473,898 Members | 2,221 Online
Bytes | Software Development & Data Engineering Community
+ Post

Home Posts Topics Members FAQ

Questions regarding large MDC tables

Hi all,

I'm currently investigating the use of MDC Tables for large data warehouse
tables.
My scenario:

A fact table with 1000 Million Rows distributed on 12 Partitions (3 physical
hosts with 4 logical partitions each).
Overall size of table is 350 GB . Each night 1.5 Million new rows will be
added
and approx. the same amount of old records will be deleted (Roll in/Roll out
with SQL INSERT/DELETE).
The table is stored in SMS tablespace with 16K Pagesize and 64 Pages
Extentsize.
The tablespace has 6 containers on each partition. Each container is on a
separate IBM ESS array. Prefetchsize is 384 (6 containers * 64 pages).
Prefetch behaves very well
with these settings (DB2_PARALLEL_I O is set). DB2 is V8.1 ESE (DPF) FP5 and
runs on AIX 5.2.

We figured out that for our choosen MDC dimensions we will have to use an
extentsize of 2 pages otherwise we would waste too much space. This
extentsize gives us headaches:

- What is an optimal prefetchsize here?
With prefetchsize of 12 (6 containers * extentsize 2) each prefetcher will
only read 32 KB of data from one container.
With a prefetchsize of 384 (which is optimal from a disk layout point of
view) will DB2 start 192 prefetchers (that would be certainly overkill)?

Further:
Does anybody have experiences with MDC tables for large warehouse tables and
is willing to share them?
Especially performance experiences when inserting/deleting during the
roll-in/roll-out of daily data ?

Unfortunately I do not have an adequate environment to test all these
issues - so any comments are highly appreciated.

TIA
Joachim

PS: Feel free to send comments by email to joklassen at web dot de
Nov 12 '05 #1
4 3421
Joachim Klassen wrote:
Hi all,

I'm currently investigating the use of MDC Tables for large data warehouse
tables.
My scenario:

A fact table with 1000 Million Rows distributed on 12 Partitions (3 physical
hosts with 4 logical partitions each).
Overall size of table is 350 GB . Each night 1.5 Million new rows will be
added
and approx. the same amount of old records will be deleted (Roll in/Roll out
with SQL INSERT/DELETE).
The table is stored in SMS tablespace with 16K Pagesize and 64 Pages
Extentsize.
The tablespace has 6 containers on each partition. Each container is on a
separate IBM ESS array. Prefetchsize is 384 (6 containers * 64 pages).
Prefetch behaves very well
with these settings (DB2_PARALLEL_I O is set). DB2 is V8.1 ESE (DPF) FP5 and
runs on AIX 5.2.

We figured out that for our choosen MDC dimensions we will have to use an
extentsize of 2 pages otherwise we would waste too much space. This
extentsize gives us headaches:

- What is an optimal prefetchsize here?
With prefetchsize of 12 (6 containers * extentsize 2) each prefetcher will
only read 32 KB of data from one container.
With a prefetchsize of 384 (which is optimal from a disk layout point of
view) will DB2 start 192 prefetchers (that would be certainly overkill)?

Further:
Does anybody have experiences with MDC tables for large warehouse tables and
is willing to share them?
Especially performance experiences when inserting/deleting during the
roll-in/roll-out of daily data ?

Unfortunately I do not have an adequate environment to test all these
issues - so any comments are highly appreciated.

TIA
Joachim

PS: Feel free to send comments by email to joklassen at web dot de

Try the new advisor in V8.2. You can simply download DB2 V8.2 onto your
laptop, mimic the stats and then apply the proposal to DB2 V8.1.5.

Cheers
Serge

--
Serge Rielau
DB2 SQL Compiler Development
IBM Toronto Lab
Nov 12 '05 #2
Serge,
thanks for the quick reply.

I already followed your suggestion but with no luck. Design advisor makes no
recommendations for MDC.
Does that mean that MDC is not recommended at all for my scenario or do I
have to refine my workload input (which is at the moment very basic with 3
Queries and 1 Insert statement).

Thanks again
Joachim

"Serge Rielau" <sr*****@ca.ibm .com> schrieb im Newsbeitrag
news:35******** *****@individua l.net...
Joachim Klassen wrote:
Hi all,

I'm currently investigating the use of MDC Tables for large data
warehouse
tables.
My scenario:

A fact table with 1000 Million Rows distributed on 12 Partitions (3
physical hosts with 4 logical partitions each).
Overall size of table is 350 GB . Each night 1.5 Million new rows will be
added
and approx. the same amount of old records will be deleted (Roll in/Roll
out with SQL INSERT/DELETE).
The table is stored in SMS tablespace with 16K Pagesize and 64 Pages
Extentsize.
The tablespace has 6 containers on each partition. Each container is on a
separate IBM ESS array. Prefetchsize is 384 (6 containers * 64 pages).
Prefetch behaves very well
with these settings (DB2_PARALLEL_I O is set). DB2 is V8.1 ESE (DPF) FP5
and runs on AIX 5.2.

