The (e, e 0 p) reaction was studied on targets of C, Fe, and Au at momentum transfers squared Q 2 of 0.6, 1.3, 1.8, and 3.3 GeV 2 in a region of kinematics dominated by quasifree electron-proton scattering. Missing energy and missing momentum distributions are reasonably well described by plane wave impulse approximation calculations with Q 2 and A dependent corrections that measure the attenuation of the final state protons. [S0031-9007 (98) The (e, e 0 p) reaction with nearly free electron-proton kinematics (quasifree) has proven to be a valuable tool to study the propagation of nucleons in the nuclear medium [1][2][3]. The relatively weak interaction of the electron with the nucleus allows the electrons to penetrate the nuclear interior and knock out protons. These studies complement nucleon-induced measurements of proton propagation in nuclei which give more emphasis to the nuclear surface. This paper reports the first results of a systematic study of the quasifree knockout of protons of 300-1800 MeV kinetic energy from carbon, iron, and gold targets. This energy range includes the minimum of the nucleon-nucleon (N-N) total cross section, the rapid rise in this cross section with energy above the pion production threshold, and extends to the long plateau in the energy dependence of the N-N total cross section. These features of the N-N interaction would be expected to be reflected in the energy dependence of attenuation of protons as they pass 5072 0031-9007͞98͞80(23)͞5072(5)$15.00

Inclusive electron scattering is measured with 4.045 GeV incident beam energy from C, Fe, and Au targets. The measured energy transfers and angles correspond to a kinematic range for Bjorken x . 1 and momentum transfers from Q 2 1 7 ͑GeV͞c͒ 2 . When analyzed in terms of the y-scaling function the data show for the first time an approach to scaling for values of the initial nucleon momenta significantly greater than the nuclear matter Fermi momentum (i.e., .0.3 GeV͞c). High energy electron scattering from nuclei can provide important information on the wave function of nucleons in the nucleus. In particular, with simple assumptions about the reaction mechanism, scaling functions can be deduced that, if shown to scale (i.e., are independent of length scale or momentum transfer), can provide information about the momentum and energy distribution of nucleons in a nucleus. Several theoretical studies [1][2][3][4] have indicated that such measurements may provide direct access to short-range nucleon-nucleon correlations.The concept of y scaling in electron-nucleus scattering was first introduced by West [5] and Kawazoe et al. [6]. They showed that in the impulse approximation, if quasielastic scattering from a nucleon in the nucleus was the dominant reaction mechanism, a scaling function F͑ y͒ could be extracted from the measured cross section which was related to the momentum distribution of the nucleons in the nucleus. In the simplest approximation the corresponding scaling variable y is the minimum momentum of the struck nucleon along the direction of the virtual photon. In general the scaling function depends on both y and momentum transfer-F͑ y, Q 2 ͒-but at sufficiently high Q 2 (2Q 2 is the square of the four-momentum transfer) the dependence on Q 2 should vanish yielding scaling. However, the simple impulse approximation picture breaks down when the final-state interactions (FSI) of the struck nucleon with the rest of the nucleus are included. Previous calculations [7][8][9][10][11][12][13][14] suggest that the contributions from final-state interactions should vanish at sufficiently high Q 2 . A previous SLAC measurement [15] suggested an approach to the scaling limit for heavy nuclei but only for low values of j yj , 0.3 GeV͞c at momentum transfers up to 3 ͑GeV͞c͒ 2 . The data presented here represent a significant increase in the Q 2 range compared to previous measurements while also extending the coverage in y.The present data were obtained in Hall C at the Thomas Jefferson National Accelerator Facility (TJNAF), using 4.045 GeV electron beams with intensities from 10-80 mA. The absolute beam energy was calibrated to 0.08% using 0.8 GeV elastic scattering from carbon and BeO targets and 4.0 GeV elastic scattering from hydrogen. The beam current was monitored with three calibrated resonant cavities. The beam energy resolution was better than 0.05% as defined by the accelerator acceptance. Solid targets of C (2.1% and 5.9% of a radiation length), Fe (1.5% and 5.8% of a radiation length), and Au (5.8% of a r...

Abstract-We found that interactive services at Bing have highly variable datacenter-side processing latencies because their processing consists of many sequential stages, parallelization across 10s-1000s of servers and aggregation of responses across the network. To improve the tail latency of such services, we use a few building blocks: reissuing laggards elsewhere in the cluster, new policies to return incomplete results and speeding up laggards by giving them more resources. Combining these building blocks to reduce the overall latency is non-trivial because for the same amount of resource (e.g., number of reissues), different stages improve their latency by different amounts. We present Kwiken, a framework that takes an end-to-end view of latency improvements and costs. It decomposes the problem of minimizing latency over a general processing DAG into a manageable optimization over individual stages. Through simulations with production traces, we show sizable gains; the 99 th percentile of latency improves by over 50% when just 0.1% of the responses are allowed to have partial results and by over 40% for 25% of the services when just 5% extra resources are used for reissues.

Although analyses of in-plane aliasing have been done for conventional computed tomography (CT) images, longitudinal aliasing in spiral CT has not been properly investigated. We propose a mathematical model of the three-dimensional (3-D) sampling scheme in spiral CT and analyze its effects on longitudinal aliasing. We investigated longitudinal aliasing as a function of the helical-interpolation algorithm, pitch, and reconstruction interval using CT simulations and actual phantom scans. Our model predicts, and we verified, that for a radially uniform object at the isocenter, the spiral sampling scheme results in spatially varying cancellation of the aliased spectral islands which, in turn, results in spatially varying longitudinal aliasing. The aliasing is minimal at the scanner isocenter, but worsens with distance from it and rapidly becomes significant. Our results agree with published results observed at the isocenter of the scanner and further provide new insight into the aliasing conditions at off-isocenter locations with respect to the pitch, interpolation algorithm, and reconstruction interval. We conclude that longitudinal aliasing at off-isocenter locations can be significant, and that its magnitude and effects cannot be predicted by measurements made only at the scanner isocenter.