github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #997] Run Ollama on AWS #483

New Issue
Closed
opened 2026-04-12 10:09:50 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-12 10:09:50 -05:00
Owner
Copy Link

Originally created by @telestia on GitHub (Nov 4, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/997

I created an AMI on which I run Ollama and made sure that it works fine, but when I create another machine from this AMI, although the OLLAMA services are active, an ollama process does not appear in the nvdia-smi command, and when I try to run my model with the ollama run command, only the loading bar returns. . I have no idea what might cause this problem, I would appreciate it if you could help me.

Originally created by @telestia on GitHub (Nov 4, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/997 I created an AMI on which I run Ollama and made sure that it works fine, but when I create another machine from this AMI, although the OLLAMA services are active, an ollama process does not appear in the nvdia-smi command, and when I try to run my model with the ollama run command, only the loading bar returns. . I have no idea what might cause this problem, I would appreciate it if you could help me.
GiteaMirror commented 2026-04-12 10:09:50 -05:00
Author
Owner
Copy Link

@igorschlum commented on GitHub (Nov 6, 2023):

Hello telestia,

An AMI, or Amazon Machine Image, is a template that contains a software configuration for launching an instance on AWS. An instance is a virtual server in the cloud that runs the software in the AMI. You can launch multiple instances from the same AMI, or different types of instances depending on your needs. For example, you can launch an instance with an AMI that has an operating system, an application server, and applications pre-installed, or you can launch an instance with an AMI that has only an operating system and then install the software you want on it. You can also create your own custom AMIs, or use AMIs provided by AWS or the community. For more information, see Amazon Machine Images https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/AMIs.html or Instances and https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-instances-and-amis.html

Your AMI is with the linux version of Ollama? Do you use the Docker version of Ollama? Did you try again with version 0.1.8 ?

<!-- gh-comment-id:1793893270 --> @igorschlum commented on GitHub (Nov 6, 2023): Hello telestia, An AMI, or Amazon Machine Image, is a template that contains a software configuration for launching an instance on AWS. An instance is a virtual server in the cloud that runs the software in the AMI. You can launch multiple instances from the same AMI, or different types of instances depending on your needs. For example, you can launch an instance with an AMI that has an operating system, an application server, and applications pre-installed, or you can launch an instance with an AMI that has only an operating system and then install the software you want on it. You can also create your own custom AMIs, or use AMIs provided by AWS or the community. For more information, see Amazon Machine Images https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/AMIs.html or Instances and https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-instances-and-amis.html Your AMI is with the linux version of Ollama? Do you use the Docker version of Ollama? Did you try again with version 0.1.8 ?
GiteaMirror commented 2026-04-12 10:09:51 -05:00
Author
Owner
Copy Link

@telestia commented on GitHub (Nov 6, 2023):

I am using linux version of ollama on ubuntu 22.04. I have come a little further in solving the problem. After creating my ec2 instance and waiting for about 10 minutes, I realized that I could run it. I thought this might be related to disk speed, so I'll try creating my AMI again with a faster disk.

<!-- gh-comment-id:1794254687 --> @telestia commented on GitHub (Nov 6, 2023): I am using linux version of ollama on ubuntu 22.04. I have come a little further in solving the problem. After creating my ec2 instance and waiting for about 10 minutes, I realized that I could run it. I thought this might be related to disk speed, so I'll try creating my AMI again with a faster disk.
GiteaMirror commented 2026-04-12 10:09:51 -05:00
Author
Owner
Copy Link

@johnzhao9 commented on GitHub (Aug 1, 2024):

thanks for the lead telestia, it's definitely related to disk

for future readers, faster disks are kind of expensive however, running fio through ollama's blob where the models are stored worked with the same slow disk

on ubuntu, try this.

sudo apt update
sudo apt-get install -y fio
for file in /usr/share/ollama/.ollama/models/blobs/*; do fio --name=readfile --filename="$file" --rw=read --bs=1M --iodepth=32 --ioengine=libaio --direct=1 --status-interval=5; done

this exact issue and fio work around comes from this AWS source: Addressing I/O latency when restoring Amazon EBS volumes from EBS Snapshots
by Vishal Gupta | on 18 APR 2022

<!-- gh-comment-id:2263706469 --> @johnzhao9 commented on GitHub (Aug 1, 2024): thanks for the lead telestia, it's definitely related to disk for future readers, faster disks are kind of expensive however, running fio through ollama's blob where the models are stored worked with the same slow disk on ubuntu, try this. ``` sudo apt update sudo apt-get install -y fio for file in /usr/share/ollama/.ollama/models/blobs/*; do fio --name=readfile --filename="$file" --rw=read --bs=1M --iodepth=32 --ioengine=libaio --direct=1 --status-interval=5; done ``` this exact issue and fio work around comes from this AWS source: [Addressing I/O latency when restoring Amazon EBS volumes from EBS Snapshots by Vishal Gupta | on 18 APR 2022](https://aws.amazon.com/blogs/storage/addressing-i-o-latency-when-restoring-amazon-ebs-volumes-from-ebs-snapshots/)
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#483
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?