We figured out that for our choosen MDC dimensions we will have to use an
extentsize of 2 pages otherwise we would waste too much space. This
extentsize gives us headaches:

- What is an optimal prefetchsize here?
With prefetchsize of 12 (6 containers * extentsize 2) each prefetcher
will
only read 32 KB of data from one container.
With a prefetchsize of 384 (which is optimal from a disk layout point of
view) will DB2 start 192 prefetchers (that would be certainly overkill)?

Further:
Does anybody have experiences with MDC tables for large warehouse tables
and
is willing to share them?
Especially performance experiences when inserting/deleting during the
roll-in/roll-out of daily data ?

Unfortunately I do not have an adequate environment to test all these
issues - so any comments are highly appreciated.

TIA
Joachim

PS: Feel free to send comments by email to joklassen at web dot de

Try the new advisor in V8.2. You can simply download DB2 V8.2 onto your
laptop, mimic the stats and then apply the proposal to DB2 V8.1.5.

Cheers
Serge

--
Serge Rielau
DB2 SQL Compiler Development
IBM Toronto Lab

Nov 12 '05 #3
Joachim,

Sure it does. The V8.2 advisor handles Indexes, Partitioning, MQT, and
MDC. You sure you tried V8.2?
MDC will not improve your insert performance. It can help with delete
(rollout) and query.

Cheers
Serge
--
Serge Rielau
DB2 SQL Compiler Development
IBM Toronto Lab
Nov 12 '05 #4
"Serge Rielau" <sr*****@ca.ibm .com> schrieb im Newsbeitrag
news:35******** *****@individua l.net...
Joachim,

Sure it does. The V8.2 advisor handles Indexes, Partitioning, MQT, and
MDC. You sure you tried V8.2?
MDC will not improve your insert performance. It can help with delete
(rollout) and query.

Cheers
Serge
--
Serge Rielau
DB2 SQL Compiler Development
IBM Toronto Lab

Serge
Yes, I tried V8.2 (V8.1 FP7). I will retry with FP8 and a refined workload.
But maybe I use it wrong. Here is what I've done so far:
- captured objects ddl (tables and indexes) and statistics from the original
tables via db2look
- recreated the objects on my laptop in a non-partitioned instance (maybe
thats the problem)
- inserted some thousand records and run RUNSTATS to get entries in
SYSCOLDIST etc.. to be able to update them later
- updated the stats with the original values
- started Design advisor GUI and defined a workload consisting of 3 typical
queries and 1 insert
- started recommendations

I thought that MDC would not help with insert performance (more likely the
opposite ?) thats why we consider LOAD FROM CURSOR.
Our main goal is improvement for roll-out. Queries already perform well.
But the more informations I find about MDC in a partitioned environment the
more I doubt that they will help us in this special scenario.

Thanks for your comments
Joachim
Nov 12 '05 #5

This thread has been closed and replies have been disabled. Please start a new discussion.

