Virtual Machine and Cloud Computing Security [Slides]
Abstract not provided.
SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Abstract not provided.
Power is becoming an increasingly important concern for large super computercenters. However,todate,therehavebeenadearthofstudiesof power usage “in the wild”—on production supercomputers running pro duction workloads. In this talk, we present the initial results of an effort to characterize the power usage of three Top100 supercomputers at Los Alamos National Laboratory: Cielo, Roadrunner, and Luna (#18, #22, and #64, respectively, on the November 2012 Top500 list). Power measure ments taken both at the switchboard level and within the compute racks are presented and discussed.
Cielo is a massively parallel supercomputer funded by the DOE/NNSA Advanced Simulation and Computing (ASC) program, and operated by the Alliance for Computing at Extreme Scale (ACES), a partnership between Los Alamos National Laboratory (LANL) and Sandia National Laboratories (SNL). The primary Cielo compute platform is physically located at Los Alamos National Laboratory. This Cielo Computational Environment Usage Model documents the capabilities and the environment to be provided for the Q1 FY12 Level 2 Cielo Capability Computing (CCC) Platform Production Readiness Milestone. This document describes specific capabilities, tools, and procedures to support both local and remote users. The model is focused on the needs of the ASC user working in the secure computing environments at Lawrence Livermore National Laboratory (LLNL), Los Alamos National Laboratory, or Sandia National Laboratories, but also addresses the needs of users working in the unclassified environment. The Cielo Computational Environment Usage Model maps the provided capabilities to the tri-Lab ASC Computing Environment (ACE) Version 8.0 requirements. The ACE requirements reflect the high performance computing requirements for the Production Readiness Milestone user environment capabilities of the ASC community. A description of ACE requirements met, and those requirements that are not met, are included in each section of this document. The Cielo Computing Environment, along with the ACE mappings, has been issued and reviewed throughout the tri-Lab community.
(U) This document is written to describe issues that impact on the ability of codes to converge to a unique answer. Several issues that could be discussed in the area of mesh convergence and resolution the topics include the resolution of geometry, stress gradients, shock propagation, material softening, and localization of deformation. After a quick discussion of what mesh convergence means, this paper presents a review of some of the criteria for mesh convergence. With an introductory level understanding of these criteria, two problems are presented to demonstrate convergence and non-convergence. The first example is a simple convergent hemisphere of gas, known as the Noh problem. With this problem convergence is shown. The Noh problem is also used to investigate the impact of mesh stiffening, or viscosity. The second problem is that of an explosively driven hemisphere. The hemisphere made of U6Nb with an inner radius of 40 mm and an outer radius of 41 mm was filled with PBX9501 and tested. Test results included surface temperature and information on the breakup, such as distribution of fines. Convergence computational methods are discussed along with some proposals for obtaining mesh convergence for a wide range of applications.
Shock initiation in a plastic-bonded explosives (PBX) is due to hot spots. Current reactive burn models are based, at least heuristically, on the ignition and growth concept. The ignition phase occurs when a small localized region of high temperature (or hot spot) burns on a fast time scale. This is followed by a growth phase in which a reactive front spreads out from the hot spot. Propagating reactive fronts are deflagration waves. A key question is the deflagration speed in a PBX compressed and heated by a shock wave that generated the hot spot. Here, the ODEs for a steady deflagration wave profile in a compressible fluid are derived, along with the needed thermodynamic quantities of realistic equations of state corresponding to the reactants and products of a PBX. The properties of the wave profile equations are analyzed and an algorithm is derived for computing the deflagration speed. As an illustrative example, the algorithm is applied to compute the deflagration speed in shock compressed PBX 9501 as a function of shock pressure. The calculated deflagration speed, even at the CJ pressure, is low compared to the detonation speed. The implication of this are briefly discussed.
