Alert.png The wiki is deprecated and due to be decommissioned by the end of September 2022.
The content is being migrated to other supports, new updates will be ignored and lost.
If needed you can get in touch with EGI SDIS team using operations @ egi.eu.

EGI-Engage:TASK JRA2.4 Accelerated Computing

From EGIWiki
Jump to navigation Jump to search
EGI-Engage project: Main page WP1(NA1) WP3(JRA1) WP5(SA1) PMB Deliverables and Milestones Quality Plan Risk Plan Data Plan
Roles and
responsibilities
WP2(NA2) WP4(JRA2) WP6(SA2) AMB Software and services Metrics Project Office Procedures



Task leader: Marco Verlato/INFN
Task leader deputy: Viet Tran/IISAS

Mailing list:egi-engage-wp4.4 at mailman.egi.eu


Introduction

Accelerated computing systems deliver energy efficient and powerful HPC capabilities. Many EGI sites are providing accelerated computing technologies to enable high performance processing such as GPGPUs or MIC co-processors. Currently these accelerated capabilities are not directly supported by the EGI platforms. To use the co-processors capabilities available at resource centre level, users must directly interact with the local provider to get information about the type of resources and software libraries available and which submission queues must be used to submit tasks of accelerated computing. This task will implement the support in the information system, to expose the correct information about the accelerated computing technologies available – both software and hardware – at site level, developing a common extension of the information system structure, based on OGF GLUE standard, in order to have the capabilities published uniformly by all the sites. Users will then be able to extract all the information directly from the information system without interacting with the sites, and easily use resources provided by multiple sites. The task will also extend the HTC and cloud middleware support for co-processors, where needed, in order to provide a transparent and uniform way to allocate these resources together with CPU cores efficiently to the users.

Involved Partner

Previous work

Current activity

Roadmap

Project month

Work to be done

Goal 

(measurable)

Owner

(reponsible person)

Comments
1 Collection of people names and population of mailing lists Communications in place T. Ferrari COMPLETE
2 Review of available technologies for using GPUs in grid and cloud  

Resume of EGI-Inspire GPGPU VT

Kick-off and GPGPU VT handover meeting M.Verlato COMPLETE
3 Review of available technologies for using GPUs in grid and cloud  

Analysis of requirements from user communities

Discussion at GPGPU session at EGI Conference M.Verlato COMPLETE
4 End of review of available technologies for using GPUs in grid and cloud  

End of analysis of requirements from user communities

Technical report (to be included in Milestone 1.2 First intermediate report) and wiki pages
M.Verlato

7 Modification/configuration at site level (OS driver, hypervisor, CMF, VM images) for supporting GPUs in cloud  

Definition of abstract GPU-related CREAM JDL attributes to be mapped in the supported batch systems

Cloud site with working GPU and installation manual  

Update of the CREAM JDL specification document

M.Verlato

10 First step of integration with FedCloud (AppDB, accounting, brokering, ...)  

First implementation of GPU support in CREAM

Tools and configuration manual, possible some demonstration  

GPU-enabled CREAM prototype released for testing

M.Verlato

12 Revision of available technologies, further improvement and extension of supporting GPUs in FedCloud  

End of testing of GPU-enabled CREAM prototype and feedback collection

Technical report contributing to Project Periodic Report (first period)  

Official final CREAM release with full GPU support

M.Verlato