How to speed up data modification operations in a table with a million users, of whom only a thousand are active?

To speed up data modification operations in a table with a million users, of which only a thousand are active, it is advisable to focus on optimization specifically for active users:

• Use indexes on fields that allow quick identification of active users (for example, an index on the active = true field).
• Perform updates only on the filtered set of active users to avoid touching the entire table.
• Consider partitioning the table, isolating active users into a separate partition or table.
• Use batch updates to reduce the number of transactions.
• If possible, apply lazy updates or update data asynchronously.

Example in SQL:

UPDATE users
SET last_login = NOW()
WHERE active = TRUE;

In Java, you can use batch update with JDBC:

String sql = "UPDATE users SET last_login = ? WHERE id = ?";
try (PreparedStatement ps = connection.prepareStatement(sql)) {
    for (User user : activeUsers) {
        ps.setTimestamp(1, new Timestamp(System.currentTimeMillis()));
        ps.setLong(2, user.getId());
        ps.addBatch();
    }
    ps.executeBatch();
}