Parallelize Bash script with maximum number of processes
BashBash Problem Overview
Lets say I have a loop in Bash:
for foo in `some-command`
do
do-something $foo
done
do-something
is cpu bound and I have a nice shiny 4 core processor. I'd like to be able to run up to 4 do-something
's at once.
The naive approach seems to be:
for foo in `some-command`
do
do-something $foo &
done
This will run all do-something
s at once, but there are a couple downsides, mainly that do-something may also have some significant I/O which performing all at once might slow down a bit. The other problem is that this code block returns immediately, so no way to do other work when all the do-something
s are finished.
How would you write this loop so there are always X do-something
s running at once?
Bash Solutions
Solution 1 - Bash
Depending on what you want to do xargs also can help (here: converting documents with pdf2ps):
cpus=$( ls -d /sys/devices/system/cpu/cpu[[:digit:]]* | wc -w )
find . -name \*.pdf | xargs --max-args=1 --max-procs=$cpus pdf2ps
From the docs:
--max-procs=max-procs
-P max-procs
Run up to max-procs processes at a time; the default is 1.
If max-procs is 0, xargs will run as many processes as possible at a
time. Use the -n option with -P; otherwise chances are that only one
exec will be done.
Solution 2 - Bash
With GNU Parallel http://www.gnu.org/software/parallel/ you can write:
some-command | parallel do-something
GNU Parallel also supports running jobs on remote computers. This will run one per CPU core on the remote computers - even if they have different number of cores:
some-command | parallel -S server1,server2 do-something
A more advanced example: Here we list of files that we want my_script to run on. Files have extension (maybe .jpeg). We want the output of my_script to be put next to the files in basename.out (e.g. foo.jpeg -> foo.out). We want to run my_script once for each core the computer has and we want to run it on the local computer, too. For the remote computers we want the file to be processed transferred to the given computer. When my_script finishes, we want foo.out transferred back and we then want foo.jpeg and foo.out removed from the remote computer:
cat list_of_files | \
parallel --trc {.}.out -S server1,server2,: \
"my_script {} > {.}.out"
GNU Parallel makes sure the output from each job does not mix, so you can use the output as input for another program:
some-command | parallel do-something | postprocess
See the videos for more examples: https://www.youtube.com/playlist?list=PL284C9FF2488BC6D1
Solution 3 - Bash
maxjobs=4
parallelize () {
while [ $# -gt 0 ] ; do
jobcnt=(maxjobs=4
parallelize () {
while [ $# -gt 0 ] ; do
jobcnt=(jobs -p)
if [ ${#jobcnt[@]} -lt $maxjobs ] ; then
do-something $1 &
shift
else
sleep 1
fi
done
wait
}
parallelize arg1 arg2 "5 args to third job" arg4 ...
)
if [ ${#jobcnt[@]} -lt $maxjobs ] ; then
do-something $1 &
shift
else
sleep 1
fi
done
wait
}
parallelize arg1 arg2 "5 args to third job" arg4 ...
Solution 4 - Bash
Here an alternative solution that can be inserted into .bashrc and used for everyday one liner:
function pwait() {
while [ $(jobs -p | wc -l) -ge $1 ]; do
sleep 1
done
}
To use it, all one has to do is put &
after the jobs and a pwait call, the parameter gives the number of parallel processes:
for i in *; do
do_something $i &
pwait 10
done
It would be nicer to use wait
instead of busy waiting on the output of jobs -p
, but there doesn't seem to be an obvious solution to wait till any of the given jobs is finished instead of a all of them.
Solution 5 - Bash
Instead of a plain bash, use a Makefile, then specify number of simultaneous jobs with make -jX
where X is the number of jobs to run at once.
Or you can use wait
("man wait
"): launch several child processes, call wait
- it will exit when the child processes finish.
maxjobs = 10
foreach line in `cat file.txt` {
jobsrunning = 0
while jobsrunning < maxjobs {
do job &
jobsrunning += 1
}
wait
}
job ( ){
...
}
If you need to store the job's result, then assign their result to a variable. After wait
you just check what the variable contains.
Solution 6 - Bash
Maybe try a parallelizing utility instead rewriting the loop? I'm a big fan of xjobs. I use xjobs all the time to mass copy files across our network, usually when setting up a new database server. <http://www.maier-komor.de/xjobs.html>
Solution 7 - Bash
If you're familiar with the make
command, most of the time you can express the list of commands you want to run as a a makefile. For example, if you need to run $SOME_COMMAND on files *.input each of which produces *.output, you can use the makefile
INPUT = a.input b.input OUTPUT = $(INPUT:.input=.output)%.output : %.input $(SOME_COMMAND) $< $@
all: $(OUTPUT)
and then just run
make -j<NUMBER>
to run at most NUMBER commands in parallel.
