SELECT 
 h1.Date AS Date,
 h1.Currency AS Currency1,
 h2.Currency AS Currency2,
 h1.Account AS Account,
 SUM(h1.Size) AS Size1,
 SUM(h2.Size) AS Size2,
 h1.MaturityDate AS MaturityDate
FROM
 htable h1 JOIN htable h2 ON 
 h1.AssetClass = 'FX' AND
 h2.AssetClass = 'FX' AND
 h1.Date = h2.Date AND
 (SUBSTR(h1.Ticker, 7, 3) <> h1.Currency) AND
 (SUBSTR(h2.Ticker, 7, 3) = h2.Currency) AND
 h1.SecID = h2.SecID
GROUP BY h1.Date, h1.Currency, h2.Currency, h1.Account, h1.MaturityDate
HAVING SUM(h1.Size) <> 0 AND SUM(h2.Size) <> 0)

当查询特定日期时，上面没有GROUPBY子句的查询是快速的。
问题是，当我添加groupby时，它会变得非常慢，即使我在having子句中添加了特定的日期。
我已经为htable添加了date、assetclass、secid、currency、duritydate列的索引，以及（date、account、currency、duritydate）组合的索引。
这是在mysql中完成的。关于如何加快查询的速度，您有什么建议吗（我计划将查询用于视图定义）？谢谢！