Generalities
As for the school in Ankara we expect 40 students with courses in english. There should be 10 groups of 4 students each. The school to be 7 days long, with one day (a Sunday) free time. The location is identified as Rome, Italy. INFN Roma (in the Department of Physics , Sapienza University) confirmed to be the hosting institute. The school is planned for 9-16 February 2011. An INFN internal course will be run in parallel.
Poster
The poster of the school is
here
WEB page
The web page of the school is
here
Indico agenda
The draft agenda of the school is
here
Goals of the school
Introduce the basics of Trigger and Data acquisition
Expose the students to a maximum variety of topics
Show the TDAQ examples from small and large experiments
Make sure that the lab exercises are beneficial to the students
People
Advisory committee
Speranza Falciano (INFN Roma)
Markus Joos (CERN)
Konstantinos Kordas (Thessaloniki)
Andrew J. Lankford (UCI)
Livio Mapelli (CERN)
Robert McLaren (CERN)
Niko Neufeld (CERN)
Enrico Pasqualucci (INFN Roma)
Jorgen Petersen (CERN)
Emilio Petrolo (INFN Roma)
Gokhan Unel (UCI)
Status of local organization
Accommodation
Accommodation for students already pre-booked at Hotel Laurentia (20 double rooms). Other 20 single rooms are pre-booked for lecturers and tutors We need to define travel details by 15-20 november!
Fee
Though the school will be completely funded by sponsors, I propose to have a fee for the school and use it to provide students' accommodation, coffee breaks, lunch and social dinner.
Network
Video equipment
There is an INFN service for video. The service has been required for the school.
Teachers and tutors - Travel details
Notes:
- People travelling on the ACEOLE budget must contact Evelyne Dho before making arrangements
Draft schedule
Note: Initials after a lecture indicate that a lecturer has been identified
Preliminary homework
With respect to the first edition, it is proposed to distribute preparatory material well before the school. This material should include some linux distribution (possibly a live CD like knoppix), documentation about basics of C programming and exercises. This material should be given to the student as homework for the beginning of the school and should include the advanced features as in lecture 3 of previous edition. The morning of the first day can be used to discuss the exercises and their solutions, present C to C++ improvements, some useful trick (like in Ankara) and possibily some excercise. This would save approximately half a day and would allow to move some lectures from day 1 to day 0, freeing time for two more labs. Since in our experience homeworks at the end of the course are sent back to us only on voluntary basis (about 50% last year), we could use the ones at the beginning to give certificates - still assigning homeworks at the end.
Proposed schedule:
The proposed agenda is in
this Indico page
Review of the exercises
The assessments of the students did not indicate that any one of the 11 exercises was of bad quality (too easy, too difficult, not interesting). This picture, however should be completed from the perspective of the tutors. Therefore you are requested to add your comments to the table below. We need this information to decide which of the Ankara exercises we drop in favour of new (e.g. more H/W oriented) exercises.
Definition of terms: "Efficiency":
Was the time of the exercise used efficiently? Can the exercise be improved by providing additional material to the students? Was the H/W adequate "Content":
Did the students learn something that is relevant to general DAQ systems of was the content too exotic? Is the exercise actually helpful for the students or could they acquire the same knowledge without H/W (e.g. by reading books, writing S/W as a home work) "Comments":
Anything else
Exercise 1
Tutor: Adriana
| Category |
Assessment |
| Efficiency |
I find that the lab is very interesting concerning the topic covered and the structure of the steps to perform. It gives to the students the possibility to understand how to perform different tasks using the VME bus from the easiest to the more complex ones. |
| Content |
The different steps to perform are well organized. If a good knowledge of programming is required as a requirement to access the school, lab 11 can be aggregate to lab 1. |
| Comments |
I noticed that some students were totally unaware of what VME is (they didn't listen to Markus's lecture) and this is unavoidable. Therefore I would foresee a summary of the lecture at the beginning of the lab maybe with some questions that permit to know what they understood. Furthermore, students had different levels of knowledge about C++ programming. Between one group and another the duration of the lab was completely different. For example, one group completed the lab in less than 40 minutes and another didn't succeed in performing more then 3 steps in 2 hours. This is due to the fact that, in the first case, the student were very familiar to C++ programming and knew how to look for the correct API in the booklet provided; in the second case, I had to explain them what a #define or an #include is useful to, etc. |
Tutor: Luca
| Category |
Assessment |
| Efficiency |
