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

Jump to bottom

nextstrain VM shape tuning #112

Merged
merged 3 commits into from
Jun 11, 2020
Merged

nextstrain VM shape tuning #112

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
2a3e375
VM shape tuning based on a representative run, plus add reporting to …
dpark01 Jun 10, 2020
4acda34
fix dx_instance_type for draft tree
dpark01 Jun 10, 2020
d4f1e5f
Merge remote-tracking branch 'origin/master' into dp-nextstrain
dpark01 Jun 11, 2020
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
7 changes: 7 additions & 0 deletions pipes/WDL/tasks/tasks_demux.wdl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -239,6 +239,10 @@ task illumina_demux {
--out_zip ,,_fastqc.zip \
--threads $num_fastqc_threads" \
::: $(cat $OUT_BASENAMES)

cat /proc/uptime | cut -f 1 -d ' ' > UPTIME_SEC
cat /proc/loadavg | cut -f 3 -d ' ' > LOAD_15M
cat /sys/fs/cgroup/memory/memory.max_usage_in_bytes > MEM_BYTES
}

output {
Expand All @@ -249,6 +253,9 @@ task illumina_demux {
File unmatched_reads_bam = "unmatched/Unmatched.bam"
Array[File] raw_reads_fastqc = glob("*_fastqc.html")
Array[File] raw_reads_fastqc_zip = glob("*_fastqc.zip")
Int max_ram_gb = ceil(read_float("MEM_BYTES")/1000000000)
Int runtime_sec = ceil(read_float("UPTIME_SEC"))
Int cpu_load_15min = ceil(read_float("LOAD_15M"))
String viralngs_version = read_string("VERSION")
}

Expand Down
103 changes: 60 additions & 43 deletions pipes/WDL/tasks/tasks_nextstrain.wdl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,7 +91,6 @@ task filter_subsample_sequences {
String? exclude_where
String? include_where

Int? machine_mem_gb
String docker = "nextstrain/base"
}
parameter_meta {
Expand Down Expand Up @@ -126,6 +125,9 @@ task filter_subsample_sequences {
#cat ~{sequences_fasta} | grep \> | wc -l > IN_COUNT
grep "sequences were dropped during filtering" STDOUT | cut -f 1 -d ' ' > DROP_COUNT
grep "sequences have been written out to" STDOUT | cut -f 1 -d ' ' > OUT_COUNT
cat /proc/uptime | cut -f 1 -d ' ' > UPTIME_SEC
cat /proc/loadavg | cut -f 3 -d ' ' > LOAD_15M
cat /sys/fs/cgroup/memory/memory.max_usage_in_bytes > MEM_BYTES
}
runtime {
docker: docker
Expand All @@ -140,6 +142,9 @@ task filter_subsample_sequences {
String augur_version = read_string("VERSION")
Int sequences_dropped = read_int("DROP_COUNT")
Int sequences_out = read_int("OUT_COUNT")
Int max_ram_gb = ceil(read_float("MEM_BYTES")/1000000000)
Int runtime_sec = ceil(read_float("UPTIME_SEC"))
Int cpu_load_15min = ceil(read_float("LOAD_15M"))
}
}

