Difference between revisions of "GPGPU-FedCloud"

From EGIWiki
Jump to: navigation, search
(Status of accelerated computing in Clouds)
Line 1: Line 1:
 
= Status of accelerated computing in Clouds  =
 
= Status of accelerated computing in Clouds  =
 
<div>Need modification/support at all levels</div>  
 
<div>Need modification/support at all levels</div>  
<br>
+
<br>  
 
 
Chipset&nbsp;: HW virtualization support (otherwise some limitation)
 
  
 +
*Chipset&nbsp;: HW virtualization support (otherwise some limitation)
 
*OS level: correct kernel configuration for the accelerators  
 
*OS level: correct kernel configuration for the accelerators  
 
*Hypervisor: configuration pass-through, vGPU  
 
*Hypervisor: configuration pass-through, vGPU  
 
*CMFs: VM start, scheduler  
 
*CMFs: VM start, scheduler  
 
*FedCloud facilities: accounting, information discovery  
 
*FedCloud facilities: accounting, information discovery  
*Application: VM images with correct drivers for specific chipsets  
+
*Application: VM images with correct drivers for specific chipsets<br>
*<font face="Segoe UI">&lt;span style="font-size: 37.3333320617676px;" /&gt;</font>  
+
 
*<br>
+
 
 +
 
 +
= Accelerators =
 +
 
 +
== GPGPU (General-Purpose computing on Graphical Processing Units) ==
 +
<div>NVIDIA GPU/Tesla/GRID, AMD Radeon/FirePro, Intel HD Graphics,...</div><div>Virtualization using VGA pass-through, vGPU (GPU partitioning) - NVIDIA GRID accelerators</div><div><br></div>

Revision as of 13:33, 30 April 2015

Status of accelerated computing in Clouds

Need modification/support at all levels


  • Chipset : HW virtualization support (otherwise some limitation)
  • OS level: correct kernel configuration for the accelerators
  • Hypervisor: configuration pass-through, vGPU
  • CMFs: VM start, scheduler
  • FedCloud facilities: accounting, information discovery
  • Application: VM images with correct drivers for specific chipsets


Accelerators

GPGPU (General-Purpose computing on Graphical Processing Units)

NVIDIA GPU/Tesla/GRID, AMD Radeon/FirePro, Intel HD Graphics,...
Virtualization using VGA pass-through, vGPU (GPU partitioning) - NVIDIA GRID accelerators