NorduGrid 2016

31 May 2016, 08:30 → 3 Jun 2016, 17:00 Europe/Bratislava

Faculty of Science, Lecture hall M5 (Pavol Jozef Safarik University in Kosice)

Farid Ould-Saada (UiO) , Jozef Černák (Department of Nuclear and Subnuclear Physics, Institute of Physics, Faculty of Science, Pavol Jozef Safarik University in Kosice, Kosice, Slovak republic)

The way forward in distributed computing: evolution or revolution?
The annual NorduGrid conference will be held in Kosice, Slovakia, on June 2-3, 2016. The conference will be preceded by the technical workshop on May 31 - June 1.
	This year's theme "The way forward in distributed computing: evolution or revolution?" addresses the challenge faced by many scientific communities: how to survive the data tsunami in the shade of the clouds? The conference will cover different aspects and directions of future distributed computing systems, including: Virtualised infrastructures for data and computng Cluster resource management Exploiting supercomputers and distributed infrastructures New approaches to networked computing
On 4 June 2016 Organizers offer optional trips for conference participants and accompanying persons. Participants are kindly asked to register and indicate whether they will attend both the conference and the workshop, or either of the events, as well as the social events (the dinner and the trips). The conference is free of charge, however, the common dinner and trips are on own expense. The organisers offer free lunches in the University cafeteria; please sign up for the lunches using this form.

Tuesday, 31 May

08:30 → 09:00 Registration

09:00 → 12:00 Technical meeting session 1

09:00 Tuesday morning: Reports and Presentations

- Balazs: ARC 101 - David: A-REX Cache related services & features: the overview - Mattias: Status report from the binarians: code base, distributions, dependencies - Jon: Release management in action - Martin: ARC SDK overview - Christian: Backend status report - Andrii: Torque jobs scanning back-end experimental rewrite - Aleksandr: Proxies & delegations in ARC: a brief overview and recent changes - David: The hidden truth about how ARC CE is used in Atlas Distributed Computing - Aleksandr: Internal restructuring of A-REX for improved scalability - Balazs:Towards a new ARC.CONF configuration: status update, next steps - Oleksandr Boretsky: Integrating container-based virtualization technologies into ARC-powered grid infrastructure - Iana Kokriashkina: Grid-node software as a tool for interaction with Grid network (could be in the afternoon) - Doug Jacobsen: Ph.D,NERSC Computer Systems Engineer (afternoon) Shifter, an implementation of Linux Environment Containers for HPC

Speaker: Balazs Konya (Lunds universitet)

notes Tech_meeting_Kosice2016_notes.txt

Slides

• 2016NGInfosys_Florido.pdf
• aCT_David.pptx
• ARC101, Balazs Konya
• ARCBackends_Christian.odp
• ARC_cache_David.pptx
• ARC.CONF_Balazs
• ARC__Release_Management_in_Action_Jon.pdf
• ARC-SDK-overview_Martin.pdf
• a-rex_Aleksandr.pdf
• BinarianReport_Mattias.pdf
• delegations_Aleksander.pdf
• Docker Containers integration into ARC
• Grid-node-software_Iana.pptx
• PBS scanning rewrite, Andrii
• Shifter-NorduGrid2016.pdf

12:00 → 13:00 Lunch

13:00 → 18:20 Technical meeting session 2

13:00 Tuesday afternoon: Performance & Stability

- Protection against hanging/timeout external processes in A-REX including the infosys special use-case - Performance data: discussion - AREX behavior under heavy load

16:00 Coffee break

16:30 Shifter, an implementation of Linux Environment Containers for HPC

Speaker: Doug Jacobsen (NERSC Computer Systems Engineer)

Slides Shifter-NorduGrid2016.pdf

17:00 Tuesday late afternoon: dreaming...

- What if we could do real research & development...

Wednesday, 1 June

09:00 → 12:00 Technical meeting session 3

09:00 Wednesday morning: Planning for bigger changes

- 10 years of RTEs: time for replacement? - When can we retire the GIISes? - Resource discovery revisited - Config parsers: discussion

12:00 → 13:00 Lunch

13:00 → 18:00 Technical meeting session 4

13:00 Wednesday afternoon: mixed topics

- Planning for the next major release - Simple ARC CE configurations - VOs & ARC: short status update - Top priority bugs: discussion - AREX and Condor

18:00 → 20:00 Board Meeting

Convener: Prof. Farid Ould-Saada

Thursday, 2 June

08:30 → 09:00 Registration

09:00 → 12:30 Plenary Session 1

09:00 Welcome by the Rector and the Dean of Faculty

Speaker: Dr Jozef Černák (UPJS Kosice)

