Future Proofing Full Node Infrastructure

Future Proofing Node Infrastructure: DIS Interview with Dmitry from Quiknode

Uncategorized

In this article, we chatted with Dmitry Shklovsky from QuikNode, one of the most well-known and reliable node services out there for Ethereum, about the future of full node infrastructure and how we can look at preparing for it.

Thank you for your time Dmitry, to set some context, what are some innovations that may disrupt the current state of Node Solutions? 

Chain data storage optimizations, better compression. New, less resource-intensive clients and/or optimization of existing clients… all leading to “lighter” full node clients — some piece of software one can easily run in the background on their laptop (and eventually their phone, in their pocket 😉

There will continue to be a use-case for business and enterprise services — as they’ll need reliable and high-performance access to blockchain data, for which they will open towards leveraging a trusted third-party.

How do you foresee these innovations leading to lighter full nodes, and the need for enterprise affecting the viability of different node setups and operations? Meaning, if the goal is for everyone to run nodes, whether as a service, or on their phone, how should can they prepare today? 

The issue with running a Full or Archive Ethereum node today is that it requires a relatively powerful machine — it’s not something one can run in the background on their laptop without noticing a significant decrease in performance. Then there’s also the technical barrier, as the node clients can be too much for the average user to set up and configure (not to mention they are at times cranky and require maintenance). Also, not everyone has a spare 300+ GB SSD, which is needed to satisfy the i/o of keeping the chain in-sync at all times (or the CPU cycles, or 8+ GB of RAM). Optimizing chain storage, reducing resource usage by the node client will help improve adoption. That, or hardware will become more powerful and less expensive, so the node requirements will become moot =)

What are the current problems that, if left unattended, will break decentralized node infra?

Censorship, particularly bandwidth censorship (e.g. firewall). It’s a problem of the Internet, really. For example, a country can order all its ISPs to block an IP range or filter certain traffic, preventing data (e.g. chain/block data) from exiting or entering. We’ve seen cases where a developer from country X was not able to access a node hosted in country Y, but was able to access one hosted in country Z. Decentralization and global, redundant distribution of core infrastructure will definitely help in addressing these types of issues.

How are you approaching solving this? Are you aware of other solutions/approaches in the space to address this bandwidth censorship concern?  

We offer users the ability to launch a node in 8 locations around the world. So whether it be for better performance or to work around a networking issue, a globally-distributed setup helps. In some cases we had to contact the ISP and work with them to remove blocks & blacklisted IPs.

So,  If you had unlimited funds, and a 25-year roadmap to build out the decentralized network of the future, what would you do? 

Global-scale peer-to-peer meshing. All network-enabled devices should be able to communicate with and through each other, trustlessly. 

That Netflix video you watched on your laptop a couple of hours ago? My phone should be able to fetch it from you as long as we’re within communication proximity (and probably reward you for it), and we shouldn’t have any clue as to what each device is or who it belongs to (i.e. totally anonymous). 

Similarly, if I were to WhatsApp someone, my message should be able to get to where it’s going through other devices, anonymously. Every device essentially becomes a ‘node’ or ‘relay’ in the network. I believe there have been and are (several) projects working with this idea in mind, in this direction.

Any of these projects come to mind? 

althea.org

rightmesh.io

gotennamesh.com

What can we look forward to from QuikNode? How will this fit into the broader “Decentralized Infrastructure Solutions” that the whole space is striving for? 

We’re working to create more choices for blockchain node hosting, global distribution, and powerful tooling for Web3 developers. Also looking forward to our continued partnership with POKT to achieve a greater level of decentralization for the blockchain ecosystem!

Great Stuff! Thank you for sharing your thoughts on the future of node infrastructure.