Ravendex Stake Pool: Infrastructure – Cardano Feed

With the start of our ISPO, Ravendex had one mission in ideas, to assemble a Cardano Stake pool infrastructure that’d be protected, reliable, and provide top-class effectivity.

Though the pool’s current foremost aim is to assist the enterprise with the ISPO, the final phrase objective is to produce delegators with fantastic service as we switch earlier it. We moreover contemplate in contributing to the group, following all the on the market best practices.

With this objective in ideas, Ravendex has invested in a redundant, extreme availability, high-performance infrastructure that’s geographically distributed to cut back latency and publicity to data coronary heart incidents.

The outcome’s a pool that’s presently displaying a very low number of missed slots (efficiently 0 apart from the anticipated small number of missed slots all through epoch transition).

As the amount of delegators grows and, with it, the full of life stake will improve, assigned blocks may improve, as will the respective rewards, making them moreover further safe on a per epoch basis.


The phrases underneath are used inside the following article:

  • Block producer: the node that validates transactions and mints blocks;
  • Relay: the node that propagates blocks to the block producer and the rest of the group;
  • DS: devoted server, an event that’s solely speculated to run a Cardano node, not sharing belongings with one other corporations;
  • VPS: digital private server, an event that shares belongings with completely different VPS conditions, engaged on an underlying bodily server.

Node deployment

The block producers and relays are deployed in conditions all through numerous cloud suppliers.

For the precept conditions, devoted servers had been chosen so as to not have shocking impacts on the effectivity of the nodes, which usually happen when cloud suppliers over-allocate belongings to servers working VPS conditions.

This effectivity affect typically occurs at situations when an elevated computation power is required (epoch change, as an example) which can then translate to inconsistent, unreliable effectivity.

The precept producer and relay conditions embody DSs with 8 core Xeon CPUs and 64GBs of RAM, whereas secondary node conditions are engaged on lower-cost VPS conditions with in any case 6 CPU cores and 16GBs of RAM.

The construction accommodates:

  • One launched relay;
  • Two unannounced relays (so as to mitigate any eventual DDoS assaults and improve resiliency);
  • One on-line block producer and a standby block producer that’s reworked to foremost block producer in case the distinctive is found to be offline for any given objective.


Various aspects are thought-about to make the operation as protected as potential:

  • ● All nodes are working Debian Bullseye and all software program program packages are generally updated so as to cut back the chances of working software program program with vulnerabilities.
  • ● All nodes are firewalled, decreasing incoming connections to those strictly compulsory.
  • ● Block producer conditions are shielded from most of the people group, talking solely with the relays. No recordsdata are saved inside the BP other than these strictly needed for operation.
  • ● All nodes are generally backed up by the use of the creation of snapshots with an relevant retention interval, which can be recovered in case of need. Nodes are moreover configured in a very modular method, which makes it easy to spawn new conditions in a matter of minutes in case of need.


When any compulsory transactions are to be executed, they’re on a regular basis signed in an air-gapped offline event, that’s saved securely saved and backed up generally. This accommodates the frequent KES key rotation.

On this way, pool keys are saved protected and absolutely away from any group connection.


All relay nodes are generally checked for any latency factors to linked nodes, and adjustments are made in case of need. All relays are saved with spherical 20 connections to completely different group nodes.

What comes subsequent?

Because the ISPO progresses and full of life stake will improve, new swimming swimming pools could also be deployed to avoid saturation and diminished rewards.

In response to neighborhood ideas, Ravendex may resolve to open additional 100% RAVE swimming swimming pools or 50% RAVE swimming swimming pools.

Going into the long run, and since the ISPO includes an end, Ravendex swimming swimming pools will try to proceed providing a aggressive high-quality service with diminished expenses, bringing price to delegators that choose to keep up staking with us.

Observe Ravendex On Social Media

Web web site: https://ravendex.io

Twitter: https://twitter.com/Ravendexlabs

Telegram: https://t.me/RaversNest

Documentation: https://docs.ravendex.io/


Related Posts