1 / 12

ILC DAQ

ILC DAQ. JJRussell. What Can Be Said At This Time ?. Terminology - Front-End vs Back-End Broad shape of the DAQ Management and Social Issues Summary. Front-End vs Back-End. The Front-End is defined as The portion of the DAQ running at the beam-bucket rate The Back-End is defined as

nan
Download Presentation

ILC DAQ

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. ILC DAQ JJRussell

  2. What Can Be Said At This Time ? • Terminology - Front-End vs Back-End • Broad shape of the DAQ • Management and Social Issues • Summary

  3. Front-End vs Back-End • The Front-End is defined as • The portion of the DAQ running at the beam-bucket rate • The Back-End is defined as • The asynchronous portion responsible for • Collecting the data during inter-spill time • Organizing the data (event-building) • Too early to say much about the back-end • Extremely sensitive to available technology, except… • Debate of Commercial-Off-the-Shelf vs Custom, both have issues • COTS - failure to live up to expectations • Custom - expensive, long lead times

  4. Broad Shape of the DAQ • Duty cycle is the main driver • Now that it is known, can work with it • Main Features • Heavy buffering in the front-end • No hardware triggering

  5. Front-End • Luminosity/backgrounds force front-end buffering • Likely a fixed depth per channel • Likely tight due to a number of constraints • Buffer usage can be controlled either by • Central Trigger • Local Autonomous Decision • The latter is the choice of front-end architects • Puts a burden to get the assumptions on occupancy correct • If beginning of train noisy, will eat all the buffering

  6. Front-End - Continued • Almost certainly controlled by FPGA & ASICs • Opening guns in where to put functionality • Purpose built devices • General purpose CPUs • Power Issues • With this much stuff on the detector • Power will be an issue • Front-Ends gated off during inter-spill time

  7. Front-End - Continued • Process Control Issues • The FPGA/ASIC code is complicated • Good code management techniques must be adopted • Revision tracking, particularly in the field • Lead-time / serialization could be a major problem • Test systems must be taken seriously, both by • DAQ providing a usable system • Subsystems using it • CALICE gave a nice example of how this might work

  8. Front-End - Continued • Emphasize uniformity of front-end design • Front-End talks indicate the same general nature of the data • Need to translate this into uniformity of DAQ interface • Will greatly simplify the DAQ • Need a mechanism to achieve this • Communicate, don’t impose • Arbitration process to resolve differences

  9. No Hardware Trigger • Made possible by • Low duty cycle • Advances in technology • Bandwidth • Memory • Software filtering will control the data volume • A back-end problem • Once decided upon, likely no turning back

  10. Management / Social Issues • Team composition will be a mixture of • Cross-disciplinarians in Electronics/DAQ/Physics • Domain-specific engineers • Team composition will evolve with time • Design / Building / Commissioning / Operation • Getting a good set of requirements and solid interfaces is important, but… • Flexibility must be part of the game plan

  11. Management / Social Issues • Need to recognize the global nature of this enterprise • Global Detector Network provides a framework to minimize the importance of site selection • Allows detector to be operated from a virtual control room • If want this, must take it seriously from the beginning

  12. Summary • Need to stay in touch with • Front-End Architectures • Give uniformity at least a fighting chance • Machine Parameters • Number of buckets and bucket spacing have a large impact on the DAQ • Machine / Detector Interface • What information from the detector is useful to the accelerator operators • How should it be exchanged

More Related