Server Fault Stack Exchange
Server Fault Stack Exchange

Questions tagged [gpu]

46 questions
32
votes
7 answers

Does a server need a GPU?

Do I need a GPU on a text and console only server? No GPU as in no iGPU and dGPU. Im going to be using SSH, so I dont need a display out. Im using Linux, but the OS shouldn't affect the results
tymur999
  • 465
  • 1
  • 3
  • 8
3
votes
1 answer

What is the best metric for auto-scaling GPU instances for machine learning inference in the cloud?

We have an API in AWS with a GPU instance that does inference. We have an auto-scaler setup with the minimum and maximum number of instances, but aren’t sure which metric (GPU/CPU usage, RAM usage, average latency, etc) or combination of metrics…
elwray14
  • 31
  • 1
2
votes
0 answers

Quota 'GPUS_ALL_REGIONS' exceeded. Limit: 0.0 globally

I am trying to create the VM instance with NVIDIA K80 GPUs in Asia-East1 so, I requested to increase the quota and team have adjusted the quota. However, when I am trying to create the VM instance by selecting the NVIDIA K80 GPUs with 4 CPUs and…
Tushar Shah
  • 21
  • 2
2
votes
1 answer

Pod is stuck in PodInitializing status when an initContainer is OOMKilled

I have the following on-prem Kubernetes environment: OS: Red Hat Enterprise Linux release 8.6 (Ootpa) Kubernetes: 1.23.7 (single-node, build with kubeadm) NVIDIA driver: 515.65.01 nvidia-container-toolkit: 1.10.0-1.x86_64 (rpm) containerd:…
Daigo
  • 278
  • 1
  • 17
2
votes
1 answer

GPU Acceleration on a Windows Server without virtualization over RDP

I'm trying to find out if it's possible to run a Windows Server with one GPU which is shared between all RDP clients so that people could create a session on the server start some program with a UI which needs GPU acceleration disconnect afterwards…
ridilculous
  • 121
  • 3
2
votes
2 answers

GCP does not have enough resources available to fulfill the request for about a month

I've been trying to start my existing GCP VM that has an NVIDIA T4 GPU attached to it, for almost a month at this time. It has been working fine before but now I am constantly getting the error message: The zone '***' does not have enough resources…
masus04
  • 121
  • 4
2
votes
0 answers

Slurm srun cannot allocate ressources for GPUs - Invalid generic resource specification

I am able to launch a job on a GPU server the traditional way (using CPU and MEM as consumables): ~ srun -c 1 --mem 1M -w serverGpu1 hostname serverGpu1 but trying to use the GPUs will give an error: ~ srun -c 1 --mem 1M --gres=gpu:1 hostname srun:…
user324810
  • 121
  • 3
2
votes
1 answer

Kubernetes: How can I get which pod schedule GPU?

I have three Nvidia GPUs in my cluster, and so many pods are running in my cluster. How can I find which one of these pods schedule GPU and how many GPUs they schedule? I use this link to enable the GPU schedule in my cluster for Nvidia…
nader
  • 33
  • 7
2
votes
1 answer

Use passthrough GPU in KVM/QEMMU and display in host OS in a window

I'm new to KVM/QEMMU. I have used virtual box to run Windows 10 in a virtual machine on my Arch host system (a laptop with both integrated and discrete GPUs). Being dissatisfied with the video responsiveness of the guest machine I decided to assign…
brett
  • 121
  • 1
  • 3
2
votes
3 answers

Why are GPUs accessible from docker containers running on Linux hosts, but not on Windows or MacOS hosts?

Recent versions of docker (or any version of nvidia-docker) allow direct(?) access to the host GPU from within docker containers, with full access to CUDA APIs. This is very convenient when deploying complex machine learning inference…
Will
  • 219
  • 2
  • 7
1
vote
1 answer

always available gpu servers

I need several GPU servers for rendering. Region doesn`t matter. Preemptible - I think it is ok for me. It is important that at any time I should be able to run a couple of instances with the GPU. As I understand it, Google can take away quotas for…
Sergey Kozlov
  • 11
  • 1
1
vote
0 answers

How Do You Run GPU Task on Windows Server 2016 Remotely?

From my understanding, remotely executing GPU tasks is a little tricky because of the way sessions work on Windows. Session 0 does not have access to GPU drivers, and usually when you remotely execute a task it calls it in Session 0. In order to…
Syed Aman
  • 11
  • 2
1
vote
0 answers

DDA device out of resource(Error 12) in Hyper-V VM

We're 'borrowing' a server from an IDC that had two Tesla V100 on it. It runs Windows Server 2016. Since it has some impressive graphics capabilities, we were looking for ways to run graphic accelerated programs(CAD, Blender, etc). But the server…
iCore
  • 11
  • 3
1
vote
1 answer

Unable to use gpu in azure windows server 2016

I am trying to run a GPU intensive application(Lumion) on the Azure cloud. Image used - Windows Server 2016. Hardware - NV6_Promo with 1xK80 GPU. Any application, when launched, run without using the GPU. The system has the latest Nvidia Tesla…
Nithin Jose
  • 149
  • 7
1
vote
0 answers

Double-precision floating point performance on GPUs

I have an application which needs to run lots of double-precision floating point operations in parallel on small datasets. I've just started exploring the possibility of running these computations on a GPU. While comparing performance metrics across…
Bobface
  • 135
  • 4
1
2 3 4