Leaving the coding part up to students increase the overall time elapsed, depending on group's skill, but it helps a lot in understanding. I suggest to expand the documentations with a schema of the HW configuration, and a brief reminder of the main concepts of VME bus (single cycle, block transfer, etc.). This could help students in understanding what they are doing, in Ankara I used lectures slides but a dedicated doc will work better. Also the code can be reorganized a bit in order to group variable definitions in the related sections, to avoid jumping up and down. This can save some time, but I do not think it is feasible to reduce it to 1h. |
| Content |
This is a useful hands on session not easy to reproduce at home without proper HW/SW configuration. |
| Comments |
Docs can be improved. See comments above. |
Exercise 2
Tutor: Yi Ling
| Category |
Assessment |
| Efficiency |
The timing of 2h is generally ok. Some groups are faster while some are slower, but the versatility of the exercise allows the tutor to extend on the subjects - for example by asking more advanced questions to groups which are faster in carrying out the tasks, like the working of the CFD and the use of the Timer. |
| Content |
The content was good. I would suggest that we include a diagram of the working of the CFD, because a lot of the students missed the explanation of this during the lecture. There were also students which have no experience in operating an oscilloscope and the tutor has to take care to guide them to do this. |
| Comments |
I find that an explanation of the exercise was very important for the students. It is better to tell them the purpose of the exercise and the goal that they are expected to achieve before the students start playing with the cables. It is also important to introduce them to each of the NIM modules and their function, so that they know better which one to use during the exercise. |
Tutor: Francesca
Tutor: Luca
| Category |
Assessment |
| Efficiency |
2h hours seems ok to me. Some groups were faster, and some others get lost with LIMO cables at some point, but it is a formative experience. For faster groups I remember Wainer and Francesca designed some extra exercise, may be it is a good idea to put them on paper in the doc. Another missing point in the doc was the internal schema of the constant fraction discriminator, it helps a lot in understanding. |
| Content |
Students were asked to think about several different concepts, generally all group's member were actively involved and the average interest was very high. |
| Comments |
Working oscilloscopes improve tutors life. |
Tutor: Wainer
Exercise 3
Tutor: Markus
| Category |
Assessment |
| Efficiency |
In general the time was used in a constructive way. Some groups lost up to 30 minutes on the computation of the probability for a random coincidence. |
| Content |
In general very relevant. Many of the students have never been near NIM H/W or scintillators. This exercise is also very interesting because there is a real detector |
| Comments |
Some minor problems due the lack of LEMO Y-connectors and 50 Ohm terminators. |
Tutor: Erkcan
| Category |
Assessment |
| Efficiency |
The timing was perfect except for one group that lost time while trying to compute the probability of random coincidence. |
| Content |
There was no group (even among those with students who had previous experience with scintillators) who did not find the exercise engaging. |
| Comments |
Discussion of the random coincidence computation, if the students cannot do it in ~10 minutes, can be (was) deferred to the tea break afterwards. Discussion around a white board was useful. |
Tutor: Yi Ling
| Category |
Assessment |
| Efficiency |
The timing was very well organized (thanks a lot to the rehearsal that we had before we went to Ankara). There was one group which has not done exercise 2 before they do this exercise and thus the exercise was harder for them and they used a significantly longer time to complete the exercise. I had to help them a lot during the exercise. This problem is best to be avoided during the next school. |
| Content |
The content is very interesting. There were groups which had the potential to finish all the steps in around an hour time, so maybe the tutor should be prepared for this and prepare some extra questions or tasks for these students to complete. |
| Comments |
I find it important that we explain to the student the purpose of each step after they have completed it. Because it is relatively easy to do as the instructions say, but without understanding why we do it, it is very easy to lose track of what we are doing. I agree with Erkcan and Markus regarding the issue of the random coincidence computation - the tutor has to help them in this question. I would suggest that in the next school, we include also a diagram in the exercise to help the student understand this. |
Exercise 4
Tutor: Markus
| Category |
Assessment |
| Efficiency |
There is not too much hands-on activity for the students as it is better if the tutor drives the exercise but the tutor can demonstrate lots of effect which the students can try to anser. |
| Content |
This exercise uses the foundations laid in exercises 1, 2 and 3 to do a small experiment. The students have the opportunity to discuss the output of the DAQ system (charge and time histograms) with respect to what happens in the detector. in addition they get a feeling for the issues of scalable modular DAQ systems |
| Comments |
-- |
Tutor: Erkcan
| Category |
Assessment |
| Efficiency |
