Giter VIP home page Giter VIP logo

Comments (1)

solidiquis avatar solidiquis commented on May 20, 2024

Yeah I guess this is the opposite problem to what exa has. I originally did not want the --suppress-size option as this was meant to be a disk usage tool as well, but it became demanded enough to the point that I just thought "hey, why not?". The implementation that I approved definitely was just a monkey-patch as the ability to suppress-size and circumvent the overhead associated with how disk usage is calculated would be a bit more of a lift.

So for context on the traversal and disk usage algorithm:

  1. Traversal is done in parallel using WalkParallel. DirEntrys are transformed into Nodes and subsequently sent over a channel to another worker thread that is responsible for grouping children by their parent directory.
  2. Once traversal is complete and all of the DirEntrys are converted to Nodes, we assemble the virtual Tree and recursively traverse to compute disk usage.

The reason we need to recur in both steps is because all knowledge of filesystem hierarchies is gone once traverse in parallel. It theoretically increases throughput when dealing with filesystem I/O but then it's up to you to extrapolate your file structure again using the paths of each Node. With regard to erdtree, I felt as though compartmentalizing the traversal and disk usage calculation into the two steps above would yield good performance because you're not interweaving I/O-bound and CPU-bound workloads and you're allowing the former more costly step (in terms of time) to be done in parallel.

In a world where --suppress-size works perfectly in that you don't have to pay the overhead of double recursion (one for the file-system and one for the virtual Tree), I think we could just throw parallelism out the window and build the Tree in a single-threaded context i.e. use Walk instead of WalkParallel.

This of course is conjecture and I haven't had too much time to consider this super thoughtfully, but this is on my radar. Thanks a bunch for taking an interest in the project and making an issue out of this! I plan to do weekly minor releases for the foreseeable future and depending on how busy I am with work the fix for this will either be out this Sunday or the next. Let me know if you have any thoughts :)

from erdtree.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.