mass dependent $δ_c$ is used to evaluate the mass function for different dark energy scenarios and afterwards to predict cluster number counts, which indicate that ignoring the shear contribution can lead to biases of the order of $1σ$ in the estimation of cosmological parameters like $Ω_\mathrm{m}$, $σ_8$ or $w$.

astro-ph.CO↗

arXiv2015

Constraints on $Ω_\mathrm{m}$ and $σ_8$ from the potential-based cluster temperature function

The abundance of galaxy clusters is in principle a powerful tool to constrain cosmological parameters, especially $Ω_\mathrm{m}$ and $σ_8$, due to the exponential dependence in the high-mass regime. While the best observables are the X-ray temperature and luminosity, the abundance of galaxy clusters, however, is conventionally predicted as a function of mass. Hence, the intrinsic scatter and the uncertainties in the scaling relations between mass and either temperature or luminosity lower the reliability of galaxy clusters to constrain cosmological parameters. In this article, we further refine the X-ray temperature function for galaxy clusters by Angrick et al., which is based on the statistics of perturbations in the cosmic gravitational potential and proposed to replace the classical mass-based temperature function, by including a refined analytic merger model and compare the theoretical prediction to results from a cosmological hydrodynamical simulation. Although we find already a good agreement if we compare with a cluster temperature function based on the mass-weighted temperature, including a redshift-dependent scaling between mass-based and spectroscopic temperature yields even better agreement between theoretical model and numerical results. As a proof of concept, incorporating this additional scaling in our model, we constrain the cosmological parameters $Ω_\mathrm{m}$ and $σ_8$ from an X-ray sample of galaxy clusters and tentatively find agreement with the recent cosmic microwave background based results from the Planck mission at 1$σ$-level.

astro-ph.CO↗

arXiv2015

Effects of ghost dark energy perturbations on the evolution of spherical overdensities

While in the standard cosmological model the accelerated expansion of the Universe is explained by invoking the presence of the cosmological constant term, it is still unclear the true origin of this stunning observational fact. It is therefore interesting to explore alternatives to the simplest scenario, in particular by assuming a more general framework where the fluid responsible of the accelerated expansion is characterised by a time-dependant equation of state. Usually these models, dubbed dark energy models, are purely phenomenological, but in this work we concentrate on a theoretically justified model, the ghost dark energy model. Within the framework of the spherical collapse model, we evaluate effects of dark energy perturbations both at the linear and non-linear level and transfer these results into an observable quantity, the mass function, by speculatively taking into account contributions of dark energy to the mass of the halos. We showed that the growth rate is higher in ghost models and that perturbations enhance the number of structures with respect to the $Λ$CDM model, with stronger effects when the total mass takes into account dark energy clumps.

gr-qc↗

arXiv2015

The importance of the cosmic web and halo substructure for power spectra

In this work we study the relevance of the cosmic web and substructures on the matter and lensing power spectra measured from halo mock catalogues extracted from the N-body simulations. Since N-body simulations are computationally expensive, it is common to use faster methods that approximate the dark matter field as a set of halos. In this approximation, we replace mass concentrations in N-body simulations by a spherically symmetric Navarro-Frenk-White halo density profile. We also consider the full mass field as the sum of two distinct fields: dark matter halos ($M>9\times 10^{12}~M_{\odot}$/h) and particles not included into halos. Mock halos reproduce well the matter power spectrum, but underestimate the lensing power spectrum on large and small scales. For sources at $z_{\rm s}=1$ the lensing power spectrum is underestimated by up to 40% at $\ell\approx 10^4$ with respect to the simulated halos. The large scale effect can be alleviated by combining the mock catalogue with the dark matter distribution outside the halos. In addition, to evaluate the contribution of substructures we have smeared out the intra-halo substructures in a N-body simulation while keeping the halo density profiles unchanged. For the matter power spectrum the effect of this smoothing is only of the order of 5%, but for lensing substructures are much more important: for $\ell\approx 10^4$ the internal structures contribute 30% of the total spectrum. These findings have important implications in the way mock catalogues have to be created, suggesting that some approximate methods currently used for galaxy surveys will be inadequate for future weak lensing surveys.

astro-ph.CO↗

arXiv2015

Intrinsic size correlations in weak lensing

We present a simple model for describing intrinsic correlations for galaxy sizes based on the halo model. Studying these correlations is important both to improve our understanding of galaxy properties and because it is an important potential systematic for weak lensing size magnification measurements. Our model assumes that the density field drives these intrinsic correlations and we also model the distribution of satellite galaxies. We calculate the possible contamination to measurements of lensing convergence power spectrum from galaxy sizes, and show that the cross-correlation of intrinsic sizes with convergence is potentially an important systematic. We also explore how these intrinsic size correlations may affect surveys with different redshift depth. We find that, in this simple approach, intrinsic size correlations cannot be neglected in order to estimate lensing convergence power spectrum for constraining cosmological parameters.