When done right after exercise 3, this exercise runs very smoothly. If the student do exercise 3 and 4 with some time/days in between, the efficiency goes down a little. |
| Content |
While not as engaging as exercise 3, as it has the tutor more actively driving the exercise, it is very useful to show how DAQ systems would use computers and will scale. This is an excellent conclusion to the series of exercises 1-4: The students find out that everything they have done is really relevant to the actual large experiments like ATLAS. |
| Comments |
When done right after exercise 3, the students gain some speed, so the tutor can ask more things and thus allow them some more time to explore by themselves. |
Exercise 5
Tutor: Dominique
| Category |
Assessment |
| Efficiency |
All students were interested. In almost all groups the 2 hours were used. And for these who were fast, I had a extra excercise to do |
| Content |
| Comments |
It missed a link with a real physic application |
Tutor: Sebastian
Exercise 6
Tutor: Hannes
| Category |
Assessment |
| Efficiency |
Most groups finished in the foreseen time of 2 hours or a few minutes of extra time. For some groups, in order to save time, one of the sub-exercises was replaced by doing an estimate of the result, instead. One group finished 20 minutes early. |
| Content |
Covered: PCI in general; data acquisition via PCI; SLINK-64; timing of PCI accesses; overhead of software; DMA; low-level programming: decoding bit fields etc. For the students to judge how much they learned. |
| Comments |
The level of knowledge of simple C++ was non-uniform among the students. In most groups this was not a problem since the more proficient students took over the programming part. Most of the concepts presented in the exercise could be followed by all students nevertheless. |
Tutor: Dominique
Tutor: Sebastian
Exercise 7
Tutor: Ahmet
Exercise 8
Tutor: V. Ugur Guney
Exercise 9
Tutor: Dan
| Category |
Assessment |
| Efficiency |
For the first section (network configuration and monitoring) the allocated time of 1 hour was enough only for some groups that already had some basic knowledge in networking and scripting/programming. For most of the groups however, the time was enough only for tutor presentation and not for hands-on experience, as it was expected when planned. The lab accomplished the goal of giving an overview over network configuration, monitoring and data taking, but the time should be extended (or the information presented shrinked) in order to benefit also from the "do-it-yourself" part. |
| Content |
Network configuration and management on Linux desktop and on an HP switch using CLI. Monitoring using traffic dump and port mirroring, and snmp. Protocol study and implementation of an Event-builder software application in Python, filling blank lines of codes. The HW is required for testing and checking if the code works correctly, although some parts can be used with hardware different from the one found in setup. |
| Comments |
Although some of the participants are not necesarily directly interested in the IT infrastructure of a DAQ system (networking or system administration), the networking lab was essential in understanding all the aspects of a DAQ system, and what's behind any data communication. The lab was appreciated by the participants, some of them being interested in applying what they learnt (about network monitoring) even at home or university. |
Tutor: Jean-Christophe
| Category |
Assessment |
| Efficiency |
The time was fully used because the content is big. Two hours were too tight and did not offer enough margin for the slower groups. Some time to reply to questions was always used, and the last programming part was sometimes performed by myself, giving explanations, instead of letting them analyse the problem. |
| Content |
Students learn: Network configuration and management on Linux desktop and on an HP switch using CLI. Monitoring using traffic dump and port mirroring, and snmp. Protocol study and implementation of an Event-builder software application in Python, filling blank lines of codes. The HW is required for testing and checking if the code works correctly. |
| Comments |
It's possible to reduce the number of computer, from 5 to 3. |
Exercise 10 (a & b)
Tutor: Dan (Lab10b)
| Category |
Assessment |
| Efficiency |
Some of the groups (usually the ones where most of the students had no knowledge about a scripting/programming language) needed some extra 10-15 minutes to accomplish the task(s). As a general rule the allocated time can be succesfully managed by a tutor after some prior rehearsals. According to group level, the balance between presentation and "do-it-yourself" time ca be adjusted on the fly by the tutor so the group can get the best out of it. |
| Content |
Lab10b introduces some theoretical and practical aspects about microcontrollers, simple communication protocols and user interfaces, using scratchboard and python. The theoretical base was used to explain how python (or C) can be used to access all the sensor information on the picoboard, and how to display all the data in an user friendly way. Resistance calibration and hall efect measurement are two examples the students have to fullfill during the lab. |
| Comments |
Most parts of the lab can be reproduced at home by anyone having a picoboard. The hall effect measurement can be replace by light intensity measurement if one can't find an appropiate hall effect sensor and voltage amplifier. |
