EGI-Engage:AMB-2015-07-22
Revision as of 17:56, 21 July 2015 by Verlato (talk | contribs) (→Task JRA2.4 Accelerated Computing)
Activity Reports
WP1 (NA1) (Yannick Legre)
Milestones & Deliverables
Task NA1.1 Administrative and Financial Management
Task NA1.2 Technical Management
Task NA1.3 Quality and Risk Management
WP2 (NA2) (Sergio Andreozzi)
Milestones & Deliverables
Task NA2.1 Communication and Dissemination
(since the last meeting)
- Copyedit and layout of the EGI Strategy 2015-2020 publication
- Scoping definitions for upcoming officer opening
- Communications metrics for Q1 (Engage)
- Responding to media query! (Nature News)
- First steps towards all e-Infras CF2016: setting up steering committee
- CF2015: Feedback on guides, setting up plan for review, work on colocations, programme sketch
- CF2015: preparing for abstract review
Task NA2.2 Strategy, Business Development and Exploitation
Task NA2.3 SME/Industry Engagement and Big Data Value Chain
WP3 (JRA1) (Diego Scardaci)
Milestones & Deliverables
Task JRA1.1 Authentication and Authorisation Infrastructure
Task JRA1.2 Service Registry and Marketplace
Task JRA1.3 Accounting
Task JRA1.4 Operations Tools
Task JRA1.5 Resource Allocation – e-GRANT
WP4 (JRA2) (Tiziana Ferrari)
Milestones & Deliverables
Task JRA2.1 Federated Open Data
Gathering of requirements for open data platform by integrating expectations from various communities. Preparation of the milestone report due to by the end August.
Task JRA2.2 Federated Cloud
Task JRA2.3 e-Infrastructures Integration
Task JRA2.4 Accelerated Computing
- Attributes "GPUNumber" e "GPUMode" have been added to the command BLAH_JOB_SUBMIT. This required modifications blah_common_submit_functions.sh and server.c. The first implementation of the two new attributes have been successfully tested at CIRMMP cluster with the PBS/Torque LRMS. This required modifications to pbs_submit.sh.
- The next planned steps are:
- installing slurm LRMS on CIRMMP cluster
- add new JDL attributes in CREAM core
- After a review of the available technologies, GPGPU virtualisation in KVM/QEMU has been analysed and performance testing of passthrough technology have been carried out with the following results:
HW configuration: IBM dx360 M4 server with two NVIDIA Tesla K20 accelerators. Ubuntu 14.04.2 LTS with KVM/QEMU, PCI passthrough virtualization of GPU cards.
Tested application: NAMD molecular dynamics simulation (CUDA version), STMV test example (http://www.ks.uiuc.edu/Research/namd/).
Performance results: Tested application runs 2-3% slower in virtual machine compared to direct run on tested server. If hyperthreading is enabled on compute server, vCPUs have to be pinned to real cores so that whole cores will be dedicated to one VM. To avoid potential performance problems, hyperthreading should be switched off.
- Recently a cloud site with GPGPU support has been set up with the following layout:
Configuration: master node, 2 worker nodes (IBM dx360 M4 servers, see above) Base OS: Ubuntu 14.04.2 LTS Hypervisor: KVM Middleware: Openstack Kilo
- The next planned steps are:
- Testing the Openstack scheduler
- Start the integration with EGI-Engage Fedcloud
WP5 (SA1) (Peter Solagna)
Milestones & Deliverables
Task SA1.1 Operations Coordination
Task SA1.2 Development of Security Operations
Task SA1.3 Integration, Deployment of Grid and Cloud Platforms
WP6 (SA2) (Gergely Sipos)
Milestones & Deliverables
Task SA2.1 Training
Task SA2.2 Technical User Support
Task SA2.3 ELIXIR
Task SA2.4 BBMRI
Task SA2.5 MoBrain
Task SA2.6 DARIAH
Task SA2.7 LifeWatch
Task SA2.8 EISCAT_3D
Task SA2.9 EPOS
Task SA2.10 Disaster Mitigation