Expand Down Expand Up @@ -170,20 +175,23 @@ task augur_mafft_align {
~{true="--remove-reference" false="" remove_reference} \
--debug \
--nthreads auto
cat /proc/uptime | cut -f 1 -d ' ' > UPTIME_SEC
cat /proc/loadavg | cut -f 3 -d ' ' > LOAD_15M
cat /sys/fs/cgroup/memory/memory.max_usage_in_bytes > MEM_BYTES
}
runtime {
docker: docker
memory: select_first([machine_mem_gb, 104]) + " GB"
cpu : 16
memory: select_first([machine_mem_gb, 208]) + " GB"
cpu : 32
disks: "local-disk ${disk_space_gb} LOCAL"
preemptible: 0
dx_instance_type: "mem3_ssd2_v2_x16"
dx_instance_type: "mem3_ssd2_v2_x32"
}
output {
File aligned_sequences = "~{basename}_aligned.fasta"
File align_troubleshoot = stdout()
Int max_ram_gb = ceil(read_float("MEM_BYTES")/1000000000)
Int runtime_sec = ceil(read_float("UPTIME_SEC"))
Int cpu_load_15min = ceil(read_float("LOAD_15M"))
String augur_version = read_string("VERSION")
}
}
Expand Down Expand Up @@ -215,19 +223,23 @@ task augur_mask_sites {
else
cp "~{sequences}" "~{out_fname}"
fi
cat /proc/uptime | cut -f 1 -d ' ' > UPTIME_SEC
cat /proc/loadavg | cut -f 3 -d ' ' > LOAD_15M
cat /sys/fs/cgroup/memory/memory.max_usage_in_bytes > MEM_BYTES
}
runtime {
docker: docker
memory: "3 GB"
cpu : 2
memory: "2 GB"
cpu : 1
disks: "local-disk 100 HDD"
preemptible: 2
preemptible: 1
dx_instance_type: "mem1_ssd1_v2_x2"
}
output {
File masked_sequences = out_fname
Int max_ram_gb = ceil(read_float("MEM_BYTES")/1000000000)
Int runtime_sec = ceil(read_float("UPTIME_SEC"))
Int cpu_load_15min = ceil(read_float("LOAD_15M"))
String augur_version = read_string("VERSION")
}
}
Expand All @@ -246,8 +258,6 @@ task draft_augur_tree {
File? vcf_reference
String? tree_builder_args

Int? machine_mem_gb
Int? disk_space_gb = 750
String docker = "nextstrain/base"
}
parameter_meta {
Expand All @@ -272,10 +282,10 @@ task draft_augur_tree {
}
runtime {
docker: docker
memory: select_first([machine_mem_gb, 30]) + " GB"
cpu : 16
disks: "local-disk ${disk_space_gb} LOCAL"
dx_instance_type: "mem1_ssd1_v2_x16"
memory: "16 GB"
cpu : 32
disks: "local-disk 750 LOCAL"
dx_instance_type: "mem1_ssd1_v2_x36"
preemptible: 0
}
output {
Expand Down Expand Up @@ -313,8 +323,6 @@ task refine_augur_tree {
String? divergence_units
File? vcf_reference

Int? machine_mem_gb
Int? disk_space_gb = 750
String docker = "nextstrain/base"
}
parameter_meta {
Expand Down Expand Up @@ -353,10 +361,10 @@ task refine_augur_tree {
}
runtime {
docker: docker
memory: select_first([machine_mem_gb, 30]) + " GB"
cpu : 16
disks: "local-disk ${disk_space_gb} LOCAL"
dx_instance_type: "mem1_ssd1_v2_x16"
memory: "13 GB"
cpu : 2
disks: "local-disk 100 HDD"
dx_instance_type: "mem3_ssd1_v2_x2"
preemptible: 0
}
output {
Expand All @@ -383,7 +391,6 @@ task ancestral_traits {
File? weights
Float? sampling_bias_correction

Int? machine_mem_gb
String docker = "nextstrain/base"
}
command {
Expand All @@ -394,20 +401,25 @@ task ancestral_traits {
--columns ~{sep=" " columns} \
--output-node-data "~{basename}_nodes.json" \
~{"--weights " + weights} \
~{"--sampling-bias-correction " + sampling_bias_correction} \
~{true="--confidence" false="" confidence}
cat /proc/uptime | cut -f 1 -d ' ' > UPTIME_SEC
cat /proc/loadavg | cut -f 3 -d ' ' > LOAD_15M
cat /sys/fs/cgroup/memory/memory.max_usage_in_bytes > MEM_BYTES
}
runtime {
docker: docker
memory: select_first([machine_mem_gb, 3]) + " GB"
cpu : 2
disks: "local-disk 50 HDD"
memory: "2 GB"
cpu : 1
disks: "local-disk 50 HDD"
dx_instance_type: "mem1_ssd1_v2_x2"
preemptible: 2
preemptible: 1
}
output {
File node_data_json = "~{basename}_nodes.json"
Int max_ram_gb = ceil(read_float("MEM_BYTES")/1000000000)
Int runtime_sec = ceil(read_float("UPTIME_SEC"))
Int cpu_load_15min = ceil(read_float("LOAD_15M"))
String augur_version = read_string("VERSION")
}
}
Expand All @@ -428,7 +440,6 @@ task ancestral_tree {
File? vcf_reference
File? output_vcf

