eth2 Migration: What to Expect

It should come as no surprise that Ethereum is currently undergoing rigorous preparation for a transition to Proof of Stake via eth2 (or Serenity) in the coming months. We recently saw the introduction of an eth2 testnet block explorer – a good signal that progress is gradually being made. For those unfamiliar with Serenity and what it entails, we recommend brushing up with this resource.

Over the past few months, we’ve seen proposals for what the eth1 to eth2 migration will entail. Without going into too much detail, eth2 is an entirely new blockchain, meaning the core challenge here is how to properly port all of the state, accounts, and metadata from the existing Proof of Work chain (eth1) to a newly launched Proof of Stake beacon chain (eth2).

Similarly, eth2 marks the introduction of sharding, essentially splitting elements of the larger blockchain into smaller pieces to allow for increased throughput. In doing so, many have raised concerns that this dynamic may break the composability of many prominent DeFi applications.

Let’s take a look at what you (as an average user) can expect upon this transition.

ETH1 <-> ETH2

Perhaps the most refreshing thing for non-technical users to hear is that upon a successful transition, Ethereum transactions *should* feel the same.

“The changes and disruptions that you experience will actually be quite limited. Existing applications will keep running with no change. All account balances, contract code and contract storage (this includes ERC20 balances, active CDPs, etc) will carry over. Very different code paths would be used to package and broadcast transactions, but the functionality provided would be the same.”

In the case of something like a Vault in MakerDAO, everything will be handled by the client code, meaning there aren’t any *upgrades* you will have to perform on your end.

New Developments

Vitalik’s most recent post proposes a way to expedite the transition by allowing the eth1 system to “live” as shard 0 of eth2, allowing Ethereum to benefit from eth2 scalability faster, without as significant of a risk. This tweet sums it why this is exciting:

This proposal, if implemented, is pretty bullish for $ETH imo:

1) validators can register/degister freely, bring more liquidity into PoS w/o fear of 1way bridge (which created 2 tokens)
2) earlier use of eth2 scalability
3) earlier deprecation of PoWhttps://t.co/SMFtKfcJH1

— Su Zhu (@zhusu) December 26, 2019

But What About DeFi?

For one, it’s refreshing to see that DeFi composability has become a strong talking point on the official Ethereum research boards. From what has been discussed, it seems as though simple composability (i.e. lending Dai via Compound) or “doing something here that will soon have an effect over there” will be quite simple to port over to eth2.

Where we start to see some potential issues are with complex interactions in which contracts are calling multiple accounts across numerous different shards, or as Vitalik likes to call it,  “doing something here, then doing something over there, then doing more things here based on the results of things over there, all atomically within a single transaction”.

Drawing off the proposal, it sounds like large DeFi products will ultimately seek to live on their own shard, creating what might be described as a digital shopping mall. While it’s entirely possible to have a store ship something to your house, it will likely take longer and cost more to do than you going directly to the store and performing that transaction in person (assuming Etherum’s Amazon Prime hasn’t been invented yet).

Taking this a step further, it doesn’t seem so far fetched that a few shards may be coined as “DeFi shards” up until they reach maximum capacity. In these instances, the true question becomes:

What do I as a non-technical user need to do to interact with a specific shard?

I’d be willing to assume one of two things:

1. Shard identification will be similar to contract interaction today in which all the service provider automatically populates your transaction to ensure it is performed exactly as intended.
2. New service providers will emerge as a form of relayers, essentially performing transactions on your behalf in the event that it becomes too complicated (which is unlikely) for the average non-technical individual to digest.

Key Takeaways

Perhaps the biggest thing of note here is that as an end-user, it’s likely that all you will experience are *temporary* increased gas costs and a small amount of downtime (it’s said to take no longer than an hour) upon the migration from eth1 to eth2.

In turn, we can expect Ethereum to shift to a stake-based consensus mechanism that supports higher throughput, enhanced security and new cross-shard services to evolve.

For something that is so technically complex to achieve, the ramifications for doing so seem quite small.

If one thing is for sure, we expect the discussion surrounding composability to continue to make waves, ultimately encouraging a star pupil to emerge with what could very well be one of the most important Ethereum tools and/or contracts of all time.

Until then, we’ll be on the lookout for any updates worth sharing your way!