Buck9195/Gladstone-Bioinformatics-Workshop
1
0
Fork 0
mirror of https://github.com/gladstone-institutes/Bioinformatics-Workshops.git synced 2025-11-30 09:45:43 -08:00
Code Issues Projects Releases Packages Wiki Activity Actions

incorporate feedback from Scooter

Browse source
This commit is contained in:
Natalie Elphick 2024-04-09 09:44:29 -07:00
parent 348d061ed0
commit b0dd9c589e
9 changed files with 130 additions and 1287 deletions
Download patch file Download diff file Expand all files Collapse all files

1
.gitignore vendored
View file

@ -1,2 +1,3 @@
.DS_Store
.Rproj.user
.Rhistory

84
docs/Working_on_Wynton_Part_1.html
View file

File diff suppressed because one or more lines are too long

37
docs/Working_on_Wynton_Part_2.html
View file

File diff suppressed because one or more lines are too long

53
working-on-wynton-hpc/Working_on_Wynton_Part_1.Rmd
View file

@ -1,5 +1,6 @@
---
title: "Working on Wynton - Part 1"
title: "Working on Wynton"
subtitle: "Part 1"
author: "Natalie Elphick"
date: "April 15th, 2024"
knit: (function(input, ...) {
@ -28,7 +29,7 @@ output:
**Natalie Elphick**
Bioinformatician I
**Alex Pico**
**Alex Pico (TA)**
Bioinformatics Core Director
@ -50,7 +51,7 @@ Bioinformatics Core Director
# What is Wynton HPC?
## High-performance Computing Cluster
## High-performance Computing Cluster {.smaller-picture}
- A collection of specialized computers (nodes) connected together on a fast local network
@ -75,7 +76,7 @@ Bioinformatics Core Director
- **Login:** Submit and query jobs. SSH to development nodes. File management.
- **Development:** Compile and install software. Test job scripts. Submit and query jobs. Version control. File management.
- **Compute:** Running short and long-running job scripts.
- **Compute:** Running job scripts.
- **Transfer:** Fast in- & outbound file transfers. File management.
## The Login Nodes {.small-bullets}
@ -92,8 +93,8 @@ log1, log2 and plog (for PHI users)
## Login {.small-bullets}
- Make sure you are on the UCSF or Gladstone WiFi networks (or the respective VPN)
- **ssh [your-username]@[node-name].wynton.ucsf.edu**
- Connect to the UCSF or Gladstone WiFi networks (or the respective VPN) or using [2FA](https://wynton.ucsf.edu/hpc/get-started/duo-signup.html)
- **ssh [your-username]@[node].wynton.ucsf.edu**
```{r, engine='bash', eval=TRUE, results='markup',comment=NA, highlight=TRUE, echo=FALSE}
echo "{local}$ ssh alice@log1.wynton.ucsf.edu
@ -108,7 +109,7 @@ alice@log1.wynton.ucsf.edu's password:
- Has a set of [core software](https://wynton.ucsf.edu/hpc/software/core-software.html) installed
- e.g. git, vim, nano, make and python
- Also has access to [software repositories](https://wynton.ucsf.edu/hpc/software/software-repositories.htmll) some which are maintained by other users or research groups
- Also has access to [software repositories](https://wynton.ucsf.edu/hpc/software/software-repositories.html) some which are maintained by other users or research groups
- e.g. matlab, R and openjdk
- Cannot be logged in to directly, only from a login node
```{r, engine='bash', eval=TRUE, results='markup',comment=NA, highlight=TRUE, echo=FALSE}
@ -138,14 +139,15 @@ dt1 and dt2
## Compute Nodes {.small-bullets .big-picture}
## Compute Nodes {.small-bullets}
- Can **not** be logged in to directly
- No internet or UCSF network access
- Used to run non-interactive compute job scripts
- The software to run the job script is provided using a container
![Compute Jobs](slide_materials/compute_job_workflow.svg)
![Compute Jobs](slide_materials/compute_job_workflow.png)
@ -161,9 +163,6 @@ dt1 and dt2
## BeeGFS {.small-bullets}
- Wynton uses a *parallel* shared file system called BeeGFS
@ -196,18 +195,18 @@ dt1 and dt2
- Prefer fewer, large files over many small ones
- Distribute reading and writing over several directories
- Including compute job output and error files
- Use local scratch (/scratch) when possible
- Use local scratch (**/scratch**) when possible
- Don't include anything in **/wynton** in your default LD_LIBRARY_PATH
## Storage {.small-bullets}
- **Wynton storage is not backed up**
- /wynton/**[group_name]**/**[user]**
- /wynton/home/**[group_name]**/**[user]**
- PHI users : /wynton/protected/home/**[group_name]**/**[user]**
- User home directory - limited to 500 GiB
- /wynton/**[group_name]**
- /wynton/group/**[group_name]**
- PHI users : /wynton/protected/group/**[group_name]**
- User group directory - disk quota varies by group
- Use this directory for any analysis you want to share with your lab
- [More information on disk quotas](https://wynton.ucsf.edu/hpc/howto/storage-size.html#file-sizes-and-disk-quotas)
@ -220,9 +219,10 @@ echo 'beegfs-ctl --getquota --storagepoolid=12 --gid "$(id --group)"'
## Scratch - Temporary Storage
## Scratch - Temporary Storage {.small-bullets}
- Local **/scratch** - 0.1-1.8 TiB/node storage unique to each compute node
- Can only be accessed from the specific compute node
- Use this to store intermediate files only needed for a job
- **/wynton/scratch** and **/wynton/protected/scratch** (for PHI users)
- 703 TiB storage accessible from everywhere
- No quotas
@ -245,8 +245,10 @@ echo 'beegfs-ctl --getquota --storagepoolid=12 --gid "$(id --group)"'
## Storage Advice
- Always back up anything you store under **/wynton**
- Use **/gladstone** if you have access to it for all of your work since it is automatically backed up
- Use the scratch directories to store temporary files so they do not count against your group or user quotas
- Back up your data on **/gladstone** if you have access to it
- A large number of jobs reading and writing to these directories will be slower since it is NFS mounted not BeeGFS
- Use the scratch directories to store temporary files
# Data Transfer
@ -372,26 +374,23 @@ echo '[alice@dev1 ~]$ nextflow -v'
# Containers
## Motivation {.small-bullets}
## Motivation {.small-bullets .small-picture}
- Compute heavy jobs (high RAM, multiple cores) should be run on compute nodes
- Containers allow us to make additional software available to the compute nodes
- Also allows the use of software that might be hard to install on Rocky 8 Linux
- Improves reproducibility
![Compute Jobs](slide_materials/compute_job_workflow.svg)
![Compute Jobs](slide_materials/compute_job_workflow.png)
## Definitions {.small-bullets}
- **Virtualization:** When software mimics the functions of physical hardware to run virtual machines
- **Containers:** Implements virtualization using an *image* as its base
- **Images:** An ordered collection of root filesystem changes and the corresponding execution parameters for use within a container runtime
- **Containers:** An isolated environment for running software that is created from an *image* file, preventing conflicts with the host system.
- **Images:** An ordered collection of root filesystem changes that contain all necessary dependencies, ensuring software run identically across various computing platforms.
## Apptainer {.small-bullets}

