Design of a Resilient, High-Throughput, Persistent Storage System for the ATLAS Phase-II DAQ System

The ATLAS experiment will undergo a major upgrade to take advantage of the new conditions provided by the upgraded High-Luminosity LHC. The Trigger and Data Acquisition system (TDAQ) will record data at unprecedented rates: the detectors will be read out at 1 MHz generating around 5 TB/s of data. The Dataflow system (DF), component of TDAQ, introduces a novel design: readout data are buffered on persistent storage while the event filtering system analyses them to select 10000 events per second for a total recorded throughput of around 60 GB/s. This approach allows for decoupling the detector activity from the event selection process. New challenges then arise for DF: design and implement a distributed, reliable, persistent storage system supporting several TB/s of aggregated throughput while providing tens of PB of capacity. In this paper we first describe some of the challenges that DF is facing: data safety with persistent storage limitations, indexing of data at high-granularity in a highly-distributed system, and high-performance management of storage capacity. Then the ongoing R&D to address each of the them is presented and the performance achieved with a working prototype is shown.

Technology Tips: January 2002

We are especially pleased to offer a technology tip for Hewlett-Packard users this month. Colin Croft, a colleague from Australia, shows some potential pitfalls in using the HP—or any other calculator, for that matter—to evaluate limits. Next, coeditor Paul Gosse gives an inside look at Graph Databases (GDBs). If you have not used a GDB before, you might try to do so after you read his suggestions. In our third piece, Bob Ruzich illustrates some nice calculus extensions of the classic “ball bounce with a motion detector” activity. Appropriately enough, our featured Web site this month is Colin Croft's own site for HP calculator users.

Fusion and Rivalry of Illusory Contours

The general question is raised: “Are visual contours given directly from striate-cortex feature-detector activity?‘’ Phenomena of ‘subjective’ or ‘cognitive’ contours are examined to challenge this view, on the ground that contours can be extrapolations across low-probability gaps. The contours may be curved and may have poor ‘gestalt’ qualities—so ‘gestalt closure’ is not appropriate, but may be a sub-class of these phenomena. It is suggested that these illusory contours (and brightness differences) are generated by perceptually postulated masking objects—these being part of perceptual ‘scene analysis strategy’, since strong evidence for nearer objects is provided by improbable gaps. Experiments are reported, in which each eye is given a different ‘cognitive’ contour figure such that there are disparate but illusory contours. It is found that these are fused to give three-dimensional illusory. surfaces bowing in front of the display. Masking objects must be in front of gaps; what happens here with reversal of stereo depth? Switching the eyes often gives rivalry of the illusory contours when masking is incompatible with the stereo depth. Implications for normal stereo vision are discussed.