RSS 2.0
Sign In
# Friday, September 7, 2012

We're implementing UDT changes in the big database. Earlier, that User Defined Type was based on smallint, and now we have to use int as the base.

The impact here is manyfold:

  1. Clients of the database should be prepared to use wider types.
  2. All stored procedures, functions, triggers, and views should be updated accordingly.
  3. Impact on the database size should be analyzed.
  4. Types of columns in tables should be changed.
  5. Performance impact should be minimal.

Now, we're trying to address (3), (5) and to implement (4), while trying to keep interface with clients using old types.

As for database size impact, we have found that an index fragmentation is a  primary disk space waster (see Reorganize index in SQL Server). We have performed some partial index reorganization and can see now that we can gain back hundreds of GB of a disk space. On the other hand we use page compression, so we expect that change of types will not increase sizes of tables considerably. Indeed, our measurments show that tables will only be ~1-3% bigger.

The change of types of columns is untrivial task. The problem is that if you try to change column's type (which is part of clustered index) directly then you should temporary remove foreign keys, and to rebuild all indices. This won't work neither due to disk space required for the operation (a huge transaction log is required), nor due to availability of tables (we're talking about days or even weeks to rebuild indices).

To work-around the problem we have selected another way. For each target table T we performed the following:

  • Renamed table T to T_old;
  • Created a table T_new with required type changes;
  • Created a view named T, which is union of T_old for the dates before a split date and T_new for the dates after the split date;
  • Created instead of insert/update/delete triggers for the view T.
  • Created a procedures that move data in bulks from T_old to the T_new, update split date in view definitions, and delete data from T_old.

Note that:

  • the new view uses wider column types, so we had to change stored procedures that clients use to cast those columns back to shorter types to prevent side effects (fortunately all access to this database is through stored procedures and functions);
  • the procedures that transfer data between new and old tables may work online;
  • the quality of execution plans did not degrade due to switch from table to a view;
  • all data related to the date after the split date are inserted into T_new table.

After transfer will be complete we shall drop T_old tables, and T views, and will rename T_new tables into T.

This will complete part 4 of the whole task. Our estimations are that it will take a month or even more to complete the transfer. However solution is rather slow, the database will stay online whole this period, which is required condition.

The next task is to deal with type changes in parameters of stored procedures and column types of output result sets. We're not sure yet what's the best way to deal with it, and probably shall complain about in in next posts.

Friday, September 7, 2012 8:57:36 PM UTC  #    Comments [2] -
SQL Server puzzle | Thinking aloud | Tips and tricks
Monday, April 29, 2013 11:18:42 AM UTC
Hello... Very interesting post! I like it. I am going through a similar experience and would like to test your process in my dev env. I have a brief questions regarding the Trigger and View. What type of DDL (logic) are you using for the View and Trigger...Partitioned view? Thanks again for any assistance.-Andrew
Monday, April 29, 2013 4:23:22 PM UTC
Hello, Andrew!

Hopefully, we have finished that transition in one month.
Now, when I'm re-reading this post I think wonder if anyone can understand it. :-)

As for your question.
Our tables we already partitioned using partiton functions.

The views we have defined for the migration were partitioned in the "old style" like this:

select
Date,
...
cast(Field as int)
from
T_old
where
Date < 'yyyy-mm-dd'
union all
select
Date,
...
cast(Field as int)
from
T_new
where
Date >= 'yyyy-mm-dd'

The view's trigger were implementing correct inserts into either T_old or T_new.

All comments require the approval of the site owner before being displayed.
Name
E-mail
Home page

Comment (Some html is allowed: a@href@title, b, blockquote@cite, em, i, strike, strong, sub, super, u) where the @ means "attribute." For example, you can use <a href="" title=""> or <blockquote cite="Scott">.  

[Captcha]Enter the code shown (prevents robots):

Live Comment Preview
Archive
<September 2012>
SunMonTueWedThuFriSat
2627282930311
2345678
9101112131415
16171819202122
23242526272829
30123456
Statistics
Total Posts: 387
This Year: 3
This Month: 0
This Week: 0
Comments: 1795
Locations of visitors to this page
Disclaimer
The opinions expressed herein are our own personal opinions and do not represent our employer's view in anyway.

© 2024, Nesterovsky bros
All Content © 2024, Nesterovsky bros
DasBlog theme 'Business' created by Christoph De Baene (delarou)