Tutor: Ozgur (Lab10a)
| Category |
Assessment |
| Efficiency |
Students had enough time to try and fail, think, design/re-design, implement the design and test the implementation in an hour, with a peak-to-peak variation of plus or minus 10 minutes over all the different groups. An experienced tutor can easily control the amount of time it takes due to the flexible content and reference designs provided in case things get un-necessarily hard. 3 rehearsals before the school are needed to converge to such experience. |
| Content |
Lab10a introduces the working environment, namely Scratch, PicoBoard and Finite-State Machines (FSM) pragmatically without diving into FSM types & the theory behind and carry the students to a level where they can practically design an FSM in Qfsm tool and efficiently implement their first FSM in Scratch. A priori skill level required for this laboratory is moderate. Concepts were digested by the students without major resistance. |
| Comments |
An hour was just fine for Lab10a. |
Exercise 11
Tutor: Markus
| Category |
Assessment |
| Efficiency |
This exercise has to be driven by an experienced tutor. The students do not directly touch the H/W but there are lots of questions that the tutor can ask to them. |
| Content |
The level is probably a bit on the difficult side. Even though the students had heard the VMEbus lecture and done exercise 1 they found it difficult to explain what the bus analyzer showed. Nevertheless looking at a data transfer link at the lowest level is an interesting experience and shows the potential and limitations of bus / logic analyzers in general |
| Comments |
This exercise should stay in a 1h slot. The full 2h would be too stressful |
Tutor: Adriana
Tutor: Yi Ling
| Category |
Assessment |
| Efficiency |
Many students were totally lost during this exercise - they were either too tired from the day (this exercise was carried out during the evening) or had too little knowledge of VME bus from only one lecture. |
| Content |
The content is leaning towards the difficult side. It was interesting for the student to see what a bus analyzer can do, but good explanation and perhaps the print-out of the bus cycle might help in understanding. |
| Comments |
Because this exercise is allocated at the end of the day, the general impression I got from the student was they were already overwhelmed from too much information during the whole day. Not specifically concerning this exercise, I think this is a problem that we need to give careful consideration - the problem with information overflow. In my opinion, too many labs with too many interesting topics might not be as interesting nor as useful for the student. We have to strike a balance between the amount of lab exercise and the human ability to absorb information. |
Tutor: Luca
| Category |
Assessment |
| Efficiency |
Subject not easy and the exercise list is quite dense, It has been a bit difficult to keep students awake for the whole 2h. |
| Content |
|
| Comments |
I agree with Markus on the 1h reduction. |
The lab exercise proposals
If you want to propose a Lab exercise please add it to the list below using this
template
Below are the exercises of the school in Ankara. This is just a guideline. Nothing has been decided so far/
Lab time table (aka "Sudoku")
NOTE: The VMEbus analyzer exercise (exercise 11 of Ankara) will be reduced in volume and merged into exercise 1
Notes: G: = ID of the student group T: = Initials of the tutor
Material
Some of the material from Ankata (mostly the scintillatos) is currently stored in 4-R-5 and will be moved to the permanent lab once a room has been found. A detailed list of the available material will be added once the exercises have been selected.
Meetings/decisions/actions
Actions 22/06/ 2010
| What |
Who |
Status |
| Set-up a e-group (mailing list) with the people that have confirmed their participation in ISOTDAQ2011 |
Enrico |
Done |
| Review of old lab from tutors |
all lab tutors |
Done (partially) |
| Enquire about video equipment in Rome |
Enrico |
Done, waiting for answer |
| Ask for feedback for exercise from Turkey |
Gokhan |
|
| Organize the permanent lab |
Markus |
Ongoing (see here) |
--
YiLingHwong - 23-Jun-2010
Meeting 02/06/2010
- First proposal for the new web site
- News on funding from INFN
- Discussion of new proposed labs
- Involvement of ACEOLE people in labs
- Twiki minutes
- Possible accomodation and financial matters
- Planning
--
MarkusJoos - 18 Mar 2010
Meeting 01/09/2010
- Meeting periodicity
- Poster proposal
- Status of the labs
- Lecture organization
- Shipping
Meeting 01/09/2010
- Local organization
- Final poster version
- Web site
- Lab proposal
- Funding and fees
Meeting 17/11/2010
- Lab status
- Lab rehearsals
- Instructor to lab allocation
- Registered participants (18)
Milestones
- End of August 2010 : Labs to be defined
- September 2010 : Labs set up and testing
- mid-September 2010 : website ready, poster ready
- 1st October 2010 : Opening of registration
- October & November 2010 : Lab training and rehearsal
- January 2011 : Packing and shipping of equipments
--
YiLingHwong - 23-Jun-2010
Sandbox/SandboxArchive Web>TWikiUsers>MarkusJoos>MarkusJoosSandbox>DaqSchool2011 (2020-08-20, TWikiAdminUser)
EditAttachPDF
Sandbox/SandboxArchive Web
Create New Topic
Index
Search
Changes
Notifications
Statistics
Preferences
Copyright &© 2008-2023 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.