Questions tagged with High Performance Compute
Content language: English
Select up to 5 tags to filter
Sort by most recent
Browse through the questions and answers listed below or filter and sort to narrow down your results.
I'm running a python script which processes large amount of data and takes around 16 hrs to complete. The cpu usage is shown below in the pic. The gap between the two peaks is where, ec2-instance...
1
answers
0
votes
421
views
asked 8 months agolg...
Short backstory: I own a very small business that just started up. I live stream sporting events to our website, mostly powered by AWS. One challenge we've had is editing our very long and very large...
1
answers
0
votes
258
views
asked 8 months agolg...
Hi,
I'm trying to get s3 downloading fast. Right now, I have a p4d, and I have setup the NICs with 4 cards. The s3 client, somehow only use the primary NIC, so it never goes beyond the 100Gbps.
I...
1
answers
0
votes
230
views
asked 8 months agolg...
What should be the instance’s tenancy attribute for running it on single-tenant hardware?
1
answers
0
votes
209
views
asked 8 months agolg...
Hi,
I am trying to fine tune falcon 40b on amazon saegmaker but I keep running out of memory. I have tried using the following instances: P3.16x, 24xlarge, 12xlarge. Has anyone faced this error...
2
answers
0
votes
314
views
asked 8 months agolg...
How can I configure ParallelCluster with slurm to allow multiple jobs on nodes with multiple GPU?lg...
I'm trying to run many jobs that each use 1 GPU on ParallelCluster with the slurm scheduler. I'd like to use instances with 8 GPUs, like p3.16xlarge, and have slurm schedule 8 jobs to each node. I run...
1
answers
0
votes
618
views
asked 8 months agolg...
Hi,
I got a p4d instance and according to the [link](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-eni.html) it has 4 network cards.
To fully utilize this, do I need to assign 4 NICs for...
2
answers
0
votes
223
views
asked 8 months agolg...
Hi,
I have a t2.micro machine which is running a small application in it. When I saw the my machine is Under-provisioned then I look into details I found its EBS is under provisioned "EBS read...
0
answers
0
votes
144
views
asked 9 months agolg...
Me and my team are in the design stages for a very high-traffic, (potentially) low-latency process. We are interested in a serverless design. We would like to use Lambda, but the Burst Concurrency...
2
answers
0
votes
272
views
asked 9 months agolg...
Hello,
I created an HPC cluster with PCluster and used the free instance t2.micro.
How can I run the IOR benchmark on it without using Luster.
Are there other IO benchmarks that could be run on this...
1
answers
0
votes
167
views
asked 9 months agolg...
In our solution we are trying to support real-time updates on large scale(around million concurrent users) with low latency and keep low maintenance cost using AWS.
Some points regarding our solution...
2
answers
0
votes
259
views
asked 9 months agolg...
unknown compute instances we are getting billing
Ex: we have r5.large 1 compute instance running but here charging like 5 compute instances. can anyone look in to this case once.
2
answers
0
votes
238
views
asked 9 months agolg...