Question 1

ParityDb vs RocksDB for blockchain state storage?

Accepted Answer

ParityDb is specifically optimized for blockchain workloads with read-heavy patterns, batch writes, and reference counting for Patricia-Merkle tries, whereas RocksDB is a general-purpose key-value store. For blockchain state storage, ParityDb's design may offer better read performance and crash recovery tailored to trie structures.

Question 2

How to set up ParityDb for storing Ethereum state?

Accepted Answer

Partition data into columns for different types like state trie nodes and block headers, using hash indexes for uniformly distributed keys. Ensure adequate system memory for OS page cache to maximize read performance, and configure atomic transactions for batch writes during block imports.

Question 3

What happens if ParityDb crashes during a write batch?

Accepted Answer

On restart, ParityDb validates any existing binary log files and applies changes to restore consistency, but writes that were committed but not finalized to disk may be lost, as highlighted in the durability notes.

Question 4

Can ParityDb handle values larger than 32KB?

Accepted Answer

Yes, ParityDb supports values over 32KB through a separate multipart entry table, but the README notes that values over 16KB are rare in its intended blockchain use, and performance is optimized for smaller values.

Question 5

Does ParityDb support concurrent writes from multiple threads?

Accepted Answer

No, writes are serialized and handled by a single commit worker, though multiple concurrent readers are supported. This design ensures atomicity but limits write concurrency.

Question 6

How does ParityDb scale with available memory?

Accepted Answer

Performance scales linearly with system memory because it relies on OS page cache; more memory allows more data to be cached, improving read throughput without custom caching overhead.

ParityDB

What is ParityDB?

Overview

Use Cases

Best For

Related Projects

Found a gem we're missing?

Not Ideal For

Pros & Cons

Pros

Cons

Frequently Asked Questions