Use lateral joins to improve performance

Milliseconds instead of seconds \o/.

check_df_files

@@ -12,34 +12,52 @@ db = psycopg2.connect('')
 csr = db.cursor(cursor_factory=psycopg2.extras.DictCursor)
 
 csr.execute(
-	"""
-	with recursive t as (
-	    select '2 minutes'::interval as d
-	    union
-	    select d * 2 from t where d < '1 year'::interval
-	),
-	a as (
-	    select distinct hostname, filesystem, min(ts), max(ts)
-	    from df, t
-	    where ts >= now() - t.d
-	    group by hostname, filesystem, t.d
-	),
-	forecast as (
-	    select
-		a.hostname, a.filesystem,
-		c.ts - b.ts as d,
-		c.ts + (c.ts - b.ts) as when,
-		c.f_used + (c.f_used - b.f_used) as used,
-		(c.f_used + (c.f_used - b.f_used)) / b.f_usable * 100 as percent,
-		(c.f_used + (c.f_used - b.f_used)) > b.f_usable as capacity_exceeded
-		from a
-		    join df b on (a.hostname = b.hostname and a.filesystem = b.filesystem and a.min = b.ts)
-		    join df c on (a.hostname = c.hostname and a.filesystem = c.filesystem and a.max = c.ts)
-	)
-	select * from forecast
-	where capacity_exceeded
-	order by 1, 2, 3
-	"""
+        """
+        with recursive
+            t as (
+                select '2 minutes'::interval as d
+                union
+                select d * 2 from t where d < '1 year'::interval
+            ),
+            fs as (
+                select distinct hostname, filesystem
+                    from df
+                    where ts >= now() - '2 minutes'::interval
+                    group by hostname, filesystem
+            ),
+            forecast as (
+                select
+                    fs.hostname, fs.filesystem,
+                    new.ts + (new.ts - old.ts) as when,
+                    new.ts + (new.ts - old.ts) - now() as d,
+                    new.f_used + (new.f_used - old.f_used) as used,
+                    (new.f_used + (new.f_used - old.f_used)) / old.f_usable * 100 as percent,
+                    (new.f_used + (new.f_used - old.f_used)) > old.f_usable as capacity_exceeded
+                from fs, t,
+                    lateral (
+                        select *
+                        from df
+                        where
+                            df.hostname = fs.hostname
+                            and df.filesystem = fs.filesystem
+                            and df.ts >= now() - t.d
+                        order by ts
+                        limit 1
+                    ) old,
+                    lateral (
+                        select *
+                        from df
+                        where
+                            df.hostname = fs.hostname
+                            and df.filesystem = fs.filesystem
+                        order by ts desc
+                        limit 1
+                    ) new
+            )
+        select * from forecast
+        where capacity_exceeded
+        order by "when"
+        """
 )
 
 status = 0

check_df_space

@@ -12,34 +12,52 @@ db = psycopg2.connect('')
 csr = db.cursor(cursor_factory=psycopg2.extras.DictCursor)
 
 csr.execute(
-	"""
-	with recursive t as (
-	    select '2 minutes'::interval as d
-	    union
-	    select d * 2 from t where d < '1 year'::interval
-	),
-	a as (
-	    select distinct hostname, filesystem, min(ts), max(ts)
-	    from df, t
-	    where ts >= now() - t.d
-	    group by hostname, filesystem, t.d
-	),
-	forecast as (
-	    select
-		a.hostname, a.filesystem,
-		c.ts - b.ts as d,
-		c.ts + (c.ts - b.ts) as when,
-		c.s_used + (c.s_used - b.s_used) as used,
-		(c.s_used + (c.s_used - b.s_used)) / b.s_usable * 100 as percent,
-		(c.s_used + (c.s_used - b.s_used)) > b.s_usable as capacity_exceeded
-		from a
-		    join df b on (a.hostname = b.hostname and a.filesystem = b.filesystem and a.min = b.ts)
-		    join df c on (a.hostname = c.hostname and a.filesystem = c.filesystem and a.max = c.ts)
-	)
-	select * from forecast
-	where capacity_exceeded
-	order by 1, 2, 3
-	"""
+        """
+        with recursive
+            t as (
+                select '2 minutes'::interval as d
+                union
+                select d * 2 from t where d < '1 year'::interval
+            ),
+            fs as (
+                select distinct hostname, filesystem
+                    from df
+                    where ts >= now() - '2 minutes'::interval
+                    group by hostname, filesystem
+            ),
+            forecast as (
+                select
+                    fs.hostname, fs.filesystem,
+                    new.ts + (new.ts - old.ts) as when,
+                    new.ts + (new.ts - old.ts) - now() as d,
+                    new.s_used + (new.s_used - old.s_used) as used,
+                    (new.s_used + (new.s_used - old.s_used)) / old.s_usable * 100 as percent,
+                    (new.s_used + (new.s_used - old.s_used)) > old.s_usable as capacity_exceeded
+                from fs, t,
+                    lateral (
+                        select *
+                        from df
+                        where
+                            df.hostname = fs.hostname
+                            and df.filesystem = fs.filesystem
+                            and df.ts >= now() - t.d
+                        order by ts
+                        limit 1
+                    ) old,
+                    lateral (
+                        select *
+                        from df
+                        where
+                            df.hostname = fs.hostname
+                            and df.filesystem = fs.filesystem
+                        order by ts desc
+                        limit 1
+                    ) new
+            )
+        select * from forecast
+        where capacity_exceeded
+        order by "when"
+        """
 )
 
 status = 0