Search

Power and energy consumption are becoming key challenges for the supercomputers’ exascale race. HPC systems’ processors waist active power during communication and synchronization among the MPI processes in large-scale HPC applications. However, due to the time scale at which communication happens, transitioning into low-power states while waiting for the completion of each communication may introduce unacceptable overhead. In this article, we present COUNTDOWN, a run-time library for identifying and automatically reducing the power consumption of the CPUs during communication and synchronization. COUNTDOWN saves energy without penalizing the time-to-completion by lowering CPUs power consumption only during idle times for which power state transition overhead is negligible. This is done transparently to the user, without requiring labor-intensive and error-prone application code modifications, nor requiring recompilation of the application. We test our methodology on a production Tier-1 system. For the NAS benchmarks, COUNTDOWN saves between 6 and 50 percent energy, with a time-to-solution penalty lower than 5 percent. In a complete production—Quantum ESPRESSO—for a 3.5K cores run, COUNTDOWN saves 22.36 percent energy, with a performance penalty below 3 percent. Energy saving increases to 37 percent with a performance penalty of 6.38 percent, if the application is executed without communication tuning. [ABSTRACT FROM AUTHOR]

Abstract A Triple-foil differential Plunger for Exotic Nuclei (TPEN) has been developed to measure the lifetimes of excited states in nuclei with small production cross-sections. TPEN utilises one target foil and two degrader foils to make differential lifetime measurements: directly determining the decay function and its derivative at a single plunger distance setting. The direct measurement of the decay function and its derivative removes the requirement to measure γ -ray intensities at several target-to-degrader distances, thereby reducing the beam-time required relative to a conventional plunger with a single-degrader foil. This paper describes the commissioning of TPEN in the lifetime measurement of the first excited 4 + state in 156Dy using this differential lifetime technique. The 46(2) ps measured lifetime was found to be in good agreement with the 45.6(5) ps deduced from a previous high-statistics experiment using a conventional two-foil plunger. A comparison between the differential mode with TPEN, and that using TPEN as a conventional two-foil plunger reveals that it achieves the same statistical accuracy with a ∼ 4 -fold reduction in beam-time in the differential mode, for the reaction and experimental setup used in this work. TPEN opens up new opportunities to study exotic nuclei with lower production cross-sections that were previously not accessible. [ABSTRACT FROM AUTHOR]

The hyperfine interactions effect on the results from RDDS experiments is considered. Expressions for the influenced observables are derived within the context of application of the Differential decay curve method for data analysis. In particular, a formula for the lifetime is given where, on the one hand, measured quantities—peak areas of the shifted (fast) components and on the other hand, modeled perturbed angular distribution functions are taken into account. This formula allows for both study of the hyperfine interactions and correct derivation of the lifetime of the investigated level. [ABSTRACT FROM AUTHOR]

A new procedure for the analysis of delayed-coincidence lifetime experiments focused on the Fast-timing case is proposed following the approach of the Differential decay-curve method. Examples of application of the procedure on experimental data reveal its reliability for lifetimes even in the sub-nanosecond range. The procedure is expected to improve both precision/reliability and treatment of systematic errors and scarce data as well as to provide an option for cross-check with the results obtained by means of other analyzing methods. [ABSTRACT FROM AUTHOR]

Uncertainties in absolute plunger distances are an often overlooked component in measurements with the Recoil Distance Doppler-shift (RDDS) method when the Differential Decay Curve (DDC) method cannot be used for the analysis. It is shown how these uncertainties arise when the capacitance method is used to determine absolute distances and how they influence the obtained lifetime values of a plunger experiment. Furthermore, a new approach to obtain absolute foil separations for plunger measurements using a high precision optical distance probe is introduced, which can reduce these uncertainties by a large margin. The performance of this method is demonstrated by comparisons to the capacitance method as well as to a precise measurement of absolute distances using known lifetimes in 181Ta. [ABSTRACT FROM AUTHOR]

In this brief, a novel digital duty-cycle modulation (DDCM) scheme is outlined. The digital part, embedded in a field-programmable gate array (FPGA) chip, consists of a lookup DDCM map, a dual DDCM counting logic with interrupt service routines, and a holder. It is used as a building core of a well-tested 10–12 bits FPGA-based digital-to-analog converter (DAC). Experimental tests show that for a digital input bandwidth of 3 kHZ, the new DDDM technique for DAC achieves 40 dBc of surpious free dynamic range and 60 dB of NOISE FLOOR range. This performance is a challenge compared with most low cost oversampling DACs schemes with single decimation stage. [ABSTRACT FROM PUBLISHER]