Solution 8 - Bash
While doing this right in bash
is probably impossible, you can do a semi-right fairly easily. bstark
gave a fair approximation of right but his has the following flaws:
- Word splitting: You can't pass any jobs to it that use any of the following characters in their arguments: spaces, tabs, newlines, stars, question marks. If you do, things will break, possibly unexpectedly.
- It relies on the rest of your script to not background anything. If you do, or later you add something to the script that gets sent in the background because you forgot you weren't allowed to use backgrounded jobs because of his snippet, things will break.
Another approximation which doesn't have these flaws is the following:
scheduleAll() {
local job i=0 max=4 pids=()
for job; do
(( ++i % max == 0 )) && {
wait "${pids[@]}"
pids=()
}
bash -c "$job" & pids+=("$!")
done
wait "${pids[@]}"
}
Note that this one is easily adaptable to also check the exit code of each job as it ends so you can warn the user if a job fails or set an exit code for scheduleAll
according to the amount of jobs that failed, or something.
The problem with this code is just that:
- It schedules four (in this case) jobs at a time and then waits for all four to end. Some might be done sooner than others which will cause the next batch of four jobs to wait until the longest of the previous batch is done.
A solution that takes care of this last issue would have to use kill -0
to poll whether any of the processes have disappeared instead of the wait
and schedule the next job. However, that introduces a small new problem: you have a race condition between a job ending, and the kill -0
checking whether it's ended. If the job ended and another process on your system starts up at the same time, taking a random PID which happens to be that of the job that just finished, the kill -0
won't notice your job having finished and things will break again.
A perfect solution isn't possible in bash
.
Solution 9 - Bash
function for bash:
parallel ()
{
awk "BEGIN{print \"all: ALL_TARGETS\\n\"}{print \"TARGET_\"NR\":\\n\\t@-\"\$0\"\\n\"}END{printf \"ALL_TARGETS:\";for(i=1;i<=NR;i++){printf \" TARGET_%d\",i};print\"\\n\"}" | make $@ -f - all
}
using:
cat my_commands | parallel -j 4
Solution 10 - Bash
Really late to the party here, but here's another solution.
A lot of solutions don't handle spaces/special characters in the commands, don't keep N jobs running at all times, eat cpu in busy loops, or rely on external dependencies (e.g. GNU parallel
).
With inspiration for dead/zombie process handling, here's a pure bash solution:
function run_parallel_jobs {
local concurrent_max=$1
local callback=$2
local cmds=("${@:3}")
local jobs=( )
while [[ "${#cmds[@]}" -gt 0 ]] || [[ "${#jobs[@]}" -gt 0 ]]; do
while [[ "${#jobs[@]}" -lt $concurrent_max ]] && [[ "${#cmds[@]}" -gt 0 ]]; do
local cmd="${cmds[0]}"
cmds=("${cmds[@]:1}")
bash -c "$cmd" &
jobs+=($!)
done
local job="${jobs[0]}"
jobs=("${jobs[@]:1}")
local state="$(ps -p $job -o state= 2>/dev/null)"
if [[ "$state" == "D" ]] || [[ "$state" == "Z" ]]; then
$callback $job
else
wait $job
$callback $job $?
fi
done
}
And sample usage:
function job_done {
if [[ $# -lt 2 ]]; then
echo "PID $1 died unexpectedly"
else
echo "PID $1 exited $2"
fi
}
cmds=( \
"echo 1; sleep 1; exit 1" \
"echo 2; sleep 2; exit 2" \
"echo 3; sleep 3; exit 3" \
"echo 4; sleep 4; exit 4" \
"echo 5; sleep 5; exit 5" \
)
# cpus="$(getconf _NPROCESSORS_ONLN)"
cpus=3
run_parallel_jobs $cpus "job_done" "${cmds[@]}"
The output:
1
2
3
PID 56712 exited 1
4
PID 56713 exited 2
5
PID 56714 exited 3
PID 56720 exited 4
PID 56724 exited 5
For per-process output handling $$
could be used to log to a file, for example:
function job_done {
cat "$1.log"
}
cmds=( \
"echo 1 \$\$ >\$\$.log" \
"echo 2 \$\$ >\$\$.log" \
)
run_parallel_jobs 2 "job_done" "${cmds[@]}"
Output:
1 56871
2 56872
Solution 11 - Bash
The project I work on uses the wait command to control parallel shell (ksh actually) processes. To address your concerns about IO, on a modern OS, it's possible parallel execution will actually increase efficiency. If all processes are reading the same blocks on disk, only the first process will have to hit the physical hardware. The other processes will often be able to retrieve the block from OS's disk cache in memory. Obviously, reading from memory is several orders of magnitude quicker than reading from disk. Also, the benefit requires no coding changes.
