Mudanças entre as edições de "Cluster"

De Instituto de Física - UFRGS
 
(12 revisões intermediárias por 3 usuários não estão sendo mostradas)
Linha 1: Linha 1:
−
= Cluster Ada - Instituto de Física UFRGS =
+
= Clusters Ada and Lovelace - Instituto de Física UFRGS =
  
−
O Cluster está localizado no Instituto de Física da UFRGS, em Porto Alegre.
 
  
−
== Infraestrutura ==
+
The clusters are located at Instituto de Física da UFRGS, in Porto Alegre.
  
−
=== Software de gerenciamento ===
+
== Infraestruture ==
 +
 
 +
=== Management Software ===
  
 
<pre>
 
<pre>
Linha 13: Linha 14:
 
</pre>
 
</pre>
  
−
=== Hardware dos nodes ===
+
=== Hardware in the  ada nodes ===
  
 
<pre>
 
<pre>
 
CPU: x86_64
 
CPU: x86_64
−
RAM: varia entre 4 GB - 8 GB
+
RAM: varies between 8 GB - 16 GB
−
GPU: alguns nodes possuem NVIDIA CUDA
+
GPU: some nodes have NVIDIA CUDA
−
Storage: storage em rede com quota de 50 GB por usuário, os nodes não possuem HD local
+
Storage: storage with 50GB  quota per user
 
</pre>
 
</pre>
  
−
=== Software nos nodes ===
+
=== Hardware in the lovelace nodes ===
  
 
<pre>
 
<pre>
−
OS: Debian 8 (Jessie) x86_64
+
CPU: Ryzen (32 and 2*24 cores)
−
Pacotes instalados:
+
RAM: varia entre 64 GB
−
gcc
+
GPU: two nodes have NVIDIA CUDA
−
  docker
+
Storage: storage with 50GB quota per user
 
</pre>
 
</pre>
  
−
== Como utilizar ==
+
=== Software in the nodes ===
  
−
=== Conectar-se ao cluster-slurm ===
+
<pre>
 +
OS: Debian 8 (in cluster ada)
 +
OS: Debian 11 (in cluster lovelace)
 +
Basic packages installed:
 +
GCC
 +
gfortran
 +
python2
 +
python3
 +
</pre>
 +
 
 +
== How to use ==
 +
 
 +
=== Conect to  cluster-slurm ===
  
−
O cluster é acessível através do server cluster-slurm. Para acessar o server via SSH, use:
+
The clusters are accessible through server cluster-slurm.if.ufrgs.br (ou ada.if.ufrgr.br). To access through a unix-like system use:
 
<pre>
 
<pre>
−
ssh usuario@cluster-slurm.if.ufrgs.br
+
ssh <user>@cluster-slurm.if.ufrgs.br
 
</pre>
 
</pre>
  
−
Caso você não tenha cadastro ou não é vinculado ao Instituto de Física, solicite o cadastro enviando um email para fisica-ti@ufrgs.br.
+
or
  
−
=== Utilizando softwares no Cluster ===
+
<pre>
 +
ssh <user>@ada.if.ufrgs.br
 +
</pre>
 +
 
 +
Under windows you may use winscp.
 +
 
 +
If you are not registered, ask for registration sending an email to fisica-ti@ufrgs.br
  
−
Para que seja possível executar um programa em um job no cluster, o programa deve:
+
=== Using softwares in the cluster ===
  
−
1. Já estar instalado
+
To execute a software in a cluster job this program must:
  
−
OU
+
1. Be already installed
 +
 +
OR
  
−
2. Ser copiado para sua home (pasta do seu usuário)
+
2. Be copied to the user home  
  
 
Ex:
 
Ex:
 
<pre>
 
<pre>
−
scp meu_executavel usuario@cluster-slurm.if.ufrgs.br:~/
+
scp my_programm <user>@cluster-slurm.if.ufrgs.br:~/
 
</pre>
 
</pre>
  
−
Caso queira compilar o programa para uso no Cluster, uma das opções é usar o <code>gcc</code>.
+
If you are compiling your program in the cluster, one option is to user <code>gcc</code>.
  
 
Ex:
 
Ex:
 
<pre>
 
<pre>
 
scp -r source-code/ usuario@cluster-slurm.if.ufrgs.br:~/
 
scp -r source-code/ usuario@cluster-slurm.if.ufrgs.br:~/
−
ssh usuario@cluster-slurm.if.ufrgs.br:~/
+
ssh <user>@cluster-slurm.if.ufrgs.br:~/
 
cd source-code
 
cd source-code
−
gcc main.c
+
gcc main.c funcoes.c
 
</pre>
 
</pre>
−
Isso irá gerar um arquivo <code>a.out</code>, que é o executável.
+
This will generate file <code>a.out</code>, which is the executable.
  
