Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Where is the Elf algorithm reproduced? #25

Closed
LeeSure5986 opened this issue Nov 28, 2024 · 2 comments
Closed

Where is the Elf algorithm reproduced? #25

LeeSure5986 opened this issue Nov 28, 2024 · 2 comments

Comments

@LeeSure5986
Copy link

Hello,

I have successfully executed the script and reproduced the experimental results. However, the experimental result figure figure_1_Speed.png does not include the results of the Elf algorithm, nor did I find any code for reproducing the Elf algorithm in your projects. Could you please provide the code for reproducing the Elf algorithm?

Best regards,
Li Shuo

@lkuffo
Copy link
Collaborator

lkuffo commented Nov 28, 2024

Hi,
To benchmark Elf, we ran experiments using their codebase. We didn't have the time to re-implement Elf in C++ (as Elf was published just a few days before the deadline to submit ours).

Currently, we are trying to incorporate Elf Java code into our codebase to achieve full reproducibility of the results reported in the ALP paper, however, it is still WIP.

@lkuffo
Copy link
Collaborator

lkuffo commented Dec 6, 2024

We have added Elf benchmarks (both speed and compression ratio) to our reproducibility suite.

To run Elf benchmarks you must have Java 1.8 installed and Maven pointing to that Java version. Then, you can run Elf benchmarks as a standalone test:

mvn -f "./publication/source_code/extern/elf/pom.xml" test

After the test finish you will see the results under publication/compression_ratio_result/double/elf.csv for compression ratios, and publication/results/i4i_4xlarge/elf_raw.csv for micro-benchmarks of decompression/compression speed (measured in tuples processed per CPU cycle).

To incorporate Elf we took the branch of their code which we used for our initial benchmarks, added a FileReader to be able to read binary files of doubles and changed the BlockSize from 1000 to 1024. PR: #30

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants