2016 WLCG Collaboration Workshop
ISCTE-IUL - University Institute of Lisbon
The 2016 WLCG Workshop in Lisbon will be devoted to in-depth discussions of the current issues and how to solve them in the medium term, and to have a common brainstorming on the long term future of WLCG over the coming 10 years.
Immediately after the WLCG workshop, a DPHEP workshop will take place and cross-participation is encouraged (agenda).
-
-
10:40
Coffee break
-
Medium Term Evolution: Security and Trust Model¶Conveners: Ian Neilson (STFC RAL (GB)), Ian Peter Collier (STFC - Rutherford Appleton Lab. (GB))
-
5
Overview Medium-Term Issues in Security and Trust Model¶High-level overview of and speaker's vision on the issues and developments relevant to the medium-term (three year horizon) evolution of the WLCG Security and Trust Model.Speaker: Ian Peter Collier (STFC - Rutherford Appleton Lab. (GB))
-
6
Lightning: Trust Fabric Evolution¶Speaker: David Groep (Nikhef National institute for subatomic physics (NL))
-
7
Lightning: New paradigms for academic computing security¶Speaker: Romain Wartel (CERN)
- 8
-
9
Lightning: Trust model between experiments and sites¶Speaker: Alessandro Di Girolamo (CERN)
-
10
Discussion¶Speakers: Ian Neilson (STFC RAL (GB)), Ian Peter Collier (STFC - Rutherford Appleton Lab. (GB))
-
5
-
12:30
Lunch
-
16:00
Coffee break
-
Medium Term Evolution: Information systems, accounting and benchmarking.¶Convener: Alessandra Forti (University of Manchester (GB))
-
-
25
Introduction & goals of the day¶Focus on how to work towards a model for computing for the HL-LHC era. What are the areas where we need to invest R&D, how do we being to understand the likely costs?Speaker: Ian Bird (CERN)
-
26
Summary of top concerns of the experiments¶Speakers: Andrej Filipcic (Jozef Stefan Institute (SI)), Daniele Bonacorsi (University of Bologna), Eric Christian Lancon (CEA/IRFU,Centre d'etude de Saclay Gif-sur-Yvette (FR)), Ian Bird (CERN), Predrag Buncic (CERN), Stefan Roiser (CERN)
-
10:30
Coffee
-
27
Infrastructure models¶
- How to reduce the overall cost of the infrastructure, e.g. by:
- Reducing the amount of scattered storage
- How to procure and incorporate large scale compute resources at scale
- What is the networking model - OPN–style private?
- The use of other resources: HPC, volunteer, etc. What is the model for distributed sites (Tier 2)?
- The idea of setting up an early prototype of a data/cloud infrastructure to test ideas (a la “Zephyr” proposal) - suggestion already supported by several Tier 1s
- What about specialist sites/resources (HPC, GPU etc) for specific workflows
- Security considerations, especially access control and how that affects performance and cost
Speaker: Simone Campana (CERN) - How to reduce the overall cost of the infrastructure, e.g. by:
-
12:30
Lunch
-
28
Experiment workflows & data management tools and infrastructure¶
- Can we think about common toolsets for all experiments, given 10 years until HL-LHC?
- Input on data management convergence from previous day
- What is realistic in terms of common infrastructure and services in order to achieve commonality?
- More commonality is an expectation of funding agencies.
- Political limitations? e.g. specific project funding?
- Many sites (esp Tier 1s) have to support other sciences on the same infrastructure
Speaker: Daniele Bonacorsi (University of Bologna) -
29
Technology Evolution¶
- Should we set up a technology tracking activity (as Bernd has been doing), but perhaps now realised together with the concurrency forum and the tech lab activity – to provide a real testing and benchmarking environment (openlab here too).
- Could be a distributed collaborative effort - in common with HSF/concurrency forum
Speaker: Bernd Panzer-Steindel (CERN) -
16:00
Coffee
-
30
Undesrtanding and modelling of the distributed infrastructure & computing models¶
- How well do we understand our current workflows, their behaviour and resource needs?
- with respect to storage, remote access, networks, CPU, memory
- How well do we understand the behaviour of our current infrastructure?
- What can we do to improve this understanding in an experiment independent way?
- how independent can this be?
- What has been done already in experiments?
- What would be desirable? Ability to model ideas of infrastructure to understand performance, costs, etc.
- What is potentially common across experiments? What is specific?
- Can we derive a cost model for the infrastructure to explain the full costs of computing and the relative costs of each component?
Speaker: Markus Schulz (CERN) - How well do we understand our current workflows, their behaviour and resource needs?
-
Workshop Dinner¶ Restaurant Sacramento
Restaurant Sacramento
Calçada Sacramento 44 1200-394 Lisboa, Portugal
-
25
-
-
31
Software topics¶
- What is actually useful to do in common?