A recent research effort at LANL proposed info-gap decision theory as a framework by which to measure the predictive maturity of numerical models. Info-gap theory explores the trade-offs between accuracy, that is, the extent to which predictions reproduce the physical measurements, and robustness, that is, the extent to which predictions are insensitive to modeling assumptions. Both accuracy and robustness are necessary to demonstrate predictive maturity. However, conducting an info-gap analysis can present a formidable challenge, from the standpoint of the required computational resources. This is because a robustness function requires the resolution of multiple optimization problems. This report offers an alternative, adjoint methodology to assess the info-gap robustness of Ax = b-like numerical models solved for a solution x. Two situations that can arise in structural analysis and design are briefly described and contextualized within the info-gap decision theory framework. The treatments of the info-gap problems, using the adjoint methodology are outlined in detail, and the latter problem is solved for four separate finite element models. As compared to statistical sampling, the proposed methodology offers highly accurate approximations of info-gap robustness functions for the finite element models considered in the report, at a small fraction of the computational cost. It is noted that this report considers only linear systems; a natural follow-on study would extend the methodologies described herein to include nonlinear systems.
Project Goals are: 1. Demonstrate a paradigm shift in the simulation approach to each application on an advanced architecture. 2. Deliver a documented computational co-design knowledge-base, built upon an evolving software infrastructure 3. Foster broad, long-term research collaborations with appropriate hardware partners 4. Deepen understanding of, and our experience-base with, consistent, two-way, scale-bridging algorithms.
DC is launching a new program to recognize noteworthy achievements made by users using High Performance Computing (HPC) technologies. The program’s main goals are to: • Showcase success stories involving HPC in science and industry • Help other users better understand the benefits of adopting HPC and justify HPC investments, especially for SMBs • Demonstrate the value of HPC to funding bodies • Expand public support for increased HPC investments.
The expected or mean neutron number (or density) provides an adequate characterization of the neutron population and its dynamical excursions in most neutronic applications, in particular power reactors. Fluctuations in the neutron number, originating from the inherent randomness of neutron interactions and fission neutron multiplicities, are relatively small and ignorable for operational purposes, although measurements of the variance and time correlations provide valuable diagnostic information on fundamental reactor physics parameters. However, it is well known that there exist situations of great interest and importance in which a strictly deterministic description, or even one supplemented with a knowledge of low order statistical averages (variance, correlation), provides an incomplete and very unsatisfactory description of the state of the neutron population. These situations are marked by persistent large fluctuations in the neutron number where the emergence of a deterministic phase is suppressed. Such situations are strongly stochastic and therefore unpredictable (i.e., the mean is not representative of the actual population), and can arise either by design or by accident. Examples where the stochastic behavior of neutron populations must be taken into account include: nuclear weapon single-point safety assessment; criticality excursions in spent fuel storage and in the handling of fissile solutions in fuel fabrication and reprocessing; approach to critical under suboptimal reactor start-up conditions; preinitiation in fast burst research reactors; and weak nuclear signatures in the passive detection of nuclear materials. What distinguishes strongly stochastic neutronic systems from strongly deterministic systems is that, in the former, neutron multiplication occurs in the presence of weak neutron sources, such as spontaneous fission and background (cosmic) radiation. Weak sources (in a sense that can be made quite precise) lead to well separated fission chains (a fission chain is defined as the initial source neutron and all its subsequent progeny) in which some chains are short lived while others propagate for unusually long times. Under these conditions, fission chains do not overlap strongly and this precludes the cancellation of neutron number fluctuations necessary for the mean to become established as the dominant measure of the neutron population. The fate of individual chains then plays a defining role in the evolution of the neutron population in strongly stochastic systems, and of particular interest and importance in supercritical systems is the extinction probability, defined as the probability that the neutron chain (initiating neutron and its progeny) will be extinguished at a particular time, or its complement, the time-dependent survival probability. The time-asymptotic limit of the latter, the probability of divergence, gives the probability that the neutron population will grow without bound, and is more commonly known as the probability of initiation or just POI. The ability to numerically compute these probabilities, with high accuracy and without overly restricting the underlying physics (e.g., fission neutron multiplicity, reactivity variation) is clearly essential in developing an understanding of the behavior of strongly stochastic systems.
Network Monitoring requires a substantial use of data and error analysis to overcome issues with clusters. Zenoss and Splunk help to monitor system log messages that are reporting issues about the clusters to monitoring services. Infiniband infrastructure on a number of clusters upgraded to ibmon2. ibmon2 requires different filters to report errors to system administrators. Focus for this summer is to: (1) Implement ibmon2 filters on monitoring boxes to report system errors to system administrators using Zenoss and Splunk; (2) Modify and improve scripts for monitoring and administrative usage; (3) Learn more about networks including services and maintenance for high performance computing systems; and (4) Gain a life experience working with professionals under real world situations. Filters were created to account for clusters running ibmon2 v1.0.0-1 10 Filters currently implemented for ibmon2 using Python. Filters look for threshold of port counters. Over certain counts, filters report errors to on-call system administrators and modifies grid to show local host with issue.
