## IDONTWANT bandwidth simulation results

The simulations are named after the blob target/limit and the percentage of nodes supporting Gossipsub 1.2.0 aka IDONTWANT. All simulations use the same base Lighthouse version, with modifications to be able to disable IDONTWANT. The 4/8 simulations additionally include changes to change the blob target/limit.

The table values are the average bandwidth used between the 5th and 15th minute after genesis, over the node category indicated in the table header. The "Diff" column indicates the difference between the averages of datacenter nodes and home nodes. The "Diff" rows indicate the difference between the preceding two rows.

Datacenter nodes have 1Gbit/s up and down and no additional latency.
Home nodes have 50Mbit/s up and down and 20ms additional latency.
The simulation contains 500 datacenter nodes and 500 home nodes.

| Simulation | All nodes | DC nodes | Home nodes | *Diff*
| ------------- | ----------- | ---------------- | ----------- | ----
| **3/6, 20%** | 90.6 KiB/s | 93.9 KiB/s | 87.4 KiB/s | *-6.5 KiB/s*
| **3/6, 100%** | 87.1 KiB/s | 91.6 KiB/s | 82.6 KiB/s | *-9 KiB/s*
| *Diff* | *-3.5 KiB/s<br>(-3,86%)* | *-2.3 KiB/s<br>(-2.45%)* | *-4.8 KiB/s<br>(-5.49%)*
|
| **4/8, 20%** | 117.8 KiB/s | 123 KiB/s | 112.7 KiB/s | *-10.3 KiB/s*
| **4/8, 100%** | 112.3 KiB/s | 118.3 KiB/s | 106.4 KiB/s | *-11.9 KiB/s*
| *Diff* | *-5.5 KiB/s<br>(-4.67%)* | *-4.7 KiB/s<br>(-3.82%)* | *-6.3 KiB/s<br>(-5.59%)*

In summary, extending IDONTWANT support to all nodes reduces used bandwidth slightly. This effect does not offset the additional bandwidth use increase introduced by the blob increase. Home nodes use less bandwidth than data center nodes (even though there is plenty of remaining available bandwidth) and increased usage of IDONTWANT seems to slightly increase that effect.

Note: currently, nodes are assigned to regions. Within their own region, nodes can communicate with rather low latency, with only slight correlation to the "size" of the region. For example, nodes in "europe" have 2ms of base delay, while nodes in "south_america" have 7ms of base delay. This seriously reduces realism, but is required to make the effects of IDONTWANT noticable. A proper fix might make regions more granular or use a coordinate based calculation of latency. See the notes on the [previous try](/jQeEeWCiRiC1M8uWvZH1pg) for more information.