An overview of ETH’s L2 Ecosystem
Non-technical intro to rollups
Making Sense of Rollups, Part One: Optimistic vs. Zero Knowledge
Making Sense of Rollups, Part 2: Dispute Resolution on Arbitrum and Optimism
What Rollups are fundamentally and how they help scaling
Building DeFi on L2s
Delphi Digital on Layer 2 Rollups
Ecosystem of Ethereum Scaling Solutions
ETH L2 Ecosystem
zkPorter : A breakthrough in L2 Scaling
Best L2 Podcast
An incomplete guide to rollups
L2 Apps on Ethereum
On Rari Capital’s approach of how a yield aggregator could build on L2 - comparing scalability solutions
Delphi Clubhouse
Trustless Scalable Smart Contracts
Almost everything you need to know about optimistic rollup
Optimism vs Arbitrum
ZK vs OR
L2s for beginners
Incomplete guide to rollups
Overview of ETH's L2 Ecosystem
Caspian - L2 AMM
Scalability Wars
zkPorter: Breakthrough in L2 Scaling
ETH Scalability - L2s
Layer 2 Gold Rush
I'm worried nobody will care about rollups
What's live in L2
Scale or Die
Pantera on Scaling
THE NEXT MAJOR UNLOCK – By Emma Rose Bienvenu, Chief of Staff
After DeFi Summer of 2020, surging demand for Ethereum-hosted protocols brought sky-high gas fees and network congestion.  Thankfully, the “Layer 2 Summer” of 2021 heralds a solution, with the much-hyped launches of scaling solutions like Optimism and Offchain Labs’ Arbitrum.  Here we discuss why Layer 2 technology matters.  We start with a primer on scaling, then detail how new off-chain tools will help tackle the twin problems of transaction speed and cost.  We conclude with a discussion of what that means for the broader ecosystem, and why layer 2s will unlock the next wave of killer apps built on blockchain-based technology.
If the promise of blockchain technology was to democratize access to finance – cutting costs by cutting out rent-seeking middlemen – its great irony is that transactions on Ethereum, which hosts most DeFi protocols, remain extremely expensive.  The reason for notoriously high gas fees is that, in a distributed network like Ethereum, where every node needs to store and process every transaction: this means that the network inevitably runs up against capacity limitations as its user base grows.  Congestion drives up gas prices and processing times, making small transactions nonviable.  It also makes it all but impossible for computationally demanding DApps to run directly on an increasingly crowded blockchain.
Ethereum has, in short, become a victim of its own success.  Over the past year, as millions rushed to embrace the DeFi protocols, NFT marketplaces, and other DApps hosted on Ethereum, the demand for processing transactions exploded.  Meanwhile, the supply of processing capacity remained frustratingly stable, constrained by the computational capacity of each node and the size of individual blocks.  When more transactions compete for a limited supply of blockspace and compute capacity, gas fees go up and transactions slow down.  In 2021, Ethereum fees rose by 845% compared to the previous year; average block capacity, which stood at roughly 70% in January 2020, has risen to a sustained level of 98%.
notion image
notion image
The Problem of Scaling
Increasing the capacity of networks like Ethereum is extraordinarily challenging.  Blockchains are defined by three main characteristics: decentralization, security, and scalability.  You can pick two of the three, but if you stick to straightforward methods, you cannot have them all.  This means that improving the scalability of a blockchain – so it can process more transactions, more quickly, more cheaply – will generally erode its guarantees of security or decentralization.
This conundrum, commonly dubbed the “scalability trilemma”, has vexed proponents of blockchain technologies since the early days of the ecosystem’s development.  In 2014, Vitalik Buterin memorably promised that the Ethereum community would either solve the problem of scalability “or die trying.”  Thankfully, with recent scalability upgrades and the launch of breakthrough Layer 2 scaling tools, success in that project is finally within reach.
There are, broadly speaking, two approaches to overcoming the scalability trilemma.  “Layer 1” or “on-chain” scaling focuses on improving the blockchain itself; “Layer 2” or “off-chain” scaling looks to improve how the blockchain is used.
