FastTracker performance using the new“Variable Resolution Associative Memory”for Atlas

We report on performances of the new “variable resolution Associative Memory (AM)”applied to the reconstruction of top pair events buried in a large pile-up environment using the ATLAS detector. The AM is a VLSI processor for pattern recognition based on Content Addressable Memory (CAM) architecture. The AM is optimized for on-line track finding in high-energy physics experiments. Pattern matching is carried out by finding track candidates in coarse resolution “roads”. A large AM bank stores all trajectories of interest, called “patterns”, for a given detector resolution. The AM extracts roads compatible with a given event during detector read-out. Two important variables characterize the quality of the AM bank: its “coverage” (the fraction of tracks that match at least one pattern in the bank) and the level of “fake candidates” (roughly proportional to the number of patterns in the bank). As the luminosity increases, the fake rate increases rapidly because of the increased silicon occupancy. To counter that, we must reduce the width of our roads to improve resolution. If we increase the road resolution using current technology, the system would become very large and extremely expensive. The “variable resolution patterns” is an elegant solution to this problem. Each pattern and each detector layer within a pattern will be able to use the best resolution, but we will use a “don’t care”feature (inspired from ternary CAMs) to reduce the resolution when a lower resolution is more appropriate. In other words we can use patterns of variable shape. As a result we reduce the number of fake candidates, while keeping the efficiency high and avoiding the bank size to blow-up due to the improved resolution. We optimize the use of this idea on the reconstruction of simulated top events at high instantaneous luminosity