mysql sum多个字段千万级数据如何查询优化？

统计数据表中多个sum千万级数据超时。由于业务需要实时 所以做不来快照表

我加了索引似乎也不管用

SELECT
    f.*,
    c.c_name,
    u.username,
    sum( f.dy_sc_num ) AS dy_sc_num_total,
    sum( f.dy_gz_num ) AS dy_gz_num_total,
    sum( f.dy_like_num ) AS dy_like_num_total,
    sum( f.ks_gz_num ) AS ks_gz_num_total,
    sum( f.ks_dz_num ) AS ks_dz_num_total,
    sum( f.tt_gz_num ) AS tt_gz_num_total 
FROM
    `f_like` AS f
    LEFT JOIN f_company AS c ON c.id = f.company_id
    LEFT JOIN system_user AS u ON u.id = f.user_id
    where f.create_time >= '2024-02-27 00:00:00'
    and f.create_time <= '2024-02-27 23:59:59'
    GROUP BY f.user_id,f.device_id
    ;

后来为了不联表 我直接把快照写入进去了

SELECT
    f.*,
    
    sum( f.dy_sc_num ) AS dy_sc_num_total,
    sum( f.dy_gz_num ) AS dy_gz_num_total,
    sum( f.dy_like_num ) AS dy_like_num_total,
    sum( f.ks_gz_num ) AS ks_gz_num_total,
    sum( f.ks_dz_num ) AS ks_dz_num_total,
    sum( f.tt_gz_num ) AS tt_gz_num_total 
FROM
    `f_like` AS f
    where f.create_time >= '2024-02-27 00:00:00'
    and f.create_time <= '2024-02-27 23:59:59'
    GROUP BY f.user_id,f.device_id
    ;