Scale or Die at Accelerate 2025: The road to Decentralized Nasdaq

Solana sets its sights on disrupting traditional financial markets with a bold vision for a decentralized NASDAQ. Max Resnick from Anza unveils groundbreaking technical innovations that could revolutionize blockchain performance and market structure, potentially positioning Solana to compete head-on with centralized exchanges.

Summary

Max Resnick's presentation at Accelerate 2025 outlines Solana's ambitious roadmap to create a decentralized version of NASDAQ. This goal has been part of Solana's vision from its inception, aiming to compete not just with other blockchain projects, but with major centralized financial exchanges.

The presentation details several key technical improvements needed to achieve this vision. These include dramatically increasing transaction throughput to rival or exceed centralized exchanges, reducing block times to as low as 20 milliseconds, and implementing a novel multi-leader consensus mechanism to improve market fairness and efficiency.

Resnick also delves into the complexities of market microstructure, explaining how improvements in transaction ordering and prioritization could lead to tighter spreads and better prices for traders. This could potentially make Solana-based markets more attractive than traditional exchanges.

The talk concludes by tying these technical innovations back to the core mission of creating a decentralized alternative to NASDAQ, highlighting how each improvement contributes to making Solana a viable competitor in the global financial marketplace.

Key Points:

Increasing Transaction Throughput

One of the primary goals for Solana is to dramatically increase its transaction processing capacity. Currently, Solana's throughput is below that of Visa, which processes about 7,400 transactions per second (TPS). The aim is to surpass NASDAQ's 70,000 TPS and ultimately reach a target of one million TPS.

This ambitious goal is being actively pursued by both the Firedancer and Anza teams. Recent progress includes the activation of SIMD 207 on Mainnet, with SIMD 256 on the horizon. These improvements are expected to double the number of compute units (CUs) this year, marking significant strides towards the ultimate TPS goal.

Reducing Block Times

Another crucial aspect of Solana's development is the reduction of block times. Currently at 400 milliseconds, the implementation of Alpine Glow is expected to bring round-trip confirmation times down to about 150 milliseconds. However, Resnick envisions pushing this even further, targeting block times as low as 20 milliseconds.

To achieve this, Solana plans to implement pipelined consensus, allowing consensus to be performed on multiple blocks simultaneously. This approach could potentially lead to new blocks or sets of blocks from multiple concurrent leaders being produced every 20 milliseconds. The Alpine Glow system is particularly well-suited for this kind of pipelining, opening up possibilities for asynchronous execution and overcoming traditional blockchain speed limitations.

Improving Market Structure

A significant portion of the presentation focused on improving Solana's market structure to compete with centralized exchanges. Resnick explained the components that make up the spread in financial markets and how these affect the competitiveness of an exchange.

By analyzing real-time data from exchanges like Binance, Resnick illustrated how price volatility impacts spreads and the challenges this poses for market makers. He then proposed solutions to help limit losses for market makers, potentially allowing them to quote tighter spreads and offer better prices to traders.

Multi-Leader Consensus

One of the most innovative proposals in the presentation was the introduction of a multi-leader consensus mechanism. This approach aims to solve issues with transaction ordering and censorship resistance that can occur in single-leader systems.

In the proposed system, multiple leaders would propose blocks simultaneously. These blocks would then be merged, with transactions ordered by priority fee. This system allows for more complex and fair ordering logic, potentially prioritizing certain types of transactions (like cancellations) over others.

The multi-leader system also includes a novel approach to block propagation and consensus, using erasure coding and a two-step consensus process. This method helps prevent front-running and sandwich attacks by hiding block contents until after consensus is reached.

Facts + Figures

• Solana's current transaction throughput is below Visa's 7,400 TPS
• NASDAQ processes about 70,000 TPS
• Solana aims to reach 1 million TPS
• Current Solana block time is 400 milliseconds
• Alpine Glow is expected to reduce confirmation times to 150 milliseconds
• Future goal is to achieve 20 millisecond block times
• SIMD 207 has been activated on Mainnet
• SIMD 256 is coming soon
• Compute units are expected to double this year
• On centralized exchanges, market makers win cancellation races only 13% of the time
• The proposed multi-leader system uses erasure coding for block propagation
• The consensus process involves a two-step approach with relays reporting shred availability

Top quotes

1. "From day one, kind of the founding of Solana, it was not a project designed to kill Ethereum or anything like that. It was way more ambitious than that."

