Ultra Sound Relay

Application

Has the relay operator posted in the Call for Relays thread with the information required for application?

1. Is the relay operator publicly available?

Yes

2. Is the relay publicly listed and maintained? If yes, please link to the dashboard/webpage.

Yes (refer to Ultra Sound Relay (Mainnet) and Ultra Sound Relay (Goerli))

3. Is the relay open source? If yes, please link to the repository.

Yes, quoted reply below

1. we're running stock flashbots relay
2. custom frontend at https://relay.ultrasound.money
3. default frontend at https://relay.ultrasound.money/classic-ui
4. The custom frontend is available here (https://github.com/ultrasoundmoney/frontend/tree/main/src/relay).
5. We will shortly share the exact commit hashes we're running (for the relay and validation nodes) on a public README.
4. Are there any ethical or non-neutral considerations about the relay to disclose? This includes MEV strategies such as transaction reordering or censorship such as address filtering and sanctions/legal compliance.

The relay operator has not disclosed any such considerations and no non-neutral behavior has been observed during relay performance report compilation.

5. How does the relay source from builders and searchers? (e.g. are they internal, external 3rd parties, publicly open, private, permissioned, permissionless)?

“The relay is open to all builders to connect. We are not a builder ourselves so we do not source searchers.”

6. What is the relay operator’s policy for at-fault technical mishaps which result in missed proposals?

“We will make refunds up to 10 ETH. (For context, we are a small public good project with a tiny budget.)”

7. Do we have all the necessary disclosures and contact information about the relay operator?

Yes

Revenue

1. What is the revenue/profit sharing model (if any) of the relay?

“No revenue/profit sharing! The relay is free to use and will remain free.”

2. Is this model transparently disclosed and well understood?

N/A

Testing

1. Does the relay operator have a Goerli testnet relay?

Yes

2. How many days have node operators tested this relay? Has the relay been tested for a minimum of two weeks?

The relay has been tested by a substantial portion of Lido Node Operators on Goerli for > 2 weeks, please see

3. What was the success and failure rate of the relay’s proposed blocks?

No failed blocks were observed during the period.

4. Were there any incident or degraded performance reports from Lido Node Operators while testing the relay? (e.g. missed proposals due to missing/late payloads, validator registration errors, relay timeouts, etc.)

Some registration errors / timeouts were noticed but no missed proposals / late payloads, see

5. Were observed issues and incidents during testing investigated and resolved sufficiently in a timely manner?

From , re: validator registration timeouts Ultrasound is running a less powerful cluster on Goerli vs. mainnet due to cost constraints, meaning response times from the relay take longer. The validator registration endpoint timed out for some of these blocks.

6. Are there considerations and data to be observed and included from mainnet performance?

Based on relay monitoring and public data sources (e.g. https://www.rated.network/relays?network=mainnet and https://www.mevpanda.com/) no observations would lead a reasonable observer to believe that the relay does not operate well. A few instances of a relay reporting larger bid than actual amount delivered have been noted, though.

7. Has there been a report compiled to analyze the new relay with performance data on testnet and mainnet?

Yes, see

References/attachments (if not included in relay details)

• Follow up questions