09:20 Keynote: Science Services and Science Platforms: Using the Cloud to Democratize Discovery

Speaker: Ian Foster

Slides Discovery_Cloud_NorduGrid_June_2016.pdf

10:25 NorduGrid Collaboration

Speaker: Farid Ould-Saada

Slides WayForwardInDistributedComputing2016-_short.pptx

10:40 Coffee

11:10 State of the ARC

Speaker: Balazs Konya (Lunds universitet)

Slides ng2016_balazs_konya.pptx

11:30 ATLAS@Home: ARC-enabled volunteer computing

Speaker: David Cameron (University of Oslo)

Slides atlasathome_David_Cameron.pptx

11:50 Distributed Computing Evolution of High-Performance Computing

Speaker: Andrej Filipcic

Slides nghpc-Andrej_Filipcic.pdf

12:10 Distributed Computing as a Service with ARC

Speaker: Barbara Krasovec

Slides distributed-nordugrid-barbara_k.pdf

12:30 → 13:30 Lunch

13:30 → 20:00 Plenary Session 2

13:30 Keynote: Big Data and Big Compute in Radio Astronomy

Modern radio telescopes are large distributed sensor networks that produce huge volumes of data. Flexibility requirements force us to perform the data processing of these instruments in software. I will sketch how the LOFAR and SKA telescopes are prime examples of this trend. LOFAR currently is the largest radio telescope in the world, and SKA has exascale requirements on both processing and the network. I will focus on two interesting challenges that we face while moving to exascale instruments. First, the data volumes have become so large that they simply cannot be stored any more. This means that we have to switch from off-line processing to a real-time system. Second, changes in modern computer architecture render our current codes and algorithms inefficient. Together, these changes are disruptive, and require us to completely change our pipelines, perform completely different optimizations and redesign our algorithms. An e-Science approach that combines computer science research with domain knowledge is essential to achieve the required breakthroughs.

Speaker: Rob Van Nieuwpoort

Slides NorduGrid_slides_Rob_van_Nieuwpoort.pdf

14:30 Software and computing challenges at the high luminosity LHC

Speaker: Borut Kersevan

Slides kersevan_Kosice_Future_310516.pdf

15:10 Molecular dynamics simulations in the MolDynGrid Virtual Laboratory by means of “ARC between Grid and Cloud”

Speaker: Andrii Salnikov

Slides Nordugrid2016_MolDynGrid_Technologies.pdf

15:25 Simulation of Nonlinear Dynamics in Large Neuronal Networks Using ARC Middleware

Speaker: Oleksandr Sudakov

Slides presenta_neuro_grid.pdf

15:40 Coffee

16:10 cPouta - HPC Cloud @ CSC

Speaker: Kalle Happonen

Slides Openstack_Pouta_Kalle_Happonen.pdf

16:50 ARC as cloud front-end - Swiss experience

Speaker: Sigve Haug

Slides NGC16-ARC-Cloud-Sigve_Haug.pdf

17:10 Machine Learning Technologies to predict the ATLAS Production System Behaviour

Speaker: Maxim Gubin (Tomsk (TPU))

Slides ProdSys_ML_Maxim_Gubin.pptx

Friday, 3 June

09:00 → 12:30 Plenary Session 3

09:00 Keynote: Accelerated Computing using OpenPOWER-based Heterogeneous Computing Systems

Speaker: Christoph Hagleitner

10:00 ARC as CRAY front-end - Swiss experience

Speaker: Gianfranco Sciacca (University of Bern)

Slides Nordugrid2016-ARC-Cray_Gianfranco_Sciacca.pdf

10:20 Coffee

10:50 Towards understanding today’s and tomorrow's scheduling challenges in HPC systems

Speaker: Gonzalo Rodrigo

Slides Gonzalo_Rodrigo_TowardsUnderstanding.pdf

11:20 Generative software development method of ARC mobile clients

Speaker: Vaidas Giedrimas

11:40 Distributed computing at P.J. Safarik University

The aim of the talk is to present a brief view on the history of the distributed computing at University and partially in Slovakia, and application of the distributed computing approach that enabled us to deliver interesting scientific results. At the end I will present thoughts about a need to increase intelligence of computational systems.

Speaker: Dr Jozef Černák (UPJS)

Slides Distributed_computing_Jozef.pptx

12:00 Summary & Closing

Slides ng2016_closing.pptx

12:30 → 13:30 Lunch

14:00 → 15:30 Football