Accelerated Computing in EGI Federated Cloud

This document discusses the implementation of GPGPU technology in cloud computing, focusing on enabling technologies, cloud middleware frameworks, and a specific GPU-enabled site called IISAS-GPUCloud. It covers virtualization technologies, container-based technologies, and details about the hardware and software setup at the IISAS-GPUCloud site. Users have full control over virtual machines on this site, with support for various Virtual Organizations (VOs) and easy access via rOCCI command-line client or OpenStack Horizon portal.

• Cloud Computing
• GPGPU
• Virtualization
• OpenStack
• IISAS-GPUCloud

jae_spe Follow

Uploaded on Dec 15, 2024 | 0 Views

Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.

Presentation Transcript

1. Accelerated Computing in EGI Federated Cloud Viet Tran, Jan Astalos, Miroslav Dobrucky Institute of Informatics Slovak Academy of Sciences www.egi.eu EGI-Engage is co-funded by the Horizon 2020 Framework Programme of the European Union under grant number 654142

2. Enabling technologies For using GPGPU in clouds, the GPGPU must be supported in all layers including virtualization, CMF and higher tools Virtualization technologies KVM with PCI passthrough Widely used but with limitation Virtualized GPU is in a early stage: NVIDIA GRID vGPU (XenServer, VMWare hyperv. only) SR-IOV based AMD MxGPU (VMWare hyperv. only) Intel GVT-G recently added to Linux 4.10 kernel Container-based technologies LXD with GPU is promising But limited support in cloud middleware 12/15/2024 2

3. Enabling technologies Cloud middleware framework Openstack with PCI passthrough: supported OpenNebula with PCI passthrough: supported Openstack with LXD: limited support EGI FedCloud tools Information service: GLUE2.1 defined, deployment in progress Accounting service: work in progress 12/15/2024 3

4. IISAS-GPUCloud site Openstack GPGPU site at IISAS Hardware: IBM dx360 M4 servers with 2x Intel Xeon E5-2650v2 16 CPU cores, 64GB RAM, 1 TB storage on each WN 2x NVIDIA Tesla K20m on each WN Software Base OS: Ubuntu 16.04 LTS KVM hypervisor with PCI passthrough virtualisation of GPU cards OpenStack Mitaka middleware Newest Federated Cloud tools 12/15/2024 4

5. IISAS-GPUCloud site GPU-enabled machine types: gpu1cpu6 (1GPU + 6 CPU cores) gpu2cpu12 (2GPU +12 CPU cores) Pre-defined images with NVIDIA drivers and CUDA and OpenCL libraries installed for most used Linux distributions Users have full control over virtual machines and can install/deploy additional software/services Supported VOs: fedcloud.egi.eu, ops, dteam, moldyngrid, enmr.eu, vo.lifewatch.eu, acc-comp.egi.eu 12/15/2024 5

6. Using IISAS-GPUCloud site Via rOCCI command-line client Simply choose GPU-enable flavor (e.g. gpu2cpu12) as resource template Or via Openstack Horizon portal Graphical interface Adding support for EGI users to login via token (no username/password) 12/15/2024 6

7. IISAS-GPUCloud portal 12/15/2024 7

8. IISAS-Nebula site Hardware: IBM dx360 M4 servers with 2x Intel Xeon E5-2650v2 16 CPU cores, 64GB RAM, 1 TB storage on each WN 2x NVIDIA Tesla K20m on each WN OpenNebula 5.0 Fully integrated to EGI FedCloud, certified Nearly same image list and flavors Access via rOCCI client 12/15/2024 8

9. IISAS-LXD site Experimental site with Openstack, LXD and GPGPU More stable and flexible manipulation with GPGPU Lower overhead Limitation GPU properties are not defined Block storage supports are limited 12/15/2024 9

10. Docker support Dockers with GPGPU can be executed on GPU sites Create a VM with GPGPU-enable flavor and image Run docker with proper mapping to access GPU docker run --name=XXXXXX \ --device=/dev/nvidia0:/dev/nvidia0 \ --device=/dev/nvidia1:/dev/nvidia1 \ --device=/dev/nvidiactl:/dev/nvidiactl \ --device=/dev/nvidia-uvm:/dev/nvidia-uvm \ .. 12/15/2024 10

11. Supports User tutorial: How to use GPGPU on IISAS-GPUCloud/IISAS-Nebula Access via rOCCI client Access via OpenStack dashboard with token How to create your own GPGPU server in cloud Site admin guide How to enable GPGPU passthrough in OpenStack Additional tools Automation via scripts: NVIDIA + CUDA installer Keystone-VOMS client for getting token Keystone-voms module for Openstack Horizon All this in a wiki: https://wiki.egi.eu/wiki/GPGPU-FedCloud https://wiki.egi.eu/wiki/GPGPU-OpenNebula 12/15/2024 11

12. GPU specific VO acc-comp.egi.eu VO has been established for testing and development with GPGPU: VO image list with preinstalled GPU drivers and CUDA libraries are available via AppDB Supported only at sites with GPGPU hardware More info at https://wiki.egi.eu/wiki/Accelerated_computing_VO 12/15/2024 12

13. Accelerated computing website 12/15/2024 13

14. More information https://wiki.egi.eu/wiki/GPGPU-FedCloud https://wiki.egi.eu/wiki/GPGPU-OpenNebula https://wiki.egi.eu/wiki/Accelerated_computing_V O https://accelerated.ui.sav.sk/?page_id=21 https://horizon.ui.savba.sk/ 12/15/2024 14

15. Thank you for your attention. Questions? www.egi.eu EGI-Engage is co-funded by the Horizon 2020 Framework Programme of the European Union under grant number 654142