−
Estando acessível pelo método 1 ou 2, o programa pode ser executado no Cluster através de um <strong>JOB</strong>.
+
Being accessible by methods 1 or 2, the program can be executed in the cluster through one <strong>JOB</strong>.
  
−
OBS: Caso você execute o programa sem submetê-lo como <strong>JOB</strong>, ele não será executado nos nodes, e sim apenas no próprio server (cluster-slurm), que possui capacidades bem limitadas de processamento.
+
OBS: If you execute your executable without submitting as <strong>JOB</strong>, it will be executed in the server, not in the nodes. This is not recommended since the server computational capabilities are limited and you will be slowing down the server for everyone else.
  
 +
=== Criating and executing a Job ===
  
−
=== Criando e executando um Job ===
+
Slurm manages jobs and each job represents a program or task being executed.
  
−
O Slurm gerencia jobs, e cada job representa um programa ou tarefa sendo executado.
+
To submit a new job, you must create a script file describing the requisites and characteristics of the Job.
  
−
Para submeter um novo Job, deve-se criar um arquivo de script descrevendo os requisitos e características de execução do Job.
+
A typical example of the content of a submission script is below
−
 
+
−
Formato do arquivo abaixo.
+
  
 
Ex: <code>job.sh</code>
 
Ex: <code>job.sh</code>
Linha 85: Linha 105:
 
<pre>
 
<pre>
 
#!/bin/bash  
 
#!/bin/bash  
−
#SBATCH -n 1 # Numero de CPU cores a serem alocados
+
#SBATCH -n 1 # Number of cpus to be allocated (Despite the # these SBATCH lines are compiled by the slurm manager!)
−
#SBATCH -N 1 # Numero de nodes a serem alocados
+
#SBATCH -N 1 # Nummber of nodes to be allocated  (You don't have to use all requisites, comment with ##)
−
#SBATCH -t 0-00:05 # Tempo limite de execucao (D-HH:MM)
+
#SBATCH -t 0-00:05 # Limit execution time (D-HH:MM)
−
#SBATCH -p long # Particao (fila) a ser submetido
+
#SBATCH -p long # Partition to be submitted
 
#SBATCH --qos qos_long # QOS  
 
#SBATCH --qos qos_long # QOS  
 
    
 
    
−
# Comandos de execução do seu programa:
+
# Your program execution commands
 
./a.out
 
./a.out
 
</pre>
 
</pre>
  
−
Na opção --qos, deve-se colocar o nome da partição com o prefixo "qos_":
+
In option --qos, use the partition name with "qos_" prefix:
  
−
partição: short -> qos: qos_short -> limite de 2 semanas
+
partition: short -> qos: qos_short -> limit  2 weeks
  
−
partição: long -> qos: qos_long -> limite de 3 meses
+
partition: long -> qos: qos_long -> limit de 3 month
 
    
 
    
 +
If you run on GPU, specify the "generic resource" gpu in cluster ada:
  
−
Caso deseje rodar em GPU, é necessário especificar a fila e pedir explicitamente a ''gereric resource'' gpu:
 
 
<pre>
 
<pre>
 
#!/bin/bash  
 
#!/bin/bash  
−
#SBATCH -n 1 # Numero de CPU cores a serem alocados
+
#SBATCH -n 1  
−
#SBATCH -N 1 # Numero de nodes a serem alocados
+
#SBATCH -N 1
−
#SBATCH -t 0-00:05 # Tempo limite de execucao (D-HH:MM)
+
#SBATCH -t 0-00:05  
−
#SBATCH -p long # Particao (fila) a ser submetido
+
#SBATCH -p long  
 
#SBATCH --qos qos_long # QOS  
 
#SBATCH --qos qos_long # QOS  
 
#SBATCH --gres=gpu:1
 
#SBATCH --gres=gpu:1
Linha 116: Linha 136:
 
</pre>
 
</pre>
  
−
Para pedir alguma GPU específica, use um constraint adicionando a linha:
+
To ask for a specific gpu:
 
<pre>
 
<pre>
 
#SBATCH --constraint="gtx970"
 
#SBATCH --constraint="gtx970"
 
</pre>
 
</pre>
  
−
Para submeter o job, execute o comando
+
To submit the job, execute:
  
 
<pre>
 
<pre>
Linha 127: Linha 147:
 
</pre>
 
</pre>
  
−
== Comandos úteis ==
+
== Usefull commands ==
−
* Para listar os seus jobs:
+
* To list jobs:
 
   squeue
 
   squeue
  
−
* Para deletar um job:
+
* To list all jobs running in the cluster now:
 +
  sudo squeue
 +
 
