Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix Nextflow config syntax #265

Merged
merged 1 commit into from
Dec 3, 2024
Merged

Fix Nextflow config syntax #265

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
108 changes: 55 additions & 53 deletions runscripts/nextflow/config.template
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,25 +14,36 @@ trace {

profiles {
gcloud {
// Retry once with more RAM if OOM
process.memory = { 4.GB * task.attempt }
// Using single core is slightly slower but much cheaper
process.cpus = 1
process.executor = 'google-batch'
process.container = params.container_image
// Necessary otherwise symlinks to other files in bucket can break
process.containerOptions = '--volume /mnt/disks/BUCKET:/mnt/disks/BUCKET'
process {
withLabel: parca {
cpus = params.parca_cpus
memory = params.parca_cpus * 2.GB
}
errorStrategy = {
// Codes: 137 (out-of-memory), 50001 - 50006 (Google Batch task fail:
// https://cloud.google.com/batch/docs/troubleshooting#reserved-exit-codes)
(((task.exitStatus == 137) || (task.exitStatus >= 50001 && task.exitStatus <= 50006))
&& (task.attempt <= process.maxRetries)) ? 'retry' : 'ignore'
}
// Retry once with more RAM if OOM
memory = { 4.GB * task.attempt }
maxRetries = 1
// Using single core is slightly slower but much cheaper
cpus = 1
executor = 'google-batch'
container = params.container_image
// Necessary otherwise symlinks to other files in bucket can break
containerOptions = '--volume /mnt/disks/BUCKET:/mnt/disks/BUCKET'
// Check Google Cloud latest spot pricing / performance
machineType = {
def cpus = task.cpus
def powerOf2 = 1
while (powerOf2 < cpus && powerOf2 < 64) {
powerOf2 *= 2
}
return "t2d-standard-${powerOf2}"
}
}
process.errorStrategy = {
// Codes: 137 (out-of-memory), 50001 - 50006 (Google Batch task fail:
// https://cloud.google.com/batch/docs/troubleshooting#reserved-exit-codes)
(((task.exitStatus == 137) || (task.exitStatus >= 50001 && task.exitStatus <= 50006))
&& (task.attempt <= process.maxRetries)) ? 'retry' : 'ignore' }
// For this script to work on a Compute Engine VM, you must
// - Set default Compute Engine region and zone for your project
// - Set access scope to "Allow full access to all Cloud APIs" when
Expand All @@ -54,60 +65,51 @@ profiles {
google.batch.subnetwork = "regions/${google.location}/subnetworks/default"
docker.enabled = true
params.projectRoot = '/vEcoli'
process.maxRetries = 1
// Check Google Cloud latest spot pricing / performance
process.machineType = {
def cpus = task.cpus
def powerOf2 = 1
while (powerOf2 < cpus && powerOf2 < 64) {
powerOf2 *= 2
}
return "t2d-standard-${powerOf2}"
}
workflow.failOnIgnore = true
}
sherlock {
process.memory = {
if ( task.exitStatus in [137, 140] ) {
4.GB * task.attempt
} else {
4.GB
}
}
process.errorStrategy = {
// Codes: 140 (SLURM job limits), 143 (SLURM preemption)
// Default value for exitStatus is max integer value, this
// is a catch-all for errors that leave no exit code
((task.exitStatus in [140, 143, Integer.MAX_VALUE])
&& (task.attempt <= process.maxRetries)) ? 'retry' : 'ignore' }
// Using single core is slightly slower but can have shorter
// queue times and is less damaging to future job priority
process.cpus = 1
process.executor = 'slurm'
process.queue = 'mcovert,owners'
process.container = params.container_image
apptainer.enabled = true
process {
// Run analyses, create variants, and run ParCa locally with
// the job used to launch workflow to avoid long queue times
withLabel: short {
executor = 'local'
},
}
// ParCa 4 CPUs in ~15 min, 1 CPU in ~30 min, not too bad
withLabel: parca {
executor = 'local'
cpus = 1
memory = 2.GB
}
}
process.time = {
if ( task.exitStatus == 140 ) {
1.h * task.attempt
} else {
1.h
container = params.container_image
queue = 'mcovert,owners'
executor = 'slurm'
// Single core sims are slightly slower but can have shorter
// queue times and is less damaging to future job priority
cpus = 1
memory = {
if ( task.exitStatus in [137, 140] ) {
4.GB * task.attempt
} else {
4.GB
}
}
time = {
if ( task.exitStatus == 140 ) {
1.h * task.attempt
} else {
1.h
}
}
errorStrategy = {
// Codes: 140 (SLURM job limits), 143 (SLURM preemption)
// Default value for exitStatus is max integer value, this
// is a catch-all for errors that leave no exit code
((task.exitStatus in [140, 143, Integer.MAX_VALUE])
&& (task.attempt <= process.maxRetries)) ? 'retry' : 'ignore'
}
maxRetries = 3
}
process.maxRetries = 3
apptainer.enabled = true
params.projectRoot = "${launchDir}"
// Avoid getting queue status too frequently (can cause job status mixups)
executor.queueStatInterval = '2 min'
Expand All @@ -127,15 +129,15 @@ profiles {
workflow.failOnIgnore = true
}
standard {
process.executor = 'local'
params.projectRoot = "${launchDir}"
workflow.failOnIgnore = true
process.errorStrategy = 'ignore'
process {
withLabel: parca {
cpus = params.parca_cpus
memory = params.parca_cpus * 2.GB
}
executor = 'local'
errorStrategy = 'ignore'
}
}
}
4 changes: 1 addition & 3 deletions runscripts/workflow.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -379,9 +379,7 @@ def main():
nf_config = nf_config.replace(
"PUBLISH_DIR", os.path.dirname(os.path.dirname(out_uri))
)
nf_config = nf_config.replace(
"PARCA_CPUS", str(config["parca_options"]["cpus"])
)
nf_config = nf_config.replace("PARCA_CPUS", str(config["parca_options"]["cpus"]))

# By default, assume running on local device
nf_profile = "standard"
Expand Down
Loading