65 lines
3.0 KiB
Markdown
65 lines
3.0 KiB
Markdown
---
|
|
title: Introduction
|
|
#tags:
|
|
keywords: introduction, home, welcome, architecture, design
|
|
last_updated: 07 September 2022
|
|
#summary: "Merlin 6 cluster overview"
|
|
sidebar: merlin6_sidebar
|
|
permalink: /merlin6/introduction.html
|
|
redirect_from:
|
|
- /merlin6
|
|
- /merlin6/index.html
|
|
---
|
|
|
|
## The Merlin local HPC cluster
|
|
|
|
Historically, the local HPC clusters at PSI were named **Merlin**. Over the years,
|
|
multiple generations of Merlin have been deployed.
|
|
|
|
At present, the **Merlin local HPC cluster** contains _two_ generations of it:
|
|
* the old **Merlin5** cluster (`merlin5` Slurm cluster), and
|
|
* the newest generation **Merlin6**, which is divided in two Slurm clusters:
|
|
* `merlin6` as the Slurm CPU cluster
|
|
* `gmerlin6` as the Slurm GPU cluster.
|
|
|
|
Access to the different Slurm clusters is possible from the [**Merlin login nodes**](/merlin6/interactive.html),
|
|
which can be accessed through the [SSH protocol](/merlin6/interactive.html#ssh-access) or the [NoMachine (NX) service](/merlin6/nomachine.html).
|
|
|
|
The following image shows the Slurm architecture design for the Merlin5 & Merlin6 (CPU & GPU) clusters:
|
|
|
|
![Merlin6 Slurm Architecture Design]({{ "/images/merlin-slurm-architecture.png" }})
|
|
|
|
### Merlin6
|
|
|
|
Merlin6 is a the official PSI Local HPC cluster for development and
|
|
mission-critical applications that has been built in 2019. It replaces
|
|
the Merlin5 cluster.
|
|
|
|
Merlin6 is designed to be extensible, so is technically possible to add
|
|
more compute nodes and cluster storage without significant increase of
|
|
the costs of the manpower and the operations.
|
|
|
|
Merlin6 contains all the main services needed for running cluster, including
|
|
**login nodes**, **storage**, **computing nodes** and other *subservices*,
|
|
connected to the central PSI IT infrastructure.
|
|
|
|
#### CPU and GPU Slurm clusters
|
|
|
|
The Merlin6 **computing nodes** are mostly based on **CPU** resources. However,
|
|
it also contains a small amount of **GPU**-based resources, which are mostly used
|
|
by the BIO Division and by Deep Leaning project.
|
|
|
|
These computational resources are split into **two** different **[Slurm](https://slurm.schedmd.com/overview.html)** clusters:
|
|
* The Merlin6 CPU nodes are in a dedicated **[Slurm](https://slurm.schedmd.com/overview.html)** cluster called [**`merlin6`**](/merlin6/slurm-configuration.html).
|
|
* This is the **default Slurm cluster** configured in the login nodes: any job submitted without the option `--cluster` will be submited to this cluster.
|
|
* The Merlin6 GPU resources are in a dedicated **[Slurm](https://slurm.schedmd.com/overview.html)** cluster called [**`gmerlin6`**](/gmerlin6/slurm-configuration.html).
|
|
* Users submitting to the **`gmerlin6`** GPU cluster need to specify the option ``--cluster=gmerlin6``.
|
|
|
|
### Merlin5
|
|
|
|
The old Slurm **CPU** *merlin* cluster is still active and is maintained in a best effort basis.
|
|
|
|
**Merlin5** only contains **computing nodes** resources in a dedicated **[Slurm](https://slurm.schedmd.com/overview.html)** cluster.
|
|
* The Merlin5 CPU cluster is called [**merlin5**](/merlin5/slurm-configuration.html).
|
|
|