Running a Global Blockchain Node Infrastructure Ecosystem: How We Do It

When people think of infrastructure, we often think of the roads we drive on, bridges over rivers, or electricity power grids. What do we mean when we talk about infrastructure when it comes to blockchain?

Blockchain infrastructure is basically the decentralized deployment of different blockchains, and the overlay network that sits on top of the physical infrastructure around the globe. All software has a dependency on one or more computers. Since most blockchains are distributed in nature, there is an added dependency on those computers to be part of a network. When we talk about computers nowadays, it is important to realize that this also includes mobile phones.

How does the blockchain infrastructure we support differ from traditional web 2.0 based infrastructure such as servers or databases?

Web 2 was about community and within those communities the sharing of information. This was facilitated by a client to server model; blogs, websites, online forums, and so on. Web 3 is about more focus on the individual and semantic information, live streams, and dynamic mashing of information on the fly, more tailored to the individual, rather than the community or group. Blockchain infrastructure is mostly trustless, meaning that a much lower level of trust is needed for individuals to participate, allowing a much more diverse pool of ways for the individual to create knowledge streams, through a more unrestricted selection of methods to bring together disparate points of data.

For the global blockchain infrastructure itself, do nodes rely on cloud-based web services for hosting, bare-metal servers, or both? What is the advantage of choosing one over the other?

At Blockdaemon, we use both. The advantage of the cloud is that the responsibility of managing the hardware and deployment infrastructure is somebody else’s problem, and as a business, you can focus your time and expertise on the software being deployed. However, if you have the skills within your organization to handle the infrastructure yourself, you are then able to optimize your deployments and have tighter control over your operational expenditure. Also, you may have customers with specific requirements about how and where software is deployed, and being able to do both allows you to provide services to a much broader range of customers, as we do at Blockdaemon.

What kind of network security is needed to support a node infrastructure ecosystem?

That’s a great question, and honestly, we could discuss the answer for a whole day. The key elements that come to mind are Authentication and Authorization, Network segmentation and using something called Zero Trust architecture. Authentication and Authorization are important to make sure that you know who or what is on the network and if they should be doing what they are asking to do. Network segmentation ensures that only the right entities are part of any particular part of the whole network, using things like firewalls and software-defined networks. Zero trust architecture is also needed to ensure that even if an attacker gets into a private network, they are still blocked from being able to do further damage.

What kind of metrics are important for monitoring the health of a node network?

It’s important to always know that nodes are always available, so we use health checks just for that. Connectivity of nodes to the rest of the network is also important and so we need to continually check for things like network segregation. Other things that can affect a node network are unwanted traffic, so it is important to monitor what the packets look like moving from one node to another. The speed of packet movement is also important so another metric that is important is network latency.

What measures are needed to ensure global node infrastructure is robust and resilient?

You need to take care to think about the interconnectivity of networks around the globe, and ensure that regardless of which country the user or machine is in, that there is enough capacity to serve the machine or user’s needs regardless of use-case. Networks are inherently unreliable and so with that in mind, there needs to be duplication of all your services so that as one becomes unavailable, another can automatically take its place.

What kind of people or companies rely on blockchain infrastructure?

If I can use Blockdaemon customers as a base-line, those customers range from individuals looking to have more control over their operations, to professional and academic groups doing research and analytics, scaling then, all the way up to large financial institutions who need enterprise-grade delivery and management of large clusters of blockchain infrastructure.

Why is access to this infrastructure needed for commercial applications?

Blockchain infrastructure access will allow further developments in several fields. Just a few that come to mind are Finance, Supply Chains, and Voting.

Financial applications rely on trust and access to blockchain infrastructure allows the development and deployment of next-generation applications to allow a wider audience to both participate and benefit. Supply chains also depend on high levels of trust. Blockchain infrastructure allows new trustless ways of bringing goods from the other side of the world in a trustless way, previously depending on building relationships over time and with high business risk. Voting is interesting too and has already been used in several places. One example is in Sierra Leone back in 2018 using the Agora blockchain.

What role do node networks play in keeping the blockchain ecosystem decentralized?

Each node that participates in any given network becomes part of the network, and as more nodes are added to the network, the node network organically spreads both geographically and in a custodial manner, ensuring that the node networks are more decentralized.

What are the common challenges with running a node infrastructure network at a global scale?

Networks fail continuously. Machines fail. Hard drives break. Software fails, and usually at the worst of times. These factors alone present significant challenges, making sure systems are resilient, that every node has a failover plan, that there is a disaster recovery plan in place, and all of this done in an automated way.

What are the advantages of operating nodes and backup nodes in multiple geographic locations?

We can provide a rapid service to customers, wherever they are in the world, and in case of a data center or network outage, there is an alternative solution in another location or another network. This is something that our growing team of engineers deals with on a daily basis.

For a globally accessible node ecosystem, what, if any, are the advantages of having nodes located in the geographical proximity of users and customers?

Having services located as close to the customer as possible means less network traffic and chances of disruption, lower latency meaning that the customer has a much better experience and in some cases, lower costs, for both customers and service providers.

Traditionally we have seen a lot of focus on supporting Proof of Work networks such as Bitcoin and Ethereum. Do you see a paradigm shift potentially occurring where Proof of Work nodes are overtaken by Proof of Stake nodes as the prevailing infrastructure on a global level or do you see coexistence?

There will definitely be room for both to coexist. Bitcoin uses proof of work, and is analogous to gold, being of comparatively high value compared to other currencies, yet difficult to transact at scale. As Proof-of-stake continues to evolve and grow, there will be increased ways to model businesses and financial tools. However, the increasing scarcity of Bitcoin will arguably keep its value climbing higher and different layers will allow the virtual trading of bitcoin at a high scale, rivaling that of today’s credit markets. At Blockdaemon, we are particularly interested in this scenario and as such, have a roadmap that aligns with part of our strategy that caters for both.

This article was written by Lloyd Moore, VP of Engineering at Blockdaemon