astro-ph.CO↗

arXiv2015

Evolution of spherical overdensities in holographic dark energy models

In this work we investigate the spherical collapse model in flat FRW dark energy universes. We consider the Holographic Dark Energy (HDE) model as a dynamical dark energy scenario with a slowly time-varying equation-of-state (EoS) parameter $w_{\rm de}$ in order to evaluate the effects of the dark energy component on structure formation in the universe. We first calculate the evolution of density perturbations in the linear regime for both phantom and quintessence behavior of the HDE model and compare the results with standard Einstein-de Sitter (EdS) and $Λ$CDM models. We then calculate the evolution of two characterizing parameters in the spherical collapse model, i.e., the linear density threshold $δ_{\rm c}$ and the virial overdensity parameter $Δ_{\rm vir}$. We show that in HDE cosmologies the growth factor $g(a)$ and the linear overdensity parameter $δ_{\rm c}$ fall behind the values for a $Λ$CDM universe while the virial overdensity $Δ_{\rm vir}$ is larger in HDE models than in the $Λ$CDM model. We also show that the ratio between the radius of the spherical perturbations at the virialization and turn-around time is smaller in HDE cosmologies than that predicted in a $Λ$CDM universe. Hence the growth of structures starts earlier in HDE models than in $Λ$CDM cosmologies and more concentrated objects can form in this case. It has been shown that the non-vanishing surface pressure leads to smaller virial radius and larger virial overdensity $Δ_{\rm vir}$. We compare the predicted number of halos in HDE cosmologies and find out that in general this value is smaller than for $Λ$CDM models at higher redshifts and we compare different mass function prescriptions. Finally, we compare the results of the HDE models with observations.

gr-qc↗

arXiv2014

Raytracing simulations of coupled dark energy models

Dark matter and dark energy are usually assumed to couple only gravitationally. An extension to this picture is to model dark energy as a scalar field coupled directly to cold dark matter. This coupling leads to new physical effects, such as a fifth-force and a time-dependent dark matter particle mass. In this work we examine the impact that coupling has on weak lensing statistics by constructing realistic simulated weak-lensing maps using raytracing techniques through N-body cosmological simulations. We construct maps for different lensing quantities, covering a range of scales from a few arcminutes to several degrees. The concordance $Λ$CDM model is compared to different coupled dark energy models, described either by an exponential scalar field potential (standard coupled dark energy scenario) or by a SUGRA potential (bouncing model). We analyse several statistical quantities and our results, with sources at low redshifts are largely consistent with previous work on CMB lensing by Carbone et al., 2013. The most significant differences from the $Λ$CDM model are due to the enhanced growth of the perturbations and to the effective friction term in non-linear dynamics. For the most extreme models, we see differences in the power spectra up to 40% compared to the $Λ$CDM model. The different time evolution of the linear matter overdensity can account for most of the differences, but when controlling for this using a $Λ$CDM model having the same normalization, the overall signal is smaller due to the effect of the friction term appearing in the equation of motion for dark matter particles.

astro-ph.CO↗

arXiv2014

Can observational growth rate data favour the clustering dark energy models?

Under the commonly used assumption that clumped objects can be well described by a spherical top-hat matter density profile, we investigate the evolution of the cosmic growth index in clustering dark energy (CDE) scenarios on sub-horizon scales. We show that the evolution of the growth index $γ(z)$ strongly depends on the equation-of-state (EoS) parameter and on the clustering properties of the dark energy (DE) component. Performing a $χ^2$ analysis, we show that CDE models have a better fit to observational growth rate data points with respect to the concordance $Λ$CDM model. We finally determine $γ(z)$ using an exponential parametrization and demonstrate that the growth index in CDE models presents large variations with cosmic redshift. In particular it is smaller (larger) than the theoretical value for the $Λ$CDM model, $γ_Λ\simeq0.55$, in the recent past (at the present time).

astro-ph.CO↗

arXiv2013

A comparison of structure formation in minimally and non-minimally coupled quintessence models

We study structure formation in non-minimally coupled dark energy models, where there is a coupling in the Lagrangian between a quintessence scalar field and gravity via the Ricci scalar. We consider models with a range of different non-minimal coupling strengths and compare these to minimally coupled quintessence models with time-dependent dark energy densities. The equations of state of the latter are tuned to either reproduce the equation of state of the non-minimally coupled models or their background history. Thereby they provide a reference to study the unique imprints of coupling on structure formation. We show that the coupling between gravity and the scalar field, which effectively results in a time-varying gravitational constant G, is not negligible and its effect can be distinguished from a minimally coupled model. We extend previous work on this subject by showing that major differences appear in the determination of the mass function at high masses, where we observe differences of the order of 40% at z=0. Our new results concern effects on the non-linear matter power spectrum and on the lensing signal (differences of ~10% for both quantities), where we find that non-minimally coupled models could be distinguished from minimally coupled ones.

astro-ph.CO↗

arXiv2013

Hydrodynamical chemistry simulations of the SZ effect and the impacts from primordial non-Gaussianities