- Tools or support for common automated and intelligent build/test/validation services?
- More ? Common libraries etc? Role of HSF
- What is usefully shareable?
- Detailed software performance analysis and tools for analysis
- All this should be in the context of the HSF. Some activities towards common (lower-level) build tools was started already. Organise a dedicated HSF workshop on this?-
- Do experiments actually want to commit to the HSF for common software interests?
- Given today's (and expected evolution of) processor and machine architecture, is today's data architecture (i.e. ROOT) appropriate - do we need to re-think completely?
Speaker: Benedikt Hegner (CERN)Benedikt Hegner showed 6 slides.
Frank Wuerthwein: Go where the money is. KNL will be relevant in the US. For CMS Reconstruction will be the most cpu intensive data processing activity, and of that Tracking is the biggest part
Marco Cattaneo: C++14 is coming, we all need to become familiar and develop expertise, need for common support
Michel Jouvin : HSF TN’s were started in order to address this, so that expertise can be shared
Mattias Wadenstein : easier to deal with small projects; validation - use state of the art compilers, haven’t done anything yet on behalf of the entire community
Fons Rademakers :
- Need to keep tracking new (software) technologies such as Apple/Swift
- Fast memories will have an impact - build a demonstrator to study them
- ROOT format is used for storage model but this may be outdated - need to look at new technologies such as Kudu, C++ Object Storage
- Interesting results coming out of taking a fresh look - Intel competition on Xenon Phi - student managed to increase the speed of neurobiology code by a factor of x320
- Make HSF a legal entity (cf Apache Foundation), handle Copyright in a better way, helps to get funding from industry, profit from work done in Machine Learning
Benedikt Hegner :
- several comments to make..
- legal entity : this is a chicken/egg situation, HSF started small by a few volunteers, need to reach a critical mass and demonstrate feasibility of initiative, now need more participation from the community
- Concurrency Forum fosters 1st round of demonstrators, now need to start a second round e.g. fast memories
- Different languages have been looked at e.g. Go, but no real outcome, take a look at Julia, ….
Ian Bird :
- profiling tools were mentioned, it would be good to have measures of what can be achieved.
- savings of 10% here and there are crucial since they have a big impact on costs. If they exist they should be used
- I’d like to see some metrics of what can be done
Danilo Piparo : we need to build a matrix of costs, for example see what we are prepared to give up; one example is precision, floating point give up IEEE performance
Liz Sexton-Kennedy : we have one person to do it for the experiment, this is a fundable model, have developer and user work closely together to achieve results
Latchezar Betev : for simulation code is focused in one place, therefore optimisation can be done for whole community, surprised more emphasis has not been given to this in the discussion
Marco Cattaneo :
- careers are important, need to raise the profile of developers
- not a part-time job for a student, engineers are important
- middleware is not a job for a physicist, get this on the radar of management
Stefan Roiser : try to involve SE institutes - those close to our physics depts. and collaborate with computer scientists; in LHCb we need to change our framework and in particular the data model in order to exploit vectorisation for example, currently we are not using SIMD
Ian Bird: there are ways of giving recognition (see mail thread that was circulating earlier this year; idea is to look at code that is accessed/used in git repositories such that articles can reference what has been used in a piece of work, this allows developers to build their CVs
Danilo Piparo:
- I agree its important to work with computer scientists, however the biggest role is still played by physicists;
- have to understand the impact of any software change on physics results
Benedikt Hegner : yes we need physicists for validating impact on physics results, but this shouldn’t prevent us for being outward-looking and learning from other fields, there is a place for non-physicists and this will help us to avoid re-inventing the wheel
Miron Livny : From experience, solutions owned by experiment(s) don’t become common, only solutions that are owned outside the experiments have a good chance to become common solutions.
Liz Sexton-Kennedy : Disagrees. Frontier and xrootd were solutions that grew out of a single experiment (CDF, BaBar). The solution has to be proven to work at scale for somebody before it is accepted by another. Common solutions need to be open source because it increases the trust between developers from different organizations.
Jeff :
- Here in Netherlands we benefit from participation of non-physicists
- However experience shows it is hard to rethink an algorithm if you are not a physicist
- Try community building i.e. one good place where people can talk to each other
Oxana : we need to lower the threashold for invoving non-physicists who would like to help
Frank Wuerthwein : If I look in CMS, the biggest contribution to tracking came from 5 people, 3 of whom have permanent jobs; I’d suggest that physicists have a harder time becoming permanent
Benedikt Hegner :
- HSF is everybody, you are all invited to contribute, need to assemble people who are keen to do technology tracking
- In reply to question from John, the HSF Workshop will take place at LAL, Orsay, Paris and the current favoured time is first week of May, time hopefully to be fixed early next week
note taker : John Harvey
-
10:00
Coffee
-
31