A generalization of an earlier proposed version of the Differential decay curve method is presented for the analysis of Doppler-shift attenuation lifetime measurements. The lifetime is derived directly from the line shapes of the depopulating and feeding transitions without any assumptions about or fitting of the time dependence of the population of the corresponding levels except for unobserved feeding when relevant. Fitting of the line shapes is also not necessary. The only approximation involved is related to the continuous treatment of the nuclear scattering events in the Monte Carlo simulation needed. Tests with simulated and real data reveal good reliability of this method. We propose also a new precise procedure where the lifetime is derived by fitting the time dependence of the population of the level of interest using the line shape of the depopulating transition and the difference of the spectra of the depopulating and feeding transitions. Practical application to simulated and real data proves the applicability of the new procedure. [ABSTRACT FROM AUTHOR]

Abstract: The Recoil-Distance Doppler-Shift (RDDS) or Plunger technique is a well established method to measure lifetimes of excited nuclear states in the pico-second range. In standard RDDS experiments, the velocities of the nuclei of interest emerging from a usually thin target foil are distributed around a mean velocity with a relatively narrow width and it is sufficient to assume that all nuclei move with the average velocity. In this paper we investigate the influence of a broader velocity distribution especially for lifetimes τ determined using the DDCM and its basic relation and simulated experimental data (R(x) decay curve of the level of interest, feeding decay curves). It turned out that it is favorable to use instead of . Further, deviations from the correct lifetimes practically vanish at target to stopper separations close to the maximum amplitude of the function . As a consequence in a plunger experiment target-to-stopper separations should be selected symmetrically around the maximum amplitude of the function in order to minimize the effect of a broad velocity distribution. [Copyright &y& Elsevier]

Abstract: A new version of the differential decay curve method is proposed for the analysis of Doppler-shift attenuation lifetime measurements. The lifetime is derived directly from the line shapes of the depopulating and feeding transitions on the basis of the Blaugrund approximation without including any assumptions or fitting of the time dependence of the population of the corresponding levels. For specific simulated cases, the method shows promise for its applicability. In the future, we intend to generalize the method for the case of an arbitrary multi-detector setup. [Copyright &y& Elsevier]

Abstract: A new plunger device has been designed and built to measure the lifetimes of unbound states in exotic nuclei beyond the proton drip-line. The device has been designed to work in both vacuum and dilute-gas environments made possible through the introduction of a low-voltage stepping motor. DPUNS will be used in conjunction with the gas-filled separator RITU and the vacuum separator MARA at the accelerator laboratory of the University of Jyväskylä, Finland, to measure the lifetimes of excited states with low population cross-sections. This is achieved by eliminating the need for a carbon foil to isolate the helium gas of RITU from the beam line thus reducing the background from beam-foil reactions. The inclusion of a high-sampling rate data acquisition card increases further the sensitivity of the device. The plunger will be used to address many key facets of nuclear structure physics with particular emphasis on the effect of deformation on proton emission rates. [Copyright &y& Elsevier]

A charge plunger device has been commissioned based on the DPUNS plunger (Taylor et al., 2013) using the in-flight mass separator MARA at the University of Jyväskylä. The 152 Sm (32 S ,4 n)180 Pt reaction was used to populate excited states in 180 Pt. A lifetime measurement of the 2 1 + state was performed by applying the charge plunger technique, which relies on the detection of the charge state-distribution of recoils rather than the detection of the emitted γ rays. This state was a good candidate to test the charge plunger technique as it has a known lifetime and depopulates through a converted transition that competes strongly with γ -ray emission. The lifetime of the 2 1 + state was measured to be 480 (10) ps , which is consistent with previously reported lifetimes that relied on the standard γ -ray techniques. The charge plunger technique is a complementary approach to lifetime measurements of excited states that depopulate through both γ -ray emission and internal conversion. In cases where it is not possible to detect Doppler-shifted γ rays, for example, in heavy nuclei where internal conversion dominates, it may well be the only feasible lifetime analysis approach. [ABSTRACT FROM AUTHOR]

A new method of extracting the γ -ray intensities necessary to perform lifetime measurements using the differential decay curve method (DDCM) is presented in this work, the unresolved Doppler-shifted components method (UDCM). The UDCM allows for a DDCM analysis to be performed using a γ -ray transition for which the fully Doppler-shifted and degraded components are unresolvable in energy and so are detected as a single peak. This technique was used to measure the known lifetime of the yrast 2 1 + state in 50 Mn with a depopulating transition that does not have resolvable fully Doppler-shifted and degraded components. The lifetime measured through applying the UDCM was consistent with the standard DDCM measurement of the 2 1 + state. Use of the UDCM allows for DDCM lifetime measurements to be made using transitions of smaller γ -ray energies, smaller recoil velocities and, in some cases, with a smaller uncertainty. In contrast to a standard DDCM analysis, a UDCM analysis is also independent of the widths of the fully Doppler-shifted and degraded components and as a result they do not need to be determined. [ABSTRACT FROM AUTHOR]