2. "We need to obviously increase bandwidth, increase the number of TPS that we can do. NASDAQ does about 70,000. Visa does about 7,400 per second. We're still kind of below Visa right now. We'd love to be at a million TPS."

3. "I'd like to see us get to 20 millisecond block times."

4. "Market makers pick up pennies and lose dollars."

5. "Can we help limit those losses for market makers so that they can quote tighter?"

6. "Given that we're going to have to do batching, it would be nice if we could process the cancels before the takes. That would reduce the spread."

7. "The problem in the single-leader case is that one person decides what goes on chain. So why don't we just let there be more?"

8. "This allows us to prioritize cancels."

9. "It effectively hides the contents of the block until after a consensus is reached, at which point we can replay the block."

10. "We need to be able to give applications the power to sequence their own transactions. And doing that will allow them to tighten spreads, and hopefully we can achieve decentralized NASDAQ."

Questions Answered

What is Solana's ultimate goal in terms of competing with traditional financial markets?

Solana's ultimate goal is to create a decentralized version of NASDAQ. This means competing not just with other blockchain projects, but with major centralized financial exchanges like the New York Stock Exchange, NASDAQ, and CME. The aim is to achieve comparable trading volumes and offer competitive spreads, essentially providing a decentralized alternative to these established financial powerhouses.

How does Solana plan to increase its transaction processing capacity?

Solana aims to dramatically increase its transaction throughput from its current level (below Visa's 7,400 TPS) to eventually reach 1 million TPS. This is being pursued through various technical improvements, including the activation of SIMD 207 on Mainnet and the upcoming implementation of SIMD 256. These enhancements are expected to double the number of compute units this year, significantly boosting Solana's processing capacity.

What improvements are planned for Solana's block times?

Solana is working on reducing its block times from the current 400 milliseconds to as low as 20 milliseconds. This will be achieved through a combination of improvements, including the implementation of Alpine Glow, which is expected to bring round-trip confirmation times down to about 150 milliseconds. Further improvements will involve pipelined consensus, allowing for consensus on multiple blocks simultaneously and potentially producing new blocks every 20 milliseconds.

How does Solana plan to improve market structure to compete with centralized exchanges?

Solana plans to improve its market structure by implementing a multi-leader consensus mechanism and allowing for more sophisticated transaction ordering. This includes prioritizing certain types of transactions (like cancellations) over others, which can help market makers limit losses and potentially offer tighter spreads. The system also aims to prevent front-running and sandwich attacks by hiding block contents until after consensus is reached.

What is the multi-leader consensus mechanism proposed for Solana?

The multi-leader consensus mechanism involves multiple leaders proposing blocks simultaneously. These blocks are then merged, with transactions ordered by priority fee. This system allows for more complex and fair ordering logic, potentially prioritizing certain types of transactions over others. It also includes a novel approach to block propagation using erasure coding and a two-step consensus process, which helps prevent front-running and improves overall market fairness.

How does the proposed system aim to prevent front-running and sandwich attacks?

The proposed multi-leader system uses a unique approach to block propagation and consensus. Blocks are shredded and erasure coded, then sent to relays. The relays report which shreds they have to a consensus leader, who then packs this information into a block for consensus. This process effectively hides the contents of the block until after consensus is reached, making it much more difficult for malicious actors to front-run or sandwich trades.

Why is improving transaction ordering important for Solana's competitiveness?

Improving transaction ordering is crucial for Solana to offer competitive spreads and attract traders from traditional markets. By allowing for more sophisticated ordering logic, such as prioritizing cancellations over new orders, the system can help market makers limit losses and quote tighter spreads. This, in turn, can lead to better prices for traders, making Solana-based markets potentially more attractive than traditional exchanges.

How does Solana plan to compete with the speed of centralized exchanges?

Solana aims to compete with the speed of centralized exchanges through a combination of increased transaction throughput and reduced block times. By targeting 1 million TPS and block times as low as 20 milliseconds, Solana could potentially offer confirmation times comparable to or even faster than many centralized exchanges. This speed, combined with improved market structure, could make Solana a viable alternative to traditional financial markets.