Merge pull request #14 from ULHPC/AWS-doc

[AWS] -- second call -- updated documentation
ULHPC · Jul 27, 2023 · bb4fb0d · bb4fb0d
2 parents 8c48192 + d24352c
commit bb4fb0d
Show file tree

Hide file tree

Showing 4 changed files with 345 additions and 4 deletions.
diff --git a/docs/aws/connection.md b/docs/aws/connection.md
@@ -8,7 +8,7 @@ Once your account has been enabled, you can connect to the cluster using ssh. Co
 To create a direct connection, use the command below (using your specific cluster name if it differs from workshop-cluster).
 
 ```bash
-ssh -i sshkey.pem username@ec2-44-209-227-16.compute-1.amazonaws.com
+ssh -i id_rsa username@ec2-52-5-167-162.compute-1.amazonaws.com 
 ```
 This will open a direct, non-graphical connection in the terminal. To exit the remote terminal session, use the standard Linux command “exit”.
 
@@ -17,7 +17,7 @@ Alternatively, you may want to save the configuration of this connection (and cr
 ```bash
 Host aws-ulhpc-access
   User username
-  Hostname ec2-44-209-227-16.compute-1.amazonaws.com
+  Hostname ec2-52-5-167-162.compute-1.amazonaws.com 
   IdentityFile ~/.ssh/id_rsa
   IdentitiesOnly yes
 ```

diff --git a/docs/aws/images/queues.png b/docs/aws/images/queues.png
diff --git a/docs/aws/overview.md b/docs/aws/overview.md
@@ -19,7 +19,7 @@ After initial project selection by a steering committee that includes representa
 
 ## System description and environment 
 
-The AWS Parallel Cluster based on the new Graviton3 (all instances and storage located in US-EAST-1) will provide cloud computing services to Uni.lu that will support the development, design, and testing of numerical codes, which traditionally demands a lot of compute power. This will give researchers an accessible, easy-to-use, end-to-end environment in which they can validate their simulation codes on ARM64 architectures, including servers, personal computers, and Internet of Things (IoT). The cluster will consist in two main partitions and jobs will be submitted using the [Slurm scheduler](https://slurm.schedmd.com/documentation.html) :
+The AWS Parallel Cluster based on the new [HPC-based Graviton3 instances](https://aws.amazon.com/ec2/instance-types/hpc7g/) (all instances and storage located in US-EAST-1) will provide cloud computing services to Uni.lu that will support the development, design, and testing of numerical codes, which traditionally demands a lot of compute power. This will give researchers an accessible, easy-to-use, end-to-end environment in which they can validate their simulation codes on ARM64 architectures, including servers, personal computers, and Internet of Things (IoT). The cluster will consist in two main partitions and jobs will be submitted using the [Slurm scheduler](https://slurm.schedmd.com/documentation.html) :