BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Use of Associative Memories for L1 triggering in LHC environment.
DTSTART;VALUE=DATE-TIME:20120505T073000Z
DTEND;VALUE=DATE-TIME:20120505T080000Z
DTSTAMP;VALUE=DATE-TIME:20130520T060223Z
UID:indico-contribution-42@cern.ch
DESCRIPTION:Speakers: MAGALOTTI\, Daniel (INFN Sezione Perugia (IT))\nMode
 rn high energy physics experiments search for extremely rare processes hid
 den in much larger background levels. As the experiment complexity\, the a
 ccelerator backgrounds and luminosity increase we need increasingly exclus
 ive selections to efficiently select the rare events inside the huge backg
 round.\nIn the framework of the CMS experiment at LHC one of the identifie
 d challenges for future upgrade is the capability of using the tracker inf
 ormation to trigger events already at L1 (now they are used at L2). \nThis
  strategy requires massive computing power to minimize the online executio
 n time of complex tracking algorithms and the “combinatorial challenge
 ”.\nAssociative Memories (AM) have been already used in other experiment
 s (CDF) as a way to compare the tracker informations of each event to pre-
 calculated “expectations” (pattern matching) in a very short time and 
 contribute to the trigger decision. To use the AM approach for the CMS tra
 cker one of the main challenges is to make available the tracker data to t
 he AM processor in a very short time (6 s is the L1 latency for CMS). \
 n\nWe describe a possible application of AM in the CMS environment using t
 he existing hardware developed for other experiments\, the AMBslim mother 
 board consisting of 4 smaller boards\, the Local Associative Memory Banks 
 (LAMB)\, each hosting 32 AM chips to contain the stored patterns with the 
 readout logic. The ability of a single AMBSlim to process a single event i
 s much less than the amount of input data foreseen for the CMS case\, and 
 the latency strongly depends on the time necessary to load the data in the
  AM system and to process a single event.\nOne possible solution is to par
 allelize the event processing inside the AMBslim board assigning each even
 t to one LAMB. We describe the firmware implementation of this concept in 
 the current hardware and the results obtained.\n\nFinally we discuss a pos
 sible modification of the LAMB hardware in order to obtain the minimum del
 ay time for processing events.\n\nhttp://indico.cern.ch/contributionDispla
 y.py?contribId=42&sessionId=3&confId=154525
LOCATION:INFN Pisa
URL:http://indico.cern.ch/contributionDisplay.py?contribId=42&sessionId=3&
 confId=154525
END:VEVENT
END:VCALENDAR