 +
* To delete a running job:
 
   scancel [job_id]
 
   scancel [job_id]
  
−
* Para listar as partições disponíveis:
+
* To list available partitions:
 
   sinfo
 
   sinfo
  
−
* Para listar as gpus presentes nos nós:
+
* To list gpu's in the nodes:
 
   sinfo -o "%N %f"
 
   sinfo -o "%N %f"
  
−
* Para listar um resumo de todos os nós:
+
* To list characteristic of all nodes:
 
   sinfo -Nel
 
   sinfo -Nel

Edição atual tal como às 15h51min de 14 de março de 2022

Clusters Ada and Lovelace - Instituto de Física UFRGS

The clusters are located at Instituto de Física da UFRGS, in Porto Alegre.

Infraestruture

Management Software

Slurm Workload Manager

Site :https://slurm.schedmd.com/

Hardware in the ada nodes

CPU: x86_64
RAM: varies between 8 GB - 16 GB
GPU: some nodes have NVIDIA CUDA
Storage: storage with 50GB  quota per user

Hardware in the lovelace nodes

CPU: Ryzen (32 and 2*24 cores)
RAM: varia entre 64 GB
GPU: two nodes have NVIDIA CUDA
Storage: storage with 50GB  quota per user

Software in the nodes

OS: Debian 8 (in cluster ada)
OS: Debian 11 (in cluster lovelace)
Basic packages installed:
 GCC
 gfortran
 python2
 python3

How to use

Conect to cluster-slurm

The clusters are accessible through server cluster-slurm.if.ufrgs.br (ou ada.if.ufrgr.br). To access through a unix-like system use: 

ssh <user>@cluster-slurm.if.ufrgs.br

or 

ssh <user>@ada.if.ufrgs.br

Under windows you may use winscp.

If you are not registered, ask for registration sending an email to fisica-ti@ufrgs.br

Using softwares in the cluster

To execute a software in a cluster job this program must:

1. Be already installed

OR

2. Be copied to the user home

Ex: 

scp my_programm <user>@cluster-slurm.if.ufrgs.br:~/

If you are compiling your program in the cluster, one option is to user gcc.

Ex: 

scp -r source-code/ usuario@cluster-slurm.if.ufrgs.br:~/
ssh <user>@cluster-slurm.if.ufrgs.br:~/
cd source-code
gcc main.c funcoes.c

This will generate file a.out, which is the executable.

Being accessible by methods 1 or 2, the program can be executed in the cluster through one JOB.

OBS: If you execute your executable without submitting as JOB, it will be executed in the server, not in the nodes. This is not recommended since the server computational capabilities are limited and you will be slowing down the server for everyone else.

Criating and executing a Job

Slurm manages jobs and each job represents a program or task being executed.

To submit a new job, you must create a script file describing the requisites and characteristics of the Job.

A typical example of the content of a submission script is below

Ex: job.sh 

#!/bin/bash 
#SBATCH -n 1 # Number of cpus to be allocated (Despite the # these SBATCH lines are compiled by the slurm manager!)
#SBATCH -N 1 # Nummber of nodes to be allocated  (You don't have to use all requisites, comment with ##)
#SBATCH -t 0-00:05 # Limit execution time (D-HH:MM)
#SBATCH -p long # Partition to be submitted
#SBATCH --qos qos_long # QOS 
  
# Your program execution commands
./a.out

In option --qos, use the partition name with "qos_" prefix:

partition: short -> qos: qos_short -> limit 2 weeks

partition: long -> qos: qos_long -> limit de 3 month

If you run on GPU, specify the "generic resource" gpu in cluster ada: 

#!/bin/bash 
#SBATCH -n 1 
#SBATCH -N 1
#SBATCH -t 0-00:05 
#SBATCH -p long 
#SBATCH --qos qos_long # QOS 
#SBATCH --gres=gpu:1
  
# Comandos de execução do seu programa:
./a.out

To ask for a specific gpu: 

#SBATCH --constraint="gtx970"

To submit the job, execute: 

sbatch job.sh

Usefull commands

  • To list jobs: 
 squeue
  • To list all jobs running in the cluster now: 
 sudo squeue
  • To delete a running job: 
 scancel [job_id]
  • To list available partitions: 
 sinfo
  • To list gpu's in the nodes: 
 sinfo -o "%N %f"
  • To list characteristic of all nodes: 
 sinfo -Nel
Disponível em "http://wiki-old.if.ufrgs.br/index.php?title=Cluster&oldid=2019"