The two primary purposes of LANL’s Computational Physics Student Summer Workshop are (1) To educate graduate and exceptional undergraduate students in the challenges and applications of computational physics of interest to LANL, and (2) Entice their interest toward those challenges. Computational physics is emerging as a discipline in its own right, combining expertise in mathematics, physics, and computer science. The mathematical aspects focus on numerical methods for solving equations on the computer as well as developing test problems with analytical solutions. The physics aspects are very broad, ranging from low-temperature material modeling to extremely high temperature plasma physics, radiation transport and neutron transport. The computer science issues are concerned with matching numerical algorithms to emerging architectures and maintaining the quality of extremely large codes built to achieve multi-physics calculations. Although graduate programs associated with computational physics are emerging, it is apparent that the pool of U.S. citizens in this multidisciplinary field is relatively small and is typically not focused on the aspects that are of primary interest to LANL. Furthermore, more structured foundations for LANL interaction with universities in computational physics is needed; currently interactions rely almost solely on individuals’ personalities and personal contacts. Thus a tertiary purpose of the Summer Workshop is to build an educational network of LANL researchers, university professors, and emerging students to advance the field and LANL’s involvement in it. This was the second year for the Summer Workshop. Like the previous year, the workshop’s goals were achieved by bringing into LANL a select group of students recruited from across the United States and immersing them for ten weeks in lectures and interesting research projects. The lectures provided an overview of the computational physics topics of interest this year along with some detailed instruction while the projects gave the students a positive experience accomplishing technical goals. Each team consisted of two students working under one or more mentors from LANL on specific research projects associated with predefined topics. This year, the topics were on verification of simulations in plasma and radiation-hydrodynamics codes, hypervelocity material deformation, plasma physics, graphical processor unit (“GPU”) programming, multi-material mixing, turbulence modeling, computational material science, and electromagnetic pulse simulations. The students’ growth was furthered by their participation on teams where their teammates were sometimes of different academic rank. It also developed them by requiring them to produce written and oral reports that they presented to peers, mentors, and management.
Virtualization has become the standard for computing within organizations. The questions that arise when considering virtualization and cloud computing are whether these systems are, or can be, as secure as physical machines, and whether they are appropriate for production services. This paper identifies the security concerns that are associated with virtualization and cloud computing, and how they can be addressed in order to properly secure these systems. In order to accomplish this, the first goal of this paper was to identify relevant security concerns. The second goal was to identify measures that could be put in place to mitigate and address these concerns. HPC-3’s production services are evaluated in the context of available virtualization technologies, with recommendations for appropriate options.
Overview of this presentation is (1) Exascale computing - different technologies, getting there; (2) high-performance proof-of-concept MCMini - features and results; and (3) OpenCL toolkit - Oatmeal (OpenCL Automatic Memory Allocation Library) - purpose and features. Despite driver issues, OpenCL seems like a good, hardware agnostic tool. MCMini demonstrates the possibility for GPGPU-based Monte Carlo methods - it shows great scaling for HPC application and algorithmic equivalence. Oatmeal provides a flexible framework to aid in the development of scientific OpenCL codes.
We consider a constant temperature spherical shell of isotropic, homogeneous, linearly elastic material with density ρ and Lamé coefficients λ and μ. The inner and outer radii of the shell are r i and r o , respectively. We assume that the inside of the shell is a void. On the outside of the shell, we apply a uniform, time-varying pressure p(t). We also assume that the shell is initially at rest. We want to compute the jump-off time and velocity of the pressure wave, which are the first time after t = 0 at which the pressure wave from the outer surface reaches the inner surface. This analysis computes the jump-off velocity and time for both compressible and incompressible materials. This differs substantially from [3], where only incompressible materials are considered. We will consider the behavior of an impulsively loaded, exponentially decaying pressure wave p(t) = P 0 e -αt , where α ≥ 0. We notice that a constant pressure wave P(t) = P 0 is a special case (α = 0) of a decaying pressure wave. Both of these boundary conditions are considered in [3].
