On the heels of our major update to the Pocket Portal, we’ve been sharing some of the key features that we’re bringing to the table, including a new Pay As You Go plan, other flexible service options for developers, POKT rewards that help eliminate sunk costs, low latency across any chain and region, and easy multichain service.
In this post, we’ll take a closer look at how the Pocket Portal enables maximum uptime for developers.
Outages amongst Web3 infrastructure providers remain a serious pain point for both developers and end users. Widespread, lengthy outages disrupt access to blockchain data and can make dApps completely inaccessible for broad swaths of users.
For centralized providers, the risk of an outage is significantly higher, as isolated points of failure are relied on for Web3 access. Over the years, we’ve seen a number of high-profile outages that have completely disrupted access to MetaMask and other applications, for example.
These outages can come from node updates gone wrong, heavy-handed regional censorship, cloud provider issues, and other sources.
Pocket has been tackling this issue through its decentralized approach to node infrastructure services, with the enablement of maximum uptime built into the structure of its protocol.
Optimized for Uptime
Pocket is able to achieve maximum uptime for dApps and developers in large part thanks to its globally distributed network of more than 25,000 nodes.
With Pocket’s protocol structured to incentivize node runners of all shapes and sizes to participate in servicing applications, we’ve been able to build this network of thousands of nodes across nearly 30 different countries. Being able to rely on tens of thousands of nodes spread across the world brings true decentralization and redundancy to our RPC service, and this redundancy translates into virtually zero downtime for applications.
Additionally, Pocket’s nodes operate across a long list of service domains – there is no single node service provider that is dominating the set of servicer and validator nodes. In terms of the amount of POKT staked to run nodes, there is currently not a single service domain that has any higher than a 19% share of staked POKT. As a result, even if a service domain experienced issues of their own, there is still healthy redundancy across the rest of the set of diversified domains.
On top of the decentralization of nodes and service domains, Pocket’s approach to session generation plays a major part in ensuring redundancy across service nodes, and uninterrupted uptime. When dApps use Pocket Network for RPC service, they are serviced in a constant cycle of approximately hour-long “sessions,” each of which consists of 24 different nodes that are assigned to the session and responsible for servicing their data relays for that period.
With 24 nodes assigned to a single session servicing a dApp, this means there are that many extra layers of redundancy if any of these service nodes experiences issues in relaying data. Plus, Pocket’s cherry picker mechanism acts to implement further quality of service checks within those sessions. These checks address issues like improper staking, out of sync block heights, underperforming and slow nodes, and nodes failing under high loads. In combination, these quality of service checks help to prioritize the most performant nodes to deliver unstoppable uptime for dApps.
With this combination of decentralized and globally distributed nodes, diversification of service domains, redundancy of nodes within a service session, and quality of service controls, Pocket Network is able to consistently achieve upwards of a 99.8% relay success rate across the network at any given time.
To tap into maximum uptime for your dApps, developers can get set up in a matter of minutes with our Always Free tier or our Pay As You Go plan in the Pocket Portal today.