Layer 1
In the Ethereum ecosystem, the leading Layer 1 scaling proposal is referred to as “sharding”, which would split the transaction database horizontally by creating new chains, or “shards”, that decrease the quantity of data each validator needs to process.  This would allow the total volume of transactions processed on the network to exceed the compute capacity of individual nodes.  Ultimately, this would reduce the barrier to entry for new validators joining the network, increasing throughput, and decreasing the cost of transacting on the network.
There are however important limitations to Layer 1 scaling approaches.  They involve extraordinarily challenging computer science and game theory challenges, many of which have never been tackled before.  Implementing them also demands a hard fork of the protocol; as a practical matter, this requires building strong consensus among all stakeholders around each Layer 1 upgrade.  As in any complex and decentralized system, this is a daunting, time-consuming proposition.  The chronic delays that have plagued “ETH-2”, a planned hard fork upgrade of Ethereum, illustrate the difficulty, on both the technical and community alignment fronts, of implementing significant Layer 1 changes.
Layer 2
While Layer 1 solutions focus on improving the performance of the core blockchain, Layer 2 approaches look to improve how the blockchain is used.  Its proponents argue that, because distributed ledgers are inherently capacity-constrained, they should only host the highest-value transaction data.  Layer 2s migrate low-criticality operations off-chain but leave assets and cryptocurrency on Layer 2, allowing users to revert to Layer 2 at any time to either resolve disputes or reclaim their crypto-assets.  This anchors Layer 2 operations in native Layer 2 security, freeing up precious blockspace on the core blockchain.  Ultimately, this allows Layer 2 to handle in aggregate a much higher volume of transactions, more quickly and at far lower cost.
There are three leading types of layer-2 solutions: state channels, side chains, and rollups.  While all three can provide compounding gains to blockchain capacity, the Ethereum community has coalesced around rollups as the most promising path to scaling the network.  While other Layer 2 solutions accept significant tradeoffs in security or decentralization to achieve scalability, rollups accept some centralization without sacrificing trustlessness, the key priority underlying decentralization.
Rollups migrate most computation off-chain, then periodically pushing batched transactions data and the resulting state root to the Layer 1 blockchain.  By executing operations outside of mainnet but recording transaction data and/or proofs on Layer 1, rollups benefit from the security of the core blockchain while achieving greater throughput and vastly lower fees. Broadly speaking, rollups come in two flavors: ZK-Rollups and Optimistic Rollups.
ZK Rollups move computation to layer 2, and periodically batch and compress transaction data conducted off the main chain, generate a validity proof of its integrity, and post it to the Ethereum mainnet.  By posting a proof of correctness of each state transition, ZK-Rollups guarantee the validity of the on-chain state, allowing for immediate withdrawals by users.  These proofs are however complex and time-consuming to compute.  Though developers will eventually be able to use Solidity with ZK Rollup technology, they currently require that smart contracts be re-written in a custom programming language.  For the time being, this makes ZK-Rollups best-suited to projects that enable straightforward payments, like decentralized exchanges or payments platforms.
We invested in StarkWare, now the leading pioneer of ZK-Rollup technology.  In mid-2020, Starkware demonstrated the power of StarkEx, its ZK-Rollup scalability engine that supports decentralized exchanges, by using it to set up 1.3M accounts on Ethereum and seed each with an initial balance, all over a 12-hour period.  Had it run directly on the mainnet, the process would have consumed the entire capacity of the Ethereum network for 4.5 days.  StarkEx managed it with a stunningly 2.5% of Ethereum’s capacity for just 12 hours, at an average cost of $0.003 per transaction.
Unlike ZK-Rollups, Optimistic Rollups assume transactions are valid and run a fraud proof only in the event of a challenge.  Optimistic rollups rely on parties verifying Layer 2 submissions – and challenging incorrect states – to uphold the integrity of the transitions.  While computationally efficient, this forces users to wait a challenge period before they can access their funds.  Nevertheless, the scalability benefits Optimistic Rollups enable are enormous, and could cut Ethereum gas fees by 10,000+% and increase throughput by up to 200x.  Projects like Arbitrum use Optimistic Rollup technology that integrates easily with existing DApps, supporting the execution of arbitrary EVM code within the off-chain layer with minimal changes to the underlying smart contracts.  Since their August 2021 launch, more than 250 developer teams have started building on top of Arbitrum.  The project was selected by Reddit to power the launch of its own Layer 2 rollup, and Aave, Balancer, Band Protocol, Coinbase Wallet, Chainlink, Curve, DAI stablecoins, Etherscan, Dodo, Metamask, Shapeshift, Sushiswap, and Uniswap all have live or upcoming integrations with Arbitrum’s technology.
