Get in touch




Arcology is a an EVM equivalent rollup network capable of processing multiple transactions simultaneously to significantly improve scalability. It can handle tens of thousands of transactions per second, outperforming all existing high-performance blockchains. Arcology also tackles the major stateDB bottleneck, making it possible to fully realize the benefits of parallel execution.


The idea behind Arcology is to create an L2 rollup that can process multiple transactions simultaneously, dramatically improving overall throughput to handle tens of thousands of transactions per second, outperforming all existing high-performance blockchains while maintaining full EVM equivalence. To date, no other project has achieved both.

Arcology is good for all blockchain-native applications that weren’t believed possible due to poor scalability or high cost. It is a perfect choice for resource-intensive applications.

Generally, parallel execution means processing multiple transactions at the same time instead of one after another. Modern computers usually have multiple cores and processors, making it possible to improve scalability by handling many tasks simultaneously to enhance the system’s speed and efficiency.

All the transactions are executed in full parallel by multiple EVM instances to process transactions and generate state transition. Arcology has a built-in concurrency control system to protection the state consistency at all time. The concurrency control is based on an STMbased optimistic approach.

No, unlike some EVM parallelization proposals, Arcology didn’t choose to reimplement the EVM to support concurrency. Instead, it extended the original implementation, enabling multiple independent EVM instances to work simultaneously. Simply put, the parallelization happens at the intra-VM level, not the inter-VM level.

This approach makes no breaking changes to the original EVM, allowing Arcology to easily maintain equivalence as the EVM evolves. Additionally, it enables Arcology to scale across a cluster of interconnected machines, leveraging multiple cores and processors for even greater scalability and efficiency.

The node cluster is deployment mode in which a group of interconnected machines working together as a single node on Arcology network.

Conventionally a node is just a machine with a piece of client software installed (Geth for instance). Arcology allows multiple interconnected machines (physical / virtual) to share the workload of a single node.

There is always an upper limit on how many processors, memory, and storage space a single machine can have. A cluster-enabled design can effectively remove this constraint, allowing for more computational power and data storage to deliver higher throughput.

There is no theoretical limit on how fast Arcology can run. The throughput is more of a matter of resources. Arcology takes full advantage of all the resources available. 

Yes, for Ethereum, the single biggest performance bottleneck is the StateDB. Arcology addresses the StateDB bottleneck by making the Merkle Patricia Trie both parallelized and asynchronous.

EIP 648, just like its name(easy  parallelizability) suggests, its a simple parallelization design and has never been implemented so far by Ethereum L1. This approach belongs pessimistic concurrency control with account level granularity which is very coarse. Account-level concurrency granularity only brings very limited speedup in reality because of its restrictive nature.

Ask the GPT for More Concurrent Programming Related Questions

more questionsWant to know more about us?




Social network

Get in Touch

Please leave your message and we will get back to you as soon as possible
Please enable JavaScript in your browser to complete this form.