Similar topics

0
1348
by: Jack Coxen | last post by:
------_=_NextPart_001_01C3584E.5FF65B60 Content-Type: text/plain; charset="iso-8859-1" I've gone through the mailing list archives, read the appropriate parts of the manual and searched the internet and still have questions. What I'm running into is a serious performance problem with a fairly large (by my standards) database. My system setup is as follows: Dual processor Sun E250 w/1.5GB RAM running Solaris 8 (with all current
2
3743
by: Warren Wright | last post by:
Hi All, First, where can I get some questions of this sort answered? Preferably, are there good books or online guides that I can consult for these types of answers when necessary? 1. How do I know, when executing a query from Query Analyzer or otherwise, how many temporary tables will be needed, and how big they will be? 2. Where will those temporary tables be created? Always in tempdb?
1
3251
by: Invalidlastname | last post by:
Hi, I have some questions regarding to use vs.net to generate xsd from database tables then generate classes, not typed datasets, from the xsd. Basically I want to have some light-weigh classes to represent the database tables/views (I know tables/view are not business entities, but this is a good starting point and i am lazy =) Here are what I did: - used xsd.exe to generate business entities, which will be used to be passed between...
8
5285
by: Mike | last post by:
Hello, I have a few rather urgent questions that I hope someone can help with (I need to figure this out prior to a meeting tomorrow.) First, a bit of background: The company I work for is developing a web-based application, one part of which involves allowing the user the ability to page through transaction "history" information. The _summary_ history table will have the following fields: ServiceName, Date, User-Ref1, User-Ref2,...
1
4080
by: mairhtin o'feannag | last post by:
Hello, After reading the fine article on the Stinger enhancements for Linux, I have two questions. First, our application is in essence a warehouse, with 100-400 million rows for various tables. Having said that, the AIO enhancements seem to be geared to an OLTP environment. Our environment is that we load MASSIVE amounts of data every couple of months, and then run various queries against this massive influx. Seems to me that our...
4
1786
by: Bob Alston | last post by:
Some more, rather specific Access performance questions. IN a split front-end & back-end Access/Jet ONLY LAN situation, and with all query criteria fields and join fields indexed: 1. Is is good form to have a single query with base table with criteria joined to a related table - all in one query? Or should I do a two-step, first query does selection of main table and then join with other table? 2. I have a table with multiple...
4
2517
by: Drew | last post by:
I posted this to the asp.db group, but it doesn't look like there is much activity on there, also I noticed that there are a bunch of posts on here pertaining to database and asp. Sorry for cross-posting. I am trying to build a "checklist", where a user can navigate to an ASP page on the intranet which shows a list of "questions" that the user can check off. I am trying to figure out how to do this so that it is scalable, but I am...
6
1404
by: Phil Sandler | last post by:
All, I am designing a system that will involve an IIS/ASP.Net application server. The main purpose of the application server will be to load large amounts of static data into memory and do fast lookups on it. The total cache size at the time of first installation will be roughly 350MB, but may grow to as large as 10GB over the next five years. My questions:
22
3694
by: Jesse Burns | last post by:
I'm about to start working on my first large scale site (in my opinion) that will hopefully have 1000+ users a day. ok, this isn't on the google/facebook scale, but it's going to be have more hits than just family and friends. Either way, I'm planning on this site blowing up once I have enough of a feature set, so I'm concerned about performance and scalability in the long run. I've worked for a software company, but I've never...
0
9841
by: Hystou | last post by:
Most computers default to English, but sometimes we require a different language, especially when relocating. Forgot to request a specific language before your computer shipped? No problem! You can effortlessly switch the default language on Windows 10 without reinstalling. I'll walk you through it. First, let's disable language synchronization. With a Microsoft account, language settings sync across devices. To prevent any complications,...
0
11265
Oralloy
by: Oralloy | last post by:
Hello folks, I am unable to find appropriate documentation on the type promotion of bit-fields when using the generalised comparison operator "<=>". The problem is that using the GNU compilers, it seems that the internal comparison operator "<=>" tries to promote arguments from unsigned to signed. This is as boiled down as I can make it. Here is my compilation command: g++-12 -std=c++20 -Wnarrowing bit_field.cpp Here is the code in...
1
10949
by: Hystou | last post by:
Overview: Windows 11 and 10 have less user interface control over operating system update behaviour than previous versions of Windows. In Windows 11 and 10, there is no way to turn off the Windows Update option using the Control Panel or Settings app; it automatically checks for updates and installs any it finds, whether you like it or not. For most users, this new feature is actually very convenient. If you want to control the update process,...
0
9662
agi2029
by: agi2029 | last post by:
Let's talk about the concept of autonomous AI software engineers and no-code agents. These AIs are designed to manage the entire lifecycle of a software development project—planning, coding, testing, and deployment—without human intervention. Imagine an AI that can take a project description, break it down, write the code, debug it, and then launch it, all on its own.... Now, this would greatly impact the work of software developers. The idea...
0
7187
by: conductexam | last post by:
I have .net C# application in which I am extracting data from word file and save it in database particularly. To store word all data as it is I am converting the whole word file firstly in HTML and then checking html paragraph one by one. At the time of converting from word file to html my equations which are in the word document file was convert into image. Globals.ThisAddIn.Application.ActiveDocument.Select();...
0
6077
by: adsilva | last post by:
A Windows Forms form does not have the event Unload, like VB6. What one acts like?
1
4706
by: 6302768590 | last post by:
Hai team i want code for transfer the data from one system to another through IP address by using C# our system has to for every 5mins then we have to update the data what the data is updated we have to send another system
2
4296
muto222
by: muto222 | last post by:
How can i add a mobile payment intergratation into php mysql website.
3
3306
bsmnconsultancy
by: bsmnconsultancy | last post by:
In today's digital era, a well-designed website is crucial for businesses looking to succeed. Whether you're a small business owner or a large corporation in Toronto, having a strong online presence can significantly impact your brand's success. BSMN Consultancy, a leader in Website Development in Toronto offers valuable insights into creating effective websites that not only look great but also perform exceptionally well. In this comprehensive...

By using Bytes.com and it's services, you agree to our Privacy Policy and Terms of Use.

To disable or enable advertisements and analytics tracking please visit the manage ads & tracking page.