The power and promise of both Optimistic and ZK Rollups are that by anchoring transactions in native Layer 1 security, giving users the option of reverting to L1 to reclaim their assets or resolve disputes, they create game theoretic incentives for both users and operators to behave honestly.  This makes them an essential catalyst to creating secure and scalable networks on which users can transact without needing to trust centralized intermediaries or their transaction counterparties.  That is, in short, the promise of blockchain and crypto writ large.  Following the much-hyped recent launches of Layer 2 scaling projects, that promise is closer now to being realized than ever before.
What Solving for Scalability Would Mean
For blockchain-based protocols to replicate – and eventually replace – the pillars of traditional finance, they will need to match or exceed the performance of its infrastructure.  Without the help of Layer 2s, the Ethereum mainnet could only process 15-20 transactions per second (TPS).  The Bitcoin mainchain can handle just seven.  The VisaNet payments network, by comparison, handles on average roughly 1,700 TPS, and NASDAQ processes around 500-1,000 TPS.  With the launch of Layer 2s, that disparity will narrow or disappear.  Rollups alone are projected to increase TPS on Ethereum to between 1,000 and 4,000 TPS.  Other Layer 2 solutions like state channels and side chains will add to the ecosystem’s capacity, as will highly scalable Ethereum alternatives like Polkadot.  Taken together, these tools will make it possible to create a blockchain-based point of sale payment system that offers instant transfers for quick checkout times and guarantees settlement, with throughput rivalling that of the visa network.
In the short-term, the benefits of higher transaction speed and lower costs will be felt most strongly in the DeFi realm.  Applications migrating to Layer 2s will pass on their significant cost savings to users, offering lower transaction fees and lower minimum transaction sizes.  This will drive overall DeFi volumes, drawing retail users who were formerly priced out of transacting directly on Ethereum away from centralized exchanges or less-secure alternatives like Binance Smart Chain.
DeFi protocols themselves will also improve.  For example, applications like Perpetual Protocol and DYDX allow users to go long or short crypto assets using margin and leverage; when the value of posted collateral drops below a specified threshold, the collateral is liquidated automatically.  When transactions take several minutes to process, that liquidation mechanism functions sub-optimally, but in a scaled ecosystem of near-instantaneous transactions, it can function vastly more efficiently and precisely.  This dramatically increases the capital efficiency of DeFi protocols, allowing dYdX’s Starkware integration, for example, to offer users higher maximum leverage thresholds, lower penalties for liquidation, and cross-margin trading functionalities.
More broadly, the advent of high throughput, virtually free and instantaneous transactions will allow blockchain-based protocols to rival or surpass the user experience available on traditional web browsers.  One reason that attempts to build decentralized Facebook, World of Warcraft, Visa, or SecondLife have largely stalled in production is that, historically, public blockchains have lacked the capacity to handle applications that required hundreds or thousands of computations per second.  With scaling tools that increase throughput by 50-1,000x, apps that used to be too slow or computationally intensive to run on the core blockchain will become viable.  This will mark an extraordinary expansion in the size and scope of addressable markets for blockchain-based projects, unlocking the next generation of decentralized social media, virtual worlds, micropayments platforms, gaming ecosystems, and more.
For blockchain-based systems to deliver on their vision for a transparent, secure, censorship resistant, and privacy-protective financial infrastructure, expanding the capacity of the ecosystem will need to be an iterative, ongoing project.  In the immediate, Layer 2 solutions can provide powerful, flexible, compounding tools.  In time, the best approach to scaling Layer 1 may well turn out to be improving the efficiency with which it functions as a data availability engine, to optimize Layer 1’s ability to host for Layer 2s operating atop it.  In true decentralized fashion, the problem of scaling is and will continue to be tackled by a variety of complementary solutions.  For the moment, Layer 2 are leading the charge toward a scalable, compostable blockchain ecosystem – and we can’t wait to see the wave of innovation they unlock.
