Improving checkpointing intervals by considering individual job failure probabilities
Checkpointing is a popular resilience method in HPC and its efficiency highly depends on the choice of the checkpoint interval. Standard analytical approaches optimize intervals for big, long-running jobs that fail with high probability, while they are unable to minimize checkpointing overheads for jobs with a low or medium probability of failing. Nevertheless, our analysis of batch traces of four HPC systems shows that these jobs are extremely common.We therefore propose an iterative checkpointing algorithm to compute efficient intervals for jobs with a medium risk of failure. The method also supports big and long-running jobs by converging to the results of various traditional methods for…
Reappraising the appropriate calculation of a common meteorological quantity: Potential Temperature
Abstract. The potential temperature is a widely used quantity in atmospheric science since it is conserved for air's adiabatic changes of state. Its definition involves the specific heat capacity of dry air, which is traditionally assumed as constant. However, the literature provides different values of this allegedly constant parameter, which are reviewed and discussed in this study. Furthermore, we derive the potential temperature for a temperature-dependent parameterization of the specific heat capacity of dry air, thus providing a new reference potential temperature with a more rigorous basis. This new reference shows different values and vertical gradients in the upper troposphere and …
Terahertz electrical writing speed in an antiferromagnetic memory
The speed of writing of state-of-the-art ferromagnetic memories is physically limited by an intrinsic gigahertz threshold. Recently, realization of memory devices based on antiferromagnets, in which spin directions periodically alternate from one atomic lattice site to the next has moved research in an alternative direction. We experimentally demonstrate at room temperature that the speed of reversible electrical writing in a memory device can be scaled up to terahertz using an antiferromagnet. A current-induced spin-torque mechanism is responsible for the switching in our memory devices throughout the 12-order-of-magnitude range of writing speeds from hertz to terahertz. Our work opens the…
In-Situ observation of New Particle Formation in the upper troposphere/lower stratosphere of the Asian Monsoon Anticyclone
Abstract. During the monsoon season of the year 2017 the airborne StratoClim mission took place in Kathmandu, Nepal with eight mission flights of the M-55 Geophysica in the upper troposphere/lower stratosphere (UT/LS) of the Asian Monsoon Anticyclone (AMA) over northern India, Nepal and Bangladesh. More than hundred events of New Particle Formation (NPF) were observed. In total, more than two hours of flight time were spent under NPF conditions as indicated by the abundant presence of ultrafine aerosols, i.e. with particle diameters dp smaller than 15 nm, which were in-situ detected by means of condensation nuclei counting techniques. Mixing ratios of ultrafine particles (nuf) of up to ~ 50…
New particle formation inside ice clouds: In-situ observations in the tropical tropopause layer of the 2017 Asian Monsoon Anticyclone
Abstract. From 27 July to 10 August 2017 the airborne StratoClim mission took place in Kathmandu, Nepal where eight mission flights were conducted with the M-55 Geophysica up to altitudes of 20 km. New Particle Formation (NPF) was identified by the abundant presence of ultrafine aerosols, with particle diameters dp smaller than 15 nm, which were in-situ detected by means of condensation nuclei counting techniques. NPF fields in clear-skies as well as in the presence of cloud ice particles (dp > 3 µm) were encountered at upper troposphere/lowermost stratosphere (UT/LS) levels and within the Asian Monsoon Anticyclone (AMA). NPF-generated ultrafine particles in elevated concentrations (Nuf)…
On numerical broadening of particle-size spectra: a condensational growth study using PyMPDATA 1.0
This work discusses the numerical aspects of representing the condensational growth of particles in models of aerosol systems such as atmospheric clouds. It focuses on the Eulerian modelling approach, in which fixed-bin discretisation is used for the probability density function describing the particle-size spectrum. Numerical diffusion is inherent to the employment of the fixed-bin discretisation for solving the arising transport problem (advection equation describing size spectrum evolution). The focus of this work is on a technique for reducing the numerical diffusion in solutions based on the upwind scheme: the multidimensional positive definite advection transport algorithm (MPDATA). S…
Algorithmic differentiation for cloud schemes (IFS Cy43r3) using CoDiPack (v1.8.1)
Abstract. Numerical models in atmospheric sciences not only need to approximate the flow equations on a suitable computational grid, they also need to include subgrid effects of many non-resolved physical processes. Among others, the formation and evolution of cloud particles is an example of such subgrid processes. Moreover, to date there is no universal mathematical description of a cloud, hence many cloud schemes have been proposed and these schemes typically contain several uncertain parameters. In this study, we propose the use of algorithmic differentiation (AD) as a method to identify parameters within the cloud scheme, to which the output of the cloud scheme is most sensitive. We il…
On numerical broadening of particle size spectra: a condensational growth study using PyMPDATA 1.0
Abstract. The work discusses the diffusional growth in particulate systems such as atmospheric clouds. It focuses on the Eulerian modeling approach in which the evolution of the probability density function describing the particle size spectrum is carried out using a fixed-bin discretization. The numerical diffusion problem inherent to the employment of the fixed-bin discretization is scrutinized. The work focuses on the applications of MPDATA family of numerical schemes. Several MPDATA variants are explored including: infinite-gauge, non-oscillatory, third-order-terms and recursive antidiffusive correction (double pass donor cell, DPDC) options. Methodology for handling coordinate transfor…
On numerical broadening of particle size spectra: a condensational growth study using PyMPDATA
This work discusses the numerical aspects of representing the diffusional (condensational) growth in particulate systems such as atmospheric clouds. It focuses on the Eulerian modeling approach, in which the evolution of the particle size spectrum is carried out using a fixed-bin discretization associated with inherent numerical diffusion. Focus is on the applications of MPDATA numerical schemes (variants explored include: infinite-gauge, non-oscillatory, third-order-terms and recursive antidiffusive correction). Methodology for handling coordinate transformations associated with both particle size distribution variable choice and numerical grid layout are expounded. Analysis of the perform…
Impact of formulations of the homogeneous nucleation rate on ice nucleation events in cirrus
Abstract. Homogeneous freezing of solution droplets is an important pathway of ice formation in the tropopause region. The nucleation rate can be parameterized as a function of water activity, based on empirical fits and some assumptions on the underlying properties of super-cooled water, although a general theory is missing. It is not clear how nucleation events are influenced by the exact formulation of the nucleation rate or even their inherent uncertainty. In this study we investigate the formulation of the nucleation rate of homogeneous freezing of solution droplets (1) to link the formulation to the nucleation rate of pure water droplets, (2) to derive a robust and simple formulation …
Algorithmic Differentiation for Cloud Schemes
<p>Numerical models in atmospheric sciences do not only need to approximate the flow equations on a suitable computational grid, they also need to include subgrid effects of many non-resolved physical processes. Among others, the formation and evolution of cloud particles is an example of such subgrid processes. Moreover, to date there is no universal mathematical description of a cloud, hence many cloud schemes were proposed and these schemes typically contain several uncertain parameters. In this study, we propose the use of algorithmic differentiation (AD) as a method to identify parameters within the cloud scheme, to which the output of the cloud scheme is most sensitive.…