A lot of the work is done in parallel, meaning scalability is closer than it seems.
Following the successful launch of Ethereum 2.0 Phase 0, the first concrete step in building the next iteration of the protocol, Ethereum co-founder Vitalik Buterin published an updated roadmap of what comes next for the project.
Current development of Ethereum 2.0 is generally divided in phases. Phase 0 is the barebones Beacon Chain that enables staking, but has no effect on the application layer. Phase 1 would introduce sharding of data, increasing storage capabilities without directly influencing application performance. Finally, Phase 2 fully introduces transaction sharding and enables the promised thousands of transactions per second of throughput.
Buterin had said in March that this roadmap template is a vision for the next 5–10 years. The updated version is more fluid, and eliminates terms like “Phase 1” and “Phase 2” altogether. Defining features of each phase are now more independent from each other, and incorporate work done for Ethereum 1.x.
Buterin’s roadmap encompasses all Ethereum developments to provide a detailed overview of what’s next for the platform, including approximate completion bars.
The roadmap I made back in March updated with (very rough and approximate!) progress bars showing what has been done and some of the recent tweaks to the roadmap itself.
A lot has been accomplished, but still a lot remains to be done! pic.twitter.com/4zIK4aTNTh
— vitalik.eth (@VitalikButerin) December 1, 2020
Key features of the next major milestone are the transition of Ethereum 1.0 to proof-of-stake, the introduction of Eth2 light clients on Eth1, and data sharding — all of which were previously grouped under Phase 1 and 1.5. An important change in the current roadmap is the realization that the three steps are largely independent, and can be worked on in parallel.
Furthermore, enormous scalability improvements can be achieved by executing on the latter two features, as they would enable hosting rollups on a sharded data structure. Rollups are a layer two technology that offloads computation outside of the chain but guarantees its correctness through proofs stored on-chain. Due to this, data sharding vastly increases the Rollups’ room for maneuver and could enable more than 10,000 TPS as soon as light clients and data sharding are introduced.
However, progress on data sharding and light clients is still at or below 50% according to Buterin, consistent with estimates that Phase 1 will take at least one year to build.
Progress on stateless clients and the Ethereum 1.x initiative is also at less than 50% according to the chart. The cryptographic technology of polynomial commitments — which Buterin earlier said is the key to practical stateless clients — is also far from being complete. Similarly, work towards many other advanced types of cryptographic technology and an improved virtual machine is still in its early stages.
After publishing the roadmap, Buterin urged to quickly implement EIP-1559, a proposal to burn the majority of the transaction fees collected by the protocol, instead of offering them to miners.