Dune : Gas Consumption
L2 Skill Tree
A dive into Matter Labs
What is zkSync
Nick on Arbitrum
Arbitrum Ecosystem Mega Thread
What is Optimistic Ethereum
Summary : Unrolling rollups
Cross-rollup NFT Wrapper and migration idea
Why rollups + data shards are the only sustainable solution for high scalability
Deep dive into validity proofs and layer 2
ScribeDAO: Rollups - The ultimate ethereum strategy?
ScribeDAO: Why rollups and data sharing are the only solution for high scalability
ScribeDAO: All you need to know about Optimistic Rollups
Making sense of ETH L2’s Scaling Solution
Starknet vs Solana
Coinbase: Scaling Ethereum & crypto for a billion users
Optimistic rollups are the present and future of eth scaling
L2 Skill Tree
Guide to ETH Scaling Solutions
Ultra Scalable Ethereum
Halo and More: Vitalik
The ultimate ETH L2 Guide
Optimism vs Arbitrum
How Zero Knowledge Proofs became ETH´s magic bullets
zkRollups comparison
OR and ZK thread
Addressing common rollup misconceptions
Why calldata gas cost reduction is crucial for rollups
Explainer on L2s
On Vitalik’s gas cost reduction proposal
A roadmap for scaling rollups
zkSync vs Starkware
How rollup fees work
Polyna on rollups
zkRollup directory
Potential of Zk proofs
ZKRs vs ORs
Thoughts on eth l2s
Ethereum-powered ZK-Rollups: World Beaters
The basics of zkSTARKs and zkSNARKs
Incomplete guide to rollups
Almost everything you need to know about optimistic rollups
ETH powered ZK Rollups
ZK Snarks vs ZK Starks
Why L2s are the future
Starknet and the future of ETH Scaling
L2 Bridges, Tokens and fungibility
Optimistic vs ZK roll ups
What is a ZK Proof
L2 Projects launching soon
L2 tokenomics
This week in L2s
Rollups are superior than L1s
L2 happenings
Top DeFi derivatives on L2
L2 to L1 differences between optimism and starkware
Emin on Dechads
L2 tokens are coming
Rise of L2 Native apps
L2 Scaling
Architecture diagrams on mainnets/L2s
The endgame bottleneck: historical storage
Optimism protocols
What different zk terms mean
Recursive Zero Knowledge proofs.
Matter Labs vs Starkware
Hasu gets STARK-pilled
ZkRollups vs Optimistic Rollups
Introduction to Zero Knowledge Proofs
ETH scaling using ZKP
On scaling challenges and their solutions
Why L2s > L1s
On L2 growth
Polygon: A Multi-Sided Approach to ZK Scaling
L2s and ETH economics
ZK Landscape
OR vs ZK
What are ZKRollups
Privacy for pennies: Scaling Aztec’s zkRolup
Updated thoughts on modular blockchains
Privacy-protecting crypto airdrops w/ZK Proofs
Vitalik: The roads not taken
Bridge differences
4 Misconceptions about PoS vs PoW
Layer 2
What are bridges and how can we classify them?
Mega Thread about L2s
Primer on cryptographic proof systems
Deep dive into L2s Ecosystem
On bridges
Applying ZK Tech to bridges can fix issues
Types of Rollups
What’s wrong w/bridges
On Good Blockchain infrastructure
How to run an ETH Node
zkLend / JediSwap AMA
Hardware acceleration for ZKPs
Summary of blockchain scaling
Cheat sheet on roll ups/scaling
Our Network Bridging meta-issue
Decentralised Speed: Advances in ZKP
Optimistic vs Zk Rollups
Sovereign rollups
Building a defi protocol on starkness w zklend
Gubsheep on ZKP
A look at blockchain as a service
Incomplete Guide to ZK: Why ZK Matters?
On ZK Proofs
Solana vs Celestia
Subnets vs L2
Our Network 118: L2s
Zero-knowledge proofs have a lot of applications beyond blockchains.
