helix nebula science cloud - eoscpilot...helix nebula –the science cloud with grant agreement...
Post on 11-Jun-2020
7 Views
Preview:
TRANSCRIPT
Helix Nebula Science Cloud:Getting commercial suppliers andresearch organisations together
29 November 2017Bob Jones
CERNBob.Jones <at> cern.ch
Helix Nebula – The Science Cloud
Helix Nebula – The Science Cloud with Grant Agreement 687614 is a Pre-Commercial Procurement Action funded by H2020 Framework Programme
HNSciCloud Joint Pre-Commercial Procurement
Procurers: CERN, CNRS, DESY, EMBL-EBI, ESRF, IFAE, INFN, KIT, STFC, SURFSaraExperts: Trust-IT & EGI.eu
The group of procurers have committed• Procurement funds• Manpower for testing/evaluation• Use-cases with applications & data• In-house IT resources
Resulting services will be made available to end-users from many research communities
Co-funded via H2020 Grant Agreement 687614
Total procurement budget >5M€
12/1/2017
What is being procuredA hybrid cloud platform for the European research community
3
HNSciCloudPCP
Sou
rce:
Clo
ud
Co
mp
uti
ng
for
Go
vies
, DLT
So
luti
on
s,D
avid
Bla
nke
nh
orn
, V
an R
ista
uan
d C
aro
n B
eesl
ey
Combining services at the IaaS level to support science workflows
The R&D services to be developed are to be integrated withResources in data centres operated by the Buyers Group
GEANT network
HNSciCloud project phases
Preparation
•Analysis of requirements, current market offers and relevant standards
•Build stakeholder group
•Develop tender material
Implementation & Sharing
Jan’16 Dec’18
Each step is competitive - only contractors that successfully complete the previous step can bid in the next
4 Designs 3 Prototypes2 Pilots
Call-off Feb’17
Call-off Dec’17
Tender Jul’16
We are here
€ €€ €€€
Access to cloud service capacity
Bob Jones, CERN5
100C
10TB
2,000C
200TB
3,500C
350TB
3,500C
350TB
Jun
2017
Dec
2017
Feb
2018
5,000C
500TB
10,000C
1PB
Dec
2018
prototype x3 pilot x2
Functional
Testing
Scalability
Testing
End User Access
What is needed to ensure alignment of the EOSC with user’s needs and enable users to actively participate in service portfolio management?
Requirement gathering process engaging users/procurers and service providers
Users identify requirements and prioritise them
Service providers associate risk to each requirement
Collectively converge of what is achievable
All must be aware of the cost of the services
Ensures users focus on high priority requirements and are ready to verify they are satisfied
01/12/2017
7
Prioritization grid: value vs. risk
Must haves This could make
the true difference
Off-the-shelf Avoid if possible
Risk(Innovation potential from
technological/market point-of-view)
Value(Innovation potential
from end-user
perspective)
0 10 100
100
10
0
Innovative for end-user
+
Technological/market uncertainty
8
Planning Poker - Value
The cards:
Planning poker is a ‘best practice’ methodology to estimate e.g. added value, level of complexity, required
implementation effort, risk. The technique is based on domain expert evaluation and attaining consensus.
The interpretation when assessing value:
0-2 Could not care less, not needed.
3-5 OK, why not, could perhaps bring some value
13 Must have. Solution should have this requirement
20-40 WOW, home run, this would really shift the boundaries !
? Don’t know, no experience with this subject..
9
Planning Poker - Risk
The cards:
Planning poker is a ‘best practice’ methodology to estimate e.g. added value, level of complexity, required
implementation effort, risk. The technique is based on domain expert evaluation and attaining consensus.
The interpretation when assessing risk:
0-2 No worries, off-the-shelf standard solutions exist.
3-5 A frequent problem, certainly solvable.
13 Absolutely not a standard problem. A solution requires
important choices, thorough elaboration, and specific expert
effort. Success can be achieved with significant time and effort.
100 Impossible, requires breakthroughs of physical laws.
? Don’t know, no experience with this subject..
Results of risk/value assessment
Published report on the HNSciCloud Open Market Consultation
https://doi.org/10.5281/zenodo.254052
01/12/2017
Challenges
Innovative IaaS level services integrated with procurers in-house resources and public e-infrastructure as part of a hybrid cloud to support a range of scientific workloads
Compute and Storagesupport a range of virtual machine and container configurations including HPC working with datasets in the petabyte range
Network Connectivity and Federated Identity Managementprovide high-end network capacity via GEANT for the whole platform with common identity and access management
Service Payment Modelsexplore a range of purchasing options to determine those most appropriate for the scientific application workloads to be deployed
1112/1/2017
What criteria should services of the EOSC meet in order to be part of the service portfolio?
Technical – defined in the tenderRefer to relevant international stds whenever possible
Legal – defined in the tenderExclusion, selection & compliance criteria including legislation/jurisdiction under which services are operated
Signature of a common cloud services agreement by all contractors and procurers
Financial – M.E.A.T Most Economically Advantageous Tender
01/12/2017
How can certification be organized to ensure that services of the EOSC service portfolio meet a list of community-defined requirements?
Testsuite assembled by procurers to verify technical conformance
Monitoring of SLA criteria linked to payment
service availabilitysupport responsivenessperformance
etc.
01/12/2017
HNSciCloud test suite
14
CERN
• Edugain AAI
• CERN Benchmark Suite
• S3 Endpoint
• PerfSONAR Network
• VM Provisioning and Personalisation
• Grid Jobs in Docker
• Dockerized deployment of EOS+CERNBOX+SWAN
CNRS
• IAAS access via EduGain and local accounts
DESY
• HDF5_IO
EMBL
• CLOUD_BENCHMARK
• ELIXIR_AAI
• data_transparency
ESRF•FDMNESIFAE•monitoring-accounting•image creation and contextualizationINFN•BelleII_SIM•COMPASS/TGEANT_IO•INFN/GPFS_AFM•DynfarmKIT•static-batch-extension•dynamic-batch-extension•AFM-deployment-testSURFsara•BBMRi_data•dCache_io•kubertest
01/12/2017 – DI4R – Brussels
top related