Update milestones for 2024-10 (#120)

This commit is contained in:
Benjamin Arntzen 2024-09-18 15:23:42 +01:00 committed by GitHub
parent acd0329a6b
commit 7a7dc35ab8
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 38 additions and 14 deletions

View File

@ -26,7 +26,7 @@ gantt
Midscale: 2024-05-01, 2024-11-01
```
- status: 20%
- status: 40%
- CC: Wings
### Description
@ -40,13 +40,23 @@ Testing is to be done in this order of priority:
4. Test Store protocol at scale
5. Test Waku relay+store reliability with nodes going offline/online
- If nodes go online/offline, we should be able to retrieve missing messages from the store. This will also test Waku message relaying in a different way.
6. Filter and lightpush tests
7. Measure (1) and (3) in heterogenous clusters involving different node implementations such as nwaku and go-waku
8. Measure (3) with Waku peer exchange protocol used for discovery by a subset of nodes.
9. Measure (1) with a mix of nodes using Resource-restricted device reliability protocol and peer exchange, meaning a small number of nwaku nodes serve store, light push and filter protocols and a high number of clients consume them. For example, 6-10 service nodes, 200 relay nodes and 1000 light nodes.
6. Measure (1) and (3) in heterogenous clusters involving different node implementations such as nwaku and go-waku
7. Test waku shard behaviour and stability with various of numbers of shards
8. Filter and lightpush tests
9. Measure (3) with Waku peer exchange protocol used for discovery by a subset of nodes.
10. Measure (1) with a mix of nodes using Resource-restricted device reliability protocol and peer exchange, meaning a small number of nwaku nodes serve store, light push and filter protocols and a high number of clients consume them. For example, 6-10 service nodes, 200 relay nodes and 1000 light nodes.
This should include connection and node churn impact on reliability for both relay and light clients.
Additionally, perform monthly regression tests against a chosen version of Waku to ensure that no new bugs have been introduced. Produce a report on the results of the tests and any interesting findings.
### Justification
Provide a greater understanding of Waku's performance and reliability in midsized networks.
### Deliverables
Similar deliverables to 10k sim, but with focuses on smaller scale and more frequent deployments.
Frequent deployments of 1000-5000 node networks.
Documentation of both the deployment process and actual deployments.
Useful analytics for the Waku team that can be used to improve the Waku software.
Research articles such as blog posts about findings from the midscale deployments.

View File

@ -26,20 +26,23 @@ gantt
Deployer Tool: 2024-05-01, 2024-11-01
```
- status: 80%
- CC: Alberto
- status: 90%
- CC: Alberto, Wings
### Description
A first version of tool that allows deploying >10k gossipsub / waku relay nodes.
The tool should measure bandwidth usage per node and bundle the measurement data for analaysis.
The tool should measure bandwidth usage per node and bundle the measurement data for analysis.
The tool should be built in such a way that it can be used for other deployments as well.
It should allow automated, repeatable and accountable deployments.
### Justification
### Deliverables
* https://github.com/vacp2p/10ksim
* https://github.com/vacp2p/vaclab/argocd

View File

@ -26,21 +26,32 @@ gantt
Visualiser Tool: 2024-05-01, 2024-11-01
```
- status: 10%
- CC: Alberto
- status: 75%
- CC: Alberto, Alberto
### Description
A first version of tool that allows for visualising the message flow of a Waku network. It should be adaptable to other network types too (particularly Nomos, Codex)
The visualiser tools are two tools that can be used for visualising the message flow of a Waku network. They are adaptable to other network types too (particularly Nomos, Codex).
This relies on a Grafana Loki deployment to store and query logs.
They rely on either Grafana Loki or VictoriaLogs to store and query logs.
The live visualiser is used for viewing the network in real time.
The debug visualiser is used for viewing a deployment that has already taken place.
### Justification
To make it easy and intuitive to understand the message flow and propagation patterns and properties of a Waku network and apply that same understanding to other networks.
### Deliverables
A peer to peer network mapper that creates a visualisation something like this:
![alt text](image.png)
[![Visualiser Tool](visualiser-tool.png)](visualiser-tool.png)
The tool should be able to visualise the message flow of a Waku network, by lighting up nodes in a graph as they receive messages, flashing a different colour for each message (or message type).
The tool should be able to visualise the message flow of a Waku network, by lighting up nodes in a graph as they receive messages, flashing a different colour for each message (or message type).
The live visualiser is feature complete, needing only minor tweaks and bug fixes:
https://github.com/vacp2p/dst-live-visualiser
The debug visualiser is still under development, but the core functionality is available already.

Binary file not shown.

Before

Width:  |  Height:  |  Size: 506 KiB

After

Width:  |  Height:  |  Size: 206 KiB