大量查询内存使用情况

Question

我正在尝试找到一种方法来提高下采样查询内存的效率。

当前，我的查询如下所示，执行时会导致一些很高的内存峰值：

SELECT mean("value") as "mean", min("value") as "min", max("value") as "max", mode("value") as "mode"
INTO "hour"."variables"
FROM "minute"."variables"
WHERE time > now() - 1d
AND time <= now()
GROUP BY "name", time(60m)

对于每个聚合，内存在4个单独的步骤中执行下采样查询会更有效吗？

SELECT mean("value") as "mean"
INTO "hour"."variables"
FROM "minute"."variables"
WHERE time > now() - 1d
AND time <= now()
GROUP BY "name", time(60m)

SELECT min("value") as "min"
INTO "hour"."variables"
FROM "minute"."variables"
WHERE time > now() - 1d
AND time <= now()
GROUP BY "name", time(60m)

SELECT max("value") as "max"
INTO "hour"."variables"
FROM "minute"."variables"
WHERE time > now() - 1d
AND time <= now()
GROUP BY "name", time(60m)

SELECT mode("value") as "mode"
INTO "hour"."variables"
FROM "minute"."variables"
WHERE time > now() - 1d
AND time <= now()
GROUP BY "name", time(60m)