ChainSafe Cannon R&D Proposed Scope

Cannon Memory Merklization R&D

Instrumentation of FPVM:

• Create a strategy for measuring the costs incurred by the memory merkle data structure for both cache invalidation and memory merklization when running the op-program. See open issue
• • This may include contributing the fp-tests framework if appropriate

Survey and Implementation of alternative constructions:

• Refactor Cannon to allow modular switching of memory merklization strategy
• Assess alternative Merkle data structures. This may include, but is not limited to
• • Truncated hexary radix trie (see Howitzer implementation)
• • Any other novel construction that does not introduce new cryptography or assumptions
• For each construction quantify the costs for off-chain emulation and also calculate/estimate the on-chain costs and implementation complexity.
• Collate results and present comparisons