T-SQL for finding consecutive increasing values - sql-server-2008

Question

Let's say I have the following very simple schema:
Create Table MyTable (
PrimaryKey int,
Column1 datetime.
Column2 int
)
I need a query that orders the data based on Column1, and finds the first 10 consecutive rows where the value of Column2 in the current row is greater than the value of column2 in the prior row.

Answer 1

Q is used to get a ranking value rn ordered by Column1. Added in PrimaryKey in case there are ties in Column1. C is a recursive CTE that loops from the top ordered by rn incrementing cc for each increasing value of Column2. It will break from the recursion when cc reaches 10. Finally get the last 10 rows from C. The where clause takes care of the case when there are no 10 consecutive increasing values.
with Q as
(
select PrimaryKey,
Column1,
Column2,
row_number() over(order by Column1, PrimaryKey) as rn
from MyTable
),
C as
(
select PrimaryKey,
Column1,
Column2,
rn,
1 as cc
from Q
where rn = 1
union all
select Q.PrimaryKey,
Q.Column1,
Q.Column2,
Q.rn,
case
when Q.Column2 > C.Column2 then C.cc + 1
else 1
end
from Q
inner join C
on Q.rn - 1 = C.rn
where C.cc < 10
)
select top 10 *
from C
where 10 in (select cc from C)
order by rn desc
option (maxrecursion 0)
Version 2
As Martin Smith pointed out in a comment, the above query has really bad performance. The culprit is the first CTE. The version below use table variable to hold the ranked rows. The primary key directive on rn creates an index that will be used in the join in the recursive part of the query. Apart from the table variable this does the same as above.
declare #T table
(
PrimaryKey int,
Column1 datetime,
Column2 int,
rn int primary key
);
insert into #T
select PrimaryKey,
Column1,
Column2,
row_number() over(order by Column1, PrimaryKey) as rn
from MyTable;
with C as
(
select PrimaryKey,
Column1,
Column2,
rn,
1 as cc
from #T
where rn = 1
union all
select T.PrimaryKey,
T.Column1,
T.Column2,
T.rn,
case
when T.Column2 > C.Column2 then C.cc + 1
else 1
end
from #T as T
inner join C
on T.rn = C.rn + 1
where C.cc < 10
)
select top 10 *
from C
where 10 in (select cc from C)
order by rn desc
option (maxrecursion 0)