MeerKAT

Abstract:

The nearby accreting neutron star binary Sco X-1 is the closest example of ongoing relativistic jet production at high Eddington ratios. Previous radio studies have revealed that alongside mildly relativistic, radio-emitting ejecta, there is at times a much faster transfer of energy from the region of the accretion flow along the jet. The nature of this ultrarelativistic flow remains unclear and while there is some evidence for a similar phenomenon in other systems that might contain neutron stars, it has never been observed in a confirmed black hole system. We have compared these previous radio observations with a new analysis of simultaneous X-ray observations that were performed with the RXTE mission. We find that the ejection of the ultrarelativistic flow seems to be associated with the simultaneous appearance of two particular types of quasi-periodic oscillations in the X-ray power spectrum. In contrast, the mildly relativistic, radio-emitting outflows may be associated with flat-topped broad-band noise in the X-ray power spectrum. This is the first time a link, albeit tentative, has been found between these mysterious unseen flows and the accretion flow from which they are launched.

Abstract:

The study of binary pulsars enables tests of general relativity. Orbital motion in binary systems causes the apparent pulsar spin frequency to drift, reducing the sensitivity of periodicity searches. Acceleration searches are methods that account for the effect of orbital acceleration. Existing methods are currently computationally expensive, and the vast amount of data that will be produced by next-generation instruments such as the Square Kilometre Array necessitates real-time acceleration searches, which in turn requires the use of high-performance computing (HPC) platforms. We present our implementation of the correlation technique for the Fourier Domain Acceleration Search (FDAS) algorithm on Graphics Processor Units (GPUs). The correlation technique is applied as a convolution with multiple finite impulse response (FIR) filters in the Fourier domain. Two approaches are compared: the first uses the NVIDIA cuFFT library for applying Fast Fourier transforms (FFTs) on the GPU, and the second contains a custom FFT implementation in GPU shared memory. We find that the FFT shared-memory implementation performs between 1.5 and 3.2 times faster than our cuFFT-based application for smaller but sufficient filter sizes. It is also 4–6 times faster than the existing GPU and OpenMP implementations of FDAS. This work is part of the AstroAccelerate project, a many-core accelerated time-domain signal-processing library for radio astronomy.