18
working-on-wynton-hpc/Working_on_Wynton_Part_2.Rmd
View file

@ -1,5 +1,6 @@
---
title: "Working on Wynton - Part 2"
title: "Working on Wynton"
subtitle: "Part 2"
author: "Natalie Elphick"
date: "April 16th, 2024"
knit: (function(input, ...) {
@ -28,7 +29,7 @@ output:
**Natalie Elphick**
Bioinformatician I
**Alex Pico**
**Alex Pico (TA)**
Bioinformatics Core Director
@ -53,14 +54,14 @@ Bioinformatics Core Director
# Custom Containers
## Motivation {.small-bullets}
## Motivation {.small-bullets .small-picture}
- Compute heavy jobs (high RAM, multiple cores) should be run on compute nodes
- Containers allow us to make additional software available to the compute nodes
- Also allows the use of software that might be hard to install on Rocky 8 Linux
- Improves reproducibility
![Compute Jobs](slide_materials/compute_job_workflow.svg)
![Compute Jobs](slide_materials/compute_job_workflow.png)
@ -125,13 +126,14 @@ echo "docker push docker_hub_user/seurat-harmony:1.0"
echo "[alice@dev1 ~]$ apptainer pull docker://docker_hub_user/seurat-harmony:1.0"
```
## Notes on Building Custom Images {.code-small}
## Notes on Building Custom Images {.small-bullets}
- Time consuming process and uses a lot of RAM on your local machine
- Time consuming process and can use a lot of RAM on your local machine
- A good base image can save you a lot of time
- You must run `apt-get update` and `apt-get install` in the same command
- You must run **apt-get update** and **apt-get install** in the same command
- Otherwise you will encounter caching issues
- Remember to use `apt-get install -y`
- These are only for Ubuntu, for other OS run the equivalent package list retrieval and install commands together
- Remember to use **apt-get install -y**
- You will have no control over the process while it's building
# Compute Jobs

1208
working-on-wynton-hpc/renv.lock
View file

File diff suppressed because it is too large Load diff

BIN
working-on-wynton-hpc/slide_materials/compute_job_workflow.png Normal file
View file

Binary file not shown.
Side by side

After

Width:  |  Height:  |  Size: 112 KiB

3
working-on-wynton-hpc/slide_materials/compute_job_workflow.svg
View file

File diff suppressed because one or more lines are too long
Side by side

Before

Width:  |  Height:  |  Size: 109 KiB

11
working-on-wynton-hpc/style.css
View file

@ -131,7 +131,7 @@ pre, code, kbd, samp {
/* Decrease size of image, remove border, shadow and center align*/
.reveal img {
max-width: 60%;
max-width: 70%;
border: none !important;
box-shadow: none !important;
display: block !important;
@ -164,6 +164,15 @@ small {
}
.small-picture img{
max-width: 65%;
}
.smaller-picture img{
max-width: 60%;
}
/* Chage link color to sky blue */
.reveal a {
color: #0c74dc;