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

Throughput test #61

Closed
3 tasks
konradkonrad opened this issue Jun 27, 2016 · 7 comments
Closed
3 tasks

Throughput test #61

konradkonrad opened this issue Jun 27, 2016 · 7 comments

Comments

@konradkonrad
Copy link
Contributor

konradkonrad commented Jun 27, 2016

Tasklist

  • Measure throughput of a node sending mediated transactions through N channels without mediation.
  • The benchmark script must accept an argument for the minimum value for the TPS, and if the given value is not reached it should exit with a status of 1
    • The benchmark must be added to the build with the minimum value for TPS set.
@konradkonrad konradkonrad added this to the MVP milestone Jun 27, 2016
@konradkonrad konradkonrad self-assigned this Jun 27, 2016
@konradkonrad konradkonrad modified the milestones: Scalability PoC, MVP Jun 27, 2016
konradkonrad added a commit that referenced this issue Aug 31, 2016
konradkonrad added a commit that referenced this issue Aug 31, 2016
konradkonrad added a commit that referenced this issue Aug 31, 2016
konradkonrad added a commit that referenced this issue Aug 31, 2016
The playbook now installs everything necessary on a new image.
TODO:
- create DAG
- store as a new AMI (so launching multiple instances can be quicker)
konradkonrad added a commit that referenced this issue Aug 31, 2016
konradkonrad added a commit that referenced this issue Aug 31, 2016
konradkonrad added a commit that referenced this issue Sep 1, 2016
The default system python was 2.7.6 on the base AMI.
konradkonrad added a commit that referenced this issue Sep 1, 2016
There seems to be a bug in ansible, where the `editable` argument is not properly
inserted in the pip command. Explicitely adding it solves this.
konradkonrad added a commit that referenced this issue Sep 1, 2016
konradkonrad added a commit that referenced this issue Sep 1, 2016
WIP-note: dependencies on master are currently broken, so this is not
debuggable atm.
konradkonrad added a commit that referenced this issue Sep 1, 2016
konradkonrad added a commit that referenced this issue Sep 2, 2016
Installing ansible with dependencies leads to broken dependencies
for pydevp2p. By explicitely installing with `--no-deps` and excluding
ansibles cryptography modules, this can be avoided. Hopefully ansible
still works as expected...
konradkonrad added a commit that referenced this issue Sep 2, 2016
This adds a flag `--pretty` (default False) to `config_builder.py`
commands.
konradkonrad added a commit that referenced this issue Sep 2, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Oct 31, 2016
Also use `shell` module so we can resolve `$()` subcommands without
needing to read in new variables.
If a newer version of raiden should be tested, the variable `-e force_pull` can be set.
utzig pushed a commit to utzig/raiden that referenced this issue Oct 31, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Nov 1, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Nov 1, 2016
Also use `shell` module so we can resolve `$()` subcommands without
needing to read in new variables.
If a newer version of raiden should be tested, the variable `-e force_pull` can be set.
utzig pushed a commit to utzig/raiden that referenced this issue Nov 1, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Nov 3, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Nov 3, 2016
Also use `shell` module so we can resolve `$()` subcommands without
needing to read in new variables.
If a newer version of raiden should be tested, the variable `-e force_pull` can be set.
utzig pushed a commit to utzig/raiden that referenced this issue Nov 3, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Nov 4, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Nov 4, 2016
Also use `shell` module so we can resolve `$()` subcommands without
needing to read in new variables.
If a newer version of raiden should be tested, the variable `-e force_pull` can be set.
utzig pushed a commit to utzig/raiden that referenced this issue Nov 4, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Nov 17, 2016
utzig pushed a commit to utzig/raiden that referenced this issue Nov 17, 2016
Also use `shell` module so we can resolve `$()` subcommands without
needing to read in new variables.
If a newer version of raiden should be tested, the variable `-e force_pull` can be set.
utzig pushed a commit to utzig/raiden that referenced this issue Nov 17, 2016
@LefterisJP LefterisJP removed this from the Scalability Demo milestone Jan 12, 2017
@LefterisJP
Copy link
Contributor

@konradkonrad Is the throughput test viable in any way or form as of this moment? Should we try to revive it to measure throughput of our product before release so we can have proper benchmarks to support our claims?

@ulope
Copy link
Collaborator

ulope commented Jun 20, 2018

This has a lot of overlap with #652

@hackaugusto
Copy link
Contributor

hackaugusto commented Jun 20, 2018

@ulope yep, but I assume that #652 will not run on the build but on a testnet, correct?

@LefterisJP
Copy link
Contributor

@ulope Can we use the scenario player to measure throughput now?

@ulope
Copy link
Collaborator

ulope commented Oct 8, 2018

Yes, we just have to think about how we want to gather and export that data.

For gathering I see two options that can be added without much work:

  • Sum up the runtime of all transfer tasks and then divide by number of transfers
  • Add a benchmark parameter to the transfer task it so we have better control over what we measure

I'd prefer variant 2.

As for exporting, writing some form of json report should be enough.

@Dominik1999
Copy link
Contributor

@konradkonrad can we close this issue or should we somehow try to revive it?

hackaugusto pushed a commit to hackaugusto/raiden that referenced this issue Dec 5, 2019
@karlb
Copy link
Contributor

karlb commented Apr 7, 2022

Some scenarios can be considered to be throughput tests.

@karlb karlb closed this as completed Apr 7, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

8 participants