Skip to content

Commit

Permalink
SoC-2020-Ideas: Check intervals labeling for improvements first
Browse files Browse the repository at this point in the history
Christian Couder asked if there is an easy way to check that it would
provide significant performance improvements at least in some cases,
and whether can we ask the student to do that at the beginning of the
GSoC 2020.

The "Reachability labels for version control graphs.ipynb" Jupyter
Notebook on Google Colaboratory was created to answer this question
(originally for the FELINE reachability index).  Among others it can
min-post interval labels and topological levels (generation numbers),
use them for reachability queries, and load Linux kernel
commit-graph. The exploration didn't get finished, but it would be not
that difficult, I think, to at least find the amount of false negatives
for min-post interval labeling for git.git or Linux kernel repo.

  https://colab.research.google.com/drive/1V-U7_slu5Z3s5iEEMFKhLXtaxSu5xyzg

As Jupyter Notebook, it is run in the web browser.  It can either use
local runtime, or run on Google Cloud runtime.  On the other hand it
requires at least some knowledge of Python...
  • Loading branch information
jnareb committed Mar 18, 2020
1 parent a6d5982 commit ac526b1
Showing 1 changed file with 11 additions and 0 deletions.
11 changes: 11 additions & 0 deletions SoC-2020-Ideas.md
Expand Up @@ -177,6 +177,17 @@ One can start with using this labeling for making selected single
command faster, for example `--contains` queries (as it was done for
generation numbers).

Before starting on this task, at the beginning of the GSoC, it might
be good idea to check that interval labels would provide significant
performance improvements at least in some cases (and if it is not the
case, switch to working on generation numbers v2). This _could_ be
done with the help of "Reachability labels for version control
graphs.ipynb" Jupyter Notebook on Google Colaboratory[[6][]]. It was
created to answer such questions, though the exploration didn't get
finished. It would be possible with this notebook to at least find
the amount of false negatives for min-post interval labeling for
git.git or Linux kernel repo.

Next task would be, time permitting, to make it possible to update the
labeling without recomputing it from scratch, and to make it
compatible with incremental update of the commit-graph file[[7][],[3][]].
Expand Down

0 comments on commit ac526b1

Please sign in to comment.