Postgresql Blob block storage index schematic database schema.

[Andrei-Dolgolev]

We need develop index structure for provide fast access to GCS files and maintain system of storage for RDS of our cstomers.

What we store:

For now we have 5 types of structures stored per chain some of them can have additional fields or don't have some of them(that problem will be handle on crawler issue).

• Blocks
• Transactions
• Events
• Blocks_trace
• Transactions_trace

Json examples of all data structures (ethereum chain): https://gist.github.com/Andrei-Dolgolev/443ddd96d31ba72e939174fae3fdaae3

Important things from moonstram V2 events/transactions must include block_hash for possiblity to running forks detection and fixing.

Indexes

Blocks:

block_number, block_hash, block_timestamp -> (file_link)

Route to blocks: GCS/blockM..blockN/blocks.proto

Use cases: Connect with events/transactions/traces for fork protetction

Transactions:

Transactions_hash, from_address, to_address, block_hash -> (file_link)

Route to transactions: GCS/blockM..blockN/transactions.proto

Use cases: Get Transactions for customers RDS and decode them to tx_call by particular to_address. Possably restore some of transacrions in user database tx_table.(depends on transactions trace)

Events:

address, topic, block_hash, tx_hash ->(file_link)

Route to events: GCS/blockM..blockN/events.proto

Use cases: Get events for customer RDS by contract address.

Block trace:

Use cases: Restore internal calls to user RDS by to_address.

Transaction trace:

Use cases: Restore latest state of storage to user RDS