You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It does not show (or link to) more complex examples, which are non-trivial to implement, such as:
memory { 16.GB + input_file.size() / 10.GB }
The above calculation actually does not work: ERROR ~ Unknown method invocation div on Long type
So, it would helpful to show examples of:
using input file sizes in the calculations
perform
Usage scenario
More complex dynamic resource allocation examples than just memory { 2.GB * task.attempt }.
Suggest implementation
More examples in the docs. I would create just a PR, but as you can see by my failed example above, I do not know how to actually perform some of the more complex resource allocation calculations.
General comment on docs
As a more general reason for including such docs: users and developers obtain their information from ChatGPT, Claude, etc. So, the more publicly available Nextflow docs, the more helpful these LLM services will be in regards to Nextflow.
For this reason, it is best to include more details than less.
If you are worried that the docs will be too cluttered for users to find the relevant docs, that issue is becoming less of a concern as users rely more on web search via ChatGPT, Claude (possible with MCP), Perplexity, etc. to find the specific relevant docs.
The text was updated successfully, but these errors were encountered:
New feature
https://www.nextflow.io/docs/latest/process.html#dynamic-task-resources provides just one example of dynamic resource calculations:
It does not show (or link to) more complex examples, which are non-trivial to implement, such as:
The above calculation actually does not work:
ERROR ~ Unknown method invocation
divon Long type
So, it would helpful to show examples of:
Usage scenario
More complex dynamic resource allocation examples than just
memory { 2.GB * task.attempt }
.Suggest implementation
More examples in the docs. I would create just a PR, but as you can see by my failed example above, I do not know how to actually perform some of the more complex resource allocation calculations.
General comment on docs
As a more general reason for including such docs: users and developers obtain their information from ChatGPT, Claude, etc. So, the more publicly available Nextflow docs, the more helpful these LLM services will be in regards to Nextflow.
For this reason, it is best to include more details than less.
If you are worried that the docs will be too cluttered for users to find the relevant docs, that issue is becoming less of a concern as users rely more on web search via ChatGPT, Claude (possible with MCP), Perplexity, etc. to find the specific relevant docs.
The text was updated successfully, but these errors were encountered: