Conversation
A resource-agnostic (HPC or local) implementation for monolingual data analytics. In theory, this implementation should run seamlessly, the existence of HPC or not should only change the way HyperQueue is set up. In the case of local just starting a hq server and a hq worker. In the case of HPC, start a hq server and a hq allocation queue. All the procedures that were part of the map have been grouped into a single script. Parallelization should be line-based, preferably, not procedure-based, which makes it more complicated. The reduce has also been grouped into a single script, but each one on its own bash function, so all can be wrapped into the same pipeline: read concat each batch of the map, apply function, write to tmp, then move to permanent. readcorpus and readdocument positional arguments have been reodered, so the call can be just script.py <lang> and will use stdin/stdout by default. The singularity container to run each task has been added and the Dockerfile for it has some GPU stuff commented out for now (or not).
Everything in here should be now in 01.map and 02.reduce
Change input file naming Add missing srctok reduce
Fix path for intermediate files. Use HQ requested memory in sorts. Other small fixes
Not hpc-friendly having thousands of jobs trying to acquire .lock files
This allows using more ram.
The jobdefinition file now contains all the dependencies and tasks for a single lang. This way all tasks can be submitted without blocking. Also removes the need of having to open many windows, one for each lang.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.