Int? machine_mem_gb
String docker = "nextstrain/base"
}
parameter_meta {
Expand Down Expand Up @@ -456,11 +467,11 @@ task ancestral_tree {
}
runtime {
docker: docker
memory: select_first([machine_mem_gb, 7]) + " GB"
cpu : 4
disks: "local-disk 50 HDD"
dx_instance_type: "mem1_ssd1_v2_x4"
preemptible: 2
memory: "13 GB"
cpu : 2
disks: "local-disk 50 HDD"
dx_instance_type: "mem3_ssd1_v2_x2"
preemptible: 1
}
output {
File nt_muts_json = "~{basename}_nt_muts.json"
Expand All @@ -486,7 +497,6 @@ task translate_augur_tree {
File? vcf_reference_output
File? vcf_reference

Int? machine_mem_gb
String docker = "nextstrain/base"
}
command {
Expand All @@ -502,11 +512,11 @@ task translate_augur_tree {
}
runtime {
docker: docker
memory: select_first([machine_mem_gb, 3]) + " GB"
cpu : 2
memory: "2 GB"
cpu : 1
disks: "local-disk 50 HDD"
dx_instance_type: "mem1_ssd1_v2_x2"
preemptible: 2
preemptible: 1
}
output {
File aa_muts_json = "~{basename}_aa_muts.json"
Expand Down Expand Up @@ -541,11 +551,11 @@ task assign_clades_to_nodes {
}
runtime {
docker: docker
memory: "3 GB"
cpu : 2
memory: "2 GB"
cpu : 1
disks: "local-disk 50 HDD"
dx_instance_type: "mem1_ssd1_v2_x2"
preemptible: 2
preemptible: 1
}
output {
File node_clade_data_json = "~{out_basename}_node-clade-assignments.json"
Expand Down Expand Up @@ -580,20 +590,24 @@ task augur_import_beast {
~{"--tip-date-regex " + tip_date_regex} \
~{"--tip-date-format " + tip_date_format} \
~{"--tip-date-delimeter " + tip_date_delimiter}
Int max_ram_gb = ceil(read_float("MEM_BYTES")/1000000000)
cat /proc/uptime | cut -f 1 -d ' ' > UPTIME_SEC
cat /proc/loadavg | cut -f 3 -d ' ' > LOAD_15M
cat /sys/fs/cgroup/memory/memory.max_usage_in_bytes > MEM_BYTES
}
runtime {
docker: docker
memory: select_first([machine_mem_gb, 3]) + " GB"
cpu : 2
disks: "local-disk 50 HDD"
dx_instance_type: "mem1_ssd1_v2_x2"
preemptible: 2
preemptible: 1
}
output {
File tree_newick = "~{tree_basename}.nwk"
File node_data_json = "~{tree_basename}.json"
Int max_ram_gb = ceil(read_float("/sys/fs/cgroup/memory/memory.max_usage_in_bytes")/1000000000)
Int runtime_sec = ceil(read_float("UPTIME_SEC"))
Int cpu_load_15min = ceil(read_float("LOAD_15M"))
String augur_version = read_string("VERSION")
}
}
Expand All @@ -616,7 +630,6 @@ task export_auspice_json {
Array[String]? maintainers
String? title

Int? machine_mem_gb
String docker = "nextstrain/base"
}
String out_basename = basename(basename(tree, ".nwk"), "_refined_tree")
Expand Down Expand Up @@ -665,19 +678,23 @@ task export_auspice_json {
~{"--colors " + colors_tsv} \
~{"--description " + description_md} \
--output ~{out_basename}_auspice.json)
cat /proc/uptime | cut -f 1 -d ' ' > UPTIME_SEC
cat /proc/loadavg | cut -f 3 -d ' ' > LOAD_15M
cat /sys/fs/cgroup/memory/memory.max_usage_in_bytes > MEM_BYTES
}
runtime {
docker: docker
memory: select_first([machine_mem_gb, 3]) + " GB"
memory: "3 GB"
cpu : 2
disks: "local-disk 100 HDD"
dx_instance_type: "mem1_ssd1_v2_x2"
preemptible: 2
preemptible: 0
}
output {
File virus_json = "~{out_basename}_auspice.json"
Int max_ram_gb = ceil(read_float("MEM_BYTES")/1000000000)
Int runtime_sec = ceil(read_float("UPTIME_SEC"))
Int cpu_load_15min = ceil(read_float("LOAD_15M"))
String augur_version = read_string("VERSION")
}
}