Skip to main content

ES (Elasticsearch)

Use Cases​

  • Search
  • Statistics
  • In our company, the use cases are broadly as follows:
    • Trading Counter: Mainly stores historical bills, orders, order cancellations, K-line data points, growth data, and other business-related data.
    • Explore: Uses ES to store blocks, transactions (index-related), internal transactions, token transfers, addresses, tokens, token holdings, logs (indexed), and other index-related data.
    • Funds: Mainly stores asset analysis data and historical transaction records.
    • Wallet: Mainly stores transaction data.
    • NFT: Mainly stores NFT assets and transaction history.

Advantages​

  • Capable of storing massive amounts of data.
  • Effective read scenarios under various query conditions.

Disadvantages​

  • Reading from ES can have certain delays.
  • Principle: ES writes first to memory, and the data in memory is periodically flushed to disk to form files. Only data that has been written to disk can be queried; data in memory cannot be queried, hence the read delay issue.
  • No locks, no transactions.