Solution 12 - Bash
This might be good enough for most purposes, but is not optimal.
#!/bin/bash
n=0
maxjobs=10
for i in *.m4a ; do
# ( DO SOMETHING ) &
# limit jobs
if (( $(($((++n)) % $maxjobs)) == 0 )) ; then
wait # wait until all have finished (not optimal, but most times good enough)
echo $n wait
fi
done
Solution 13 - Bash
Here is how I managed to solve this issue in a bash script:
#! /bin/bash
MAX_JOBS=32
FILE_LIST=($(cat ${1}))
echo Length ${#FILE_LIST[@]}
for ((INDEX=0; INDEX < ${#FILE_LIST[@]}; INDEX=$((${INDEX}+${MAX_JOBS})) ));
do
JOBS_RUNNING=0
while ((JOBS_RUNNING < MAX_JOBS))
do
I=$((${INDEX}+${JOBS_RUNNING}))
FILE=${FILE_LIST[${I}]}
if [ "$FILE" != "" ];then
echo $JOBS_RUNNING $FILE
./M22Checker ${FILE} &
else
echo $JOBS_RUNNING NULL &
fi
JOBS_RUNNING=$((JOBS_RUNNING+1))
done
wait
done
Solution 14 - Bash
You can use a simple nested for loop (substitute appropriate integers for N and M below):
for i in {1..N}; do
(for j in {1..M}; do do_something; done & );
done
This will execute do_something N*M times in M rounds, each round executing N jobs in parallel. You can make N equal the number of CPUs you have.
Solution 15 - Bash
My solution to always keep a given number of processes running, keep tracking of errors and handle ubnterruptible / zombie processes:
function log {
echo "$1"
}
# Take a list of commands to run, runs them sequentially with numberOfProcesses commands simultaneously runs
# Returns the number of non zero exit codes from commands
function ParallelExec {
local numberOfProcesses="${1}" # Number of simultaneous commands to run
local commandsArg="${2}" # Semi-colon separated list of commands
local pid
local runningPids=0
local counter=0
local commandsArray
local pidsArray
local newPidsArray
local retval
local retvalAll=0
local pidState
local commandsArrayPid
IFS=';' read -r -a commandsArray <<< "$commandsArg"
log "Runnning ${#commandsArray[@]} commands in $numberOfProcesses simultaneous processes."
while [ $counter -lt "${#commandsArray[@]}" ] || [ ${#pidsArray[@]} -gt 0 ]; do
while [ $counter -lt "${#commandsArray[@]}" ] && [ ${#pidsArray[@]} -lt $numberOfProcesses ]; do
log "Running command [${commandsArray[$counter]}]."
eval "${commandsArray[$counter]}" &
pid=$!
pidsArray+=($pid)
commandsArrayPid[$pid]="${commandsArray[$counter]}"
counter=$((counter+1))
done
newPidsArray=()
for pid in "${pidsArray[@]}"; do
# Handle uninterruptible sleep state or zombies by ommiting them from running process array (How to kill that is already dead ? :)
if kill -0 $pid > /dev/null 2>&1; then
pidState=$(ps -p$pid -o state= 2 > /dev/null)
if [ "$pidState" != "D" ] && [ "$pidState" != "Z" ]; then
newPidsArray+=($pid)
fi
else
# pid is dead, get it's exit code from wait command
wait $pid
retval=$?
if [ $retval -ne 0 ]; then
log "Command [${commandsArrayPid[$pid]}] failed with exit code [$retval]."
retvalAll=$((retvalAll+1))
fi
fi
done
pidsArray=("${newPidsArray[@]}")
# Add a trivial sleep time so bash won't eat all CPU
sleep .05
done
return $retvalAll
}
Usage:
cmds="du -csh /var;du -csh /tmp;sleep 3;du -csh /root;sleep 10; du -csh /home"
# Execute 2 processes at a time
ParallelExec 2 "$cmds"
# Execute 4 processes at a time
ParallelExec 4 "$cmds"
Solution 16 - Bash
$DOMAINS = "list of some domain in commands"
for foo in some-command
do
eval `some-command for $DOMAINS` &
job[$i]=$!
i=$(( i + 1))
done
Ndomains=echo $DOMAINS |wc -w
for i in $(seq 1 1 $Ndomains) do echo "wait for ${job[$i]}" wait "${job[$i]}" done
in this concept will work for the parallelize. important thing is last line of eval is '&' which will put the commands to backgrounds.