Within the last decade, there has been increasing interest in the calculation of cross section sensitivity coefficients of k eff for integral experiment design and uncertainty analysis. The OECD/NEA has an Expert Group devoted to Sensitivity and Uncertainty Analysis within the Working Party for Nuclear Criticality Safety. This expert group has developed benchmarks to assess code capabilities and performance for doing sensitivity and uncertainty analysis. Phase III of a set of sensitivity benchmarks evaluates capabilities for computing sensitivity coefficients. MCNP6 has the capability to compute cross section sensitivities for k eff using continuous-energy physics. To help verify this capability, results for the Phase III benchmark cases are generated and submitted to the Expert Group for comparison. The Phase III benchmark has three cases: III.1, an array of MOX fuel pins, III.2, a series of infinite lattices of MOX fuel pins with varying pitches, and III.3 two spheres with homogeneous mixtures of UF 4 and polyethylene with different enrichments.
All activities of the Exascale Co-design Center for Materials in Extreme Environments (ExMatEx) are focused on the two ultimate goals of the project: (1) demonstrating and delivering a prototype scale-bridging materials science application based upon adaptive physics refinement, and (2) identifying the requirements for the exascale ecosystem that are necessary to perform computational materials science simulations (both single- and multi-scale). During the first year of ExMatEx, our focus was on establishing how we do computational materials science, by developing an initial suite of flexible proxy applications. These “proxy apps” are the primary vehicle for the co-design process, involving assessments and tradeoff evaluations both within the ExMatEx team, and with the entire exascale ecosystem. These interactions have formed the basis of our second year activities. The set of artifacts from these co-design interactions are the lessons learned, that are used to re-express the applications and algorithms within the context of emerging architectures, programming models, and runtime systems.
As the latest version of the fast-tube Detonation Shock Dynamics (DSD) solver is linked with the Los Alamos Lagrangian hydrocode, verification problems from a 2006 DSD report (LA-14277) have been duplicated with some of the verification criteria changed to more quantitative ones. The observed error convergence is as good as or better than reported in, quite possibly due to the careful treatment of floating point numbers to ensure that their precision level is maintained throughout the code. This report duplicates the three sample verification problems in LA-14277 using the Los Alamos ASC Lagrangian hydrocode (FLAG), official release of 3.2 Alpha6 with a few modifications. This version of FLAG is linked with the latest fast-tube Detonation Shock Dynamics (DSD) version beta 2 solver released in 2011 as part of the LanlDSD software product. New verification criteria are used for the arcwave problem where two specific locations are chosen for burn arrival time comparison. For this report FLAG's internal driver code prepares the distance function ({Psi}) and material ID fields from its hydro setup, instead of the stand-alone driver that is being utilized by the other LANL hydrocodes currently interfaced to LanlDSD. As it is implemented in version 3.2 Alpha6, the {Psi} and material ID fields and other parameters are passed from FLAG to the DSD solver directly, and the burn table is directly passed back to FLAG as part of the calling arguments. The burn-front arrival time 'exact' solutions, mentioned in the sequel for the rate-stick and 'arc-wave' problems, are computed using a pair of special-purpose Fortran codes provided by Aslam. In each case an ansatz for the form of the solution is made in which the radius from the detonator center point is used as the independent space coordinate. This leads to a simplified, problem-specific, 1D form of the governing equation. This equation is solved using 2nd-order spatial differencing and the forward Euler method on a very fine temporal and geometric mesh. The boundary conditions are handled exactly at the correct location, with second order accuracy. Care has been taken to ensure that this solution is fully converged.
We did not run with a 'cylindrically painted region'. However, we did compute two general variants of the original problem. Refinement studies where a single zone at each level of refinement contains the entire internal energy at t=0 or A 'finite' energy source which has the same physical dimensions as that for the 91 x 46 mesh, but consisting of increasing numbers of zones with refinement. Nominal mesh resolution: 91 x 46. Other mesh resolutions: 181 x 92 and 361 x 184. Note, not identical to the original specification. To maintain symmetry for the 'fixed' energy source, the mesh resolution was adjusted slightly. FLAG Lagrange or full (Eulerian) ALE was used with various options for each simulation. Observation - for either Lagrange or ALE, point or 'fixed' source, calculations converge on density and pressure with mesh resolution, but not energy, (not vorticity either).