The impacts of Compton scattering of hot cosmic gas with the cosmic microwave background radiation (Sunyaev-Zel'dovich effect, SZ) are consistently quantified in Gaussian and non-Gaussian scenarios, by means of 3D numerical, N-body, hydrodynamic simulations, including cooling, star formation, stellar evolution and metal pollution (He, C, O, Si, Fe, S, Mg, etc.) from different stellar phases, according to proper yields for individual metal species and mass-dependent stellar lifetimes. Light cones are built through the simulation outputs and samples of one hundred maps for the resulting temperature fluctuations are derived for both Gaussian and non-Gaussian primordial perturbations. From them, we estimate the possible changes due to early non-Gaussianities on: SZ maps, probability distribution functions, angular power spectra and corresponding bispectra. We find that the different growth of structures in the different cases induces significant spectral distortions only in models with large non-Gaussian parameters, $f_{\rm NL}$. In general, the overall trends are covered by the non-linear, baryonic evolution, whose feedback mechanisms tend to randomize the gas behaviour and homogenize its statistical features, quite independently from the background matter distribution. Deviations due to non-Gaussianity are almost undistinguishable for $f_{\rm NL}\lesssim 100$, remaining always at few-per-cent level, within the error bars of the Gaussian scenario. Rather extreme models with $f_{\rm NL}\sim1000$ present more substantial deviations from the Gaussian case, overcoming baryon contaminations and showing discrepancies up to a factor of a few in the spectral properties.

astro-ph.CO↗

arXiv2013

Bounded Scalar Perturbations in Bouncing Brane World Cosmologies

We examine the dynamics of scalar perturbations in closed Friedmann-Lemaître-Robertson- Walker (FLRW) universes in the framework of Brane World theory with a timelike extra dimension. In this scenario, the unperturbed Friedmann equations contain additional terms arising from the bulk-brane interaction that implement non-singular bounces in the models with a cosmological constant and non-interacting perfect fluids. The structure of the phase-space of the models allows for two basic configurations, namely, one bounce solutions or eternal universes. Assuming that the matter content of the model is given by dust and radiation, we derive the dynamical field equations for scalar hydrodynamical perturbations considering either a conformally flat (de Sitter) bulk or a perturbed bulk. We perform a numerical analysis which can shed some light on the study of cosmological scalar perturbations in bouncing brane world models. From a mathematical point of view we show that although the bounce enhances the amplitudes of scalar perturbations for one bounce models in the case of a de Sitter bulk, the amplitudes of the perturbations remain sufficiently small and bounded relative to the background values up to a certain scale. For one bounce models in the case of a perturbed bulk the amplitudes of all perturbations (apart from the Weyl fluid energy density) remain sufficiently small and bounded relative to the background values for any scale of the perturbations. We also discuss and compare the stability and bounded behaviour of the perturbations in the late accelerated phase of one bounce solutions. For eternal universes we argue that some of these features are maintained only for early times (typically of the order of the first bounce). In this sense we show that eternal solutions are highly unstable configurations considering the background model of this paper.

gr-qc↗

arXiv2012

Extended Spherical Collapse and the Accelerating Universe

The influence of the shear stress and angular momentum on the nonlinear spherical collapse model is discussed in the framework of the Einstein-de Sitter (EdS) and $Λ$CDM models. By assuming that the vacuum component is not clustering within the homogeneous nonspherical overdensities, we show how the local rotation and shear affects the linear density threshold for collapse of the non-relativistic component ($δ_\mathrm{c}$) and its virial overdensity ($Δ_\mathrm{V}$). It is also found that the net effect of shear and rotation in galactic scale is responsible for higher values of the linear overdensity parameter as compared with the standard spherical collapse model (no shear and rotation).

astro-ph.CO↗

arXiv2012

Relativistic virialization in the Spherical Collapse model for Einstein-de Sitter and ΛCDM cosmologies

Spherical collapse has turned out to be a successful semi-analytic model to study structure formation in different DE models and theories of gravity. Nevertheless, the process of virialization is commonly studied on the basis of the virial theorem of classical mechanics. In the present paper, a fully generally-relativistic virial theorem based on the Tolman-Oppenheimer-Volkoff (TOV) solution for homogeneous, perfect-fluid spheres is constructed for the Einstein-de Sitter and ΛCDM cosmologies. We investigate the accuracy of classical virialization studies on cosmological scales and consider virialization from a more fundamental point of view. Throughout, we remain within general relativity and the class of FLRW models. The virialization equation is set up and solved numerically for the virial radius, y_{vir}, from which the virial overdensity Δ_{V} is directly obtained. Leading order corrections in the post-Newtonian framework are derived and quantified. In addition, problems in the application of this formalism to dynamical DE models are pointed out and discussed explicitly. We show that, in the weak field limit, the relative contribution of the leading order terms of the post-Newtonian expansion are of the order of 10^{-3}% and the solution of Wang & Steinhardt (1998) is precisely reproduced. Apart from the small corrections, the method could provide insight into the process of virialization from a more fundamental point of view.

astro-ph.CO↗