Skip to main content

Storage

Overview​

The Mind-chaincurrently utilizes LevelDB for data storage, as well as an in-memory data store.

Throughout the Mind-chain, when modules need to interact with the underlying data store, they don't need to know which DB engine or service they're speaking to.

The DB layer is abstracted away between a module called Storage, which exports interfaces that modules query.

Each DB layer, for now only LevelDB, implements these methods separately, making sure they fit in with their implementation.

blockchain/storage/storage.go
// Storage is a generic blockchain storage
type Storage interface {
    ReadCanonicalHash(n uint64) (types.Hash, bool)
    WriteCanonicalHash(n uint64, hash types.Hash) error

    ReadHeadHash() (types.Hash, bool)
    ReadHeadNumber() (uint64, bool)
    WriteHeadHash(h types.Hash) error
    WriteHeadNumber(uint64) error

    WriteForks(forks []types.Hash) error
    ReadForks() ([]types.Hash, error)

    WriteDiff(hash types.Hash, diff *big.Int) error
    ReadDiff(hash types.Hash) (*big.Int, bool)

    WriteHeader(h *types.Header) error
    ReadHeader(hash types.Hash) (*types.Header, error)

    WriteCanonicalHeader(h *types.Header, diff *big.Int) error

    WriteBody(hash types.Hash, body *types.Body) error
    ReadBody(hash types.Hash) (*types.Body, error)

    WriteSnapshot(hash types.Hash, blob []byte) error
    ReadSnapshot(hash types.Hash) ([]byte, bool)

    WriteReceipts(hash types.Hash, receipts []*types.Receipt) error
    ReadReceipts(hash types.Hash) ([]*types.Receipt, error)

    WriteTxLookup(hash types.Hash, blockHash types.Hash) error
    ReadTxLookup(hash types.Hash) (types.Hash, bool)

    Close() error
}

LevelDB​

Prefixes​

In order to make querying the LevelDB storage deterministic, and to avoid key storage clashing, the Mind-chainleverages prefixes and sub-prefixes when storing data

blockchain/storage/keyvalue.go
// Prefixes for the key-value store
var (
    // DIFFICULTY is the difficulty prefix
    DIFFICULTY = []byte("d")

    // HEADER is the header prefix
    HEADER = []byte("h")

    // HEAD is the chain head prefix
    HEAD = []byte("o")

    // FORK is the entry to store forks
    FORK = []byte("f")

    // CANONICAL is the prefix for the canonical chain numbers
    CANONICAL = []byte("c")

    // BODY is the prefix for bodies
    BODY = []byte("b")

    // RECEIPTS is the prefix for receipts
    RECEIPTS = []byte("r")

    // SNAPSHOTS is the prefix for snapshots
    SNAPSHOTS = []byte("s")

    // TX_LOOKUP_PREFIX is the prefix for transaction lookups
    TX_LOOKUP_PREFIX = []byte("l")
)

// Sub-prefixes
var (
    HASH   = []byte("hash")
    NUMBER = []byte("number")
    EMPTY  = []byte("empty")
)

Future Plans​

The plans for the near future include adding some of the most popular DB solutions, such as:

  • PostgreSQL
  • MySQL

📜 Resources​