A thread about the advantages and misconceptions about rollups
Solana vs Celestia
ZK Proofs for anon airdrops
Innovative non-financial primitives on Solana
Tarun explains cross-chain bridging
Comparison of heterogeneous blockchain networks
Breaking down blockchain economics
Money under the bridge
Long Take: Deriving Polygon Supernets and Avalanche Subnets value accrual through the Laws of Nature
EVM Equivalence
Mega Thread about L2
With bridges, trust is a spectrum
Thread of Threads on L2s
ZK resources
Modular vs monolithic blockchains
DISRUPTORS: Modular Blockchains for Sovereign Communities With Hacktivist Mustafa Al-Bassam
Mirror weekly
Types of rollups
Layer-2 Scaling Solutions: A Framework for Comparison - Commissioned by Polygon
ZK Research recommendations?
Fundamental and applied differences between ZK scaling solutions
EVM Deep Dives Pt 4 - Opcodes in the Geth client
Building an EVM from Scratch pt 2
What are ZK-Rollups
Attempt to break Fuel Labs - importance of running validators for optimistic rollups
Foundations of blockchains: tender mint
On the long term sustainability of economic systems
The Fuel for Fast Execuiton
Scale w Subnets
CZ’s FAQ 8 - On LUNA/UST and Taking the Right Risks
Thoughts on LUNA meltdown
A case for long TRXUSDT
zkLEND ama
Bridge resources
Layer 3s
Updated view of rollups development
On why the reorg could have been a big deal but was not really
ACD call notes
Learning ZKP
On rollups
EVM Network effects
ZK Research Library
How Bridges can be the next big risk in crypto
Mental models for L1 and L2s
ZK (Validity) Rollups: Entering the General Purpose Era
Layer 2 Development
ACD Recap
L2 Token Endgame
Where to use a blockchain in non-financial applications
ETH’s Rollup Centric Future
The Bridges are dead, long live the bridges
Fullstack bridging - beating tradeoffs w modularity
Modular musings
What makes a chain good to build in
solana poses a large and growing threat to ethereum.
Rollups vs. Appchains / Subnets
Modular design overview
Sidechains are not Layer 2
ELI5 the data availability problem
Managing risk in blockchain deployments
Optimism vs. Arbitrum - A Complete Comparison
L2 Bridge Risk Framework
Rollups as sovereign chains
Modular Blockchains
ZK Landscape
Ethereum & Celestia: Designing Modular Stacks
Comparison Arbitrum & Optimism full node synch
Why rollups hit the sweet spot
Deep Dive: EVM Compatibility
Rollups: Execution Through the Modular Lens
ETH vs Cosmos
Evaluating the Cosmos Chain vs L2 Decision for Investors and Builders.
Is Fuel The Best Modular Execution Layer?
Subnets vs supernets
Decentralised VPNs
Extracting value from a L1/L2
FuelVM:An Alternative to the Ethereum Virtual Machine
EVM-Chains in the Bear Market.
Under The Radar Layer 2s: A closer look at Scroll, Aztec, Immutable X, Polygon zkEVM
State of scaling 1: dYdX, Arbitrum, Optimism
Optimistic Rollups in Practise
Polkadot vs Cosmos
Bridges - an introductions
Non EVM chains in the bear market
Layer 2 Playgrounds
Dev learning experience on Starknet and Sui
How bridges compare
ZK Proof Protocols
BTC lightning
Are L2s good or bad for ETH
Zero Knowledge Proofs
State of Scaling Issue 3
Modular Blockchains & a Rollup-centric Ethereum | Full Video
Assessing Blockchain Bridges
Why everyone should be thinking about Fuel
Best resources for understanding modular
Making blockchains suck less
Progression of the Data Availability Problem
Navigating Arbitrary Messaging Bridges: A Comparison Framework
Zero Knowledge Proofs: Primer for non-technical web3 researchers
10 Incredible resources for all-things ZK Proofs.
The inevitability of UNIchain
special edition: all things zero knowledge
ZK Bridges
Starkware, L3s and the ZK Rollup Future | Eli Ben-Sasson, Uri Kolodny
Building zk infra: which to decentralise first: the prover or the sequencer
ZK Scaling
State of scaling 8: state of zk rollups
Dawn of a modular blockchain era
The Zero-Knowledge Landscape: Part 2
Validity rollups on bitcoin
The state of ETH L2s
On sidechains, L2s, rollups and more.
A Comparison of Blockchain Execution Environments: EVM, Sealevel, MoveVM, Wasm, CairoVM
Optimistic vs ZK Rollups
ZK Proof Workshop
15-minute fundamentals with Connext | Token Terminal