---
title: "HPC On-Boarding"
canonical: "https://kb.uconn.edu/space/SH/26449903671/HPC%20On-Boarding"
format: markdown
---
#### Brief Overview

The Storrs HPC is a shared resource between all researchers on all UConn campuses. Need to be familiar with policy and resources of the cluster to properly utilize the cluster for your projects along with other members.

#### Why use High Performance Computing?

A personal computer such as a laptop or a desktop allow easy access for the user for local smaller file storage and non-intensive computational tasks. They are quite flexible and sometime portable for everyday use. 

A larger computer could be a more powerful machine that exist in your lab or office. They typically have more CPUs, memory, faster storage and possibly a powerful graphics card. These computers allow for more intensive tasks that your personal computer would struggle with. 

A high performance computing cluster has a collection of standalone computers that are networked together. It uses a software that allows for coordinated running of jobs across multiple computers to accomplish tasks much faster.

- Users do not communicate directly with a computational node. Instead, they connect to a login/head node.
- Jobs are submitted through a resource manager (SLURM) that allocates compute nodes.
- Accessed through command line interface.

> ℹ️ #### Skills Needed
> ℹ️ 
> ℹ️ 1. Proficiency in Linux Command-line
> ℹ️ 2. Knowledge about available hardware
> ℹ️ 3. Understand SLURM Scheduler
> ℹ️ 4. Familiarity with [Knowledge Base](https://kb.uconn.edu/space/SH)

#### Storrs HPC

Many partitions ([SLURM Partitions and Job Scheduling](https://uconn.atlassian.net/wiki/spaces/SH/pages/26032963610)) on the HPC that are used with differing run times.

For example, you’re able to run a job up to 12 hours in a general partition with up to 8 nodes.

In the debug partition, where people test for how much memory/resources they need to request for their full job, you can run 30 minutes with 1 node.

> ℹ️ **Warning: **Do not run jobs on the login node!

There are 3 login nodes (login4, login5, login6). You can run simple programs including file transfer, copy, removal and others. 

Let’s log on to the cluster and run our first command!

```
sinfo
```

```
(base) [anl14042@login5 ~]$ sinfo
PARTITION       AVAIL  TIMELIMIT  NODES  STATE NODELIST
GeoSciMP           up 1-00:00:00      1   drng cn423
GeoSciMP           up 1-00:00:00     29    mix cn[410,412-420,424-426,428-430,432-433,435,437-440,442-447]
GeoSciMP           up 1-00:00:00      8  alloc cn[411,421-422,427,431,434,436,441]
class              up    4:00:00      4  inval cn[406-409]
class              up    4:00:00      1  drng* cn333
class              up    4:00:00      4  comp* cn[562,567,569-570]
class              up    4:00:00      6 drain* cn[290,293,352,360,373,404]
class              up    4:00:00      1  idle* cn559
class              up    4:00:00      1  down* cn563
class              up    4:00:00      8   drng cn[334,338,353,374-376,402],gpu05
class              up    4:00:00     15  alloc cn[335-337,339-340,391,400,403,405,560-561,564-566,568]
class              up    4:00:00     39   idle cn[244-246,252,254-255,268-271,285-289,291-292,329-332,341-342,345-346,348-350,361-363,365-366,368-372,393]
debug              up      30:00      4  inval cn[406-409]
debug              up      30:00      4  comp* cn[562,567,569-570]
debug              up      30:00      4 drain* cn[290,293,360,373]
debug              up      30:00      1  idle* cn559
debug              up      30:00      1  down* cn563
debug              up      30:00      3   drng cn[374-376]
debug              up      30:00      8  alloc cn[400-401,560-561,564-566,568]
debug              up      30:00     29   idle cn[244-246,252,254-255,268-271,285-289,291-292,345-346,348-350,365,368-371],gpu[16-17]
general*           up   12:00:00     10  inval cn[343-344,354,356,359,396,406-409]
.
.
.
```

Take a look at the information displayed.


If you are not familiar with the Linux command line or would like a refresher, continue on to:

[https://uconn.atlassian.net/wiki/spaces/SH/pages/26449215565](https://uconn.atlassian.net/wiki/spaces/SH/pages/26449215565)