Will there performance improvement over separate tables vs single table with multiple partitions?

Will there performance improvement over separate tables vs single table with multiple partitions? Is advisable to have separate tables than having a single big table with partitions? Can we expect same performance having single big table with partitions? What is the recommendation approach in HANA?

Suren,
first off a friendly reminder: SCN is a public forum and for you as an SAP employee there are multiple internal forums/communities/JAM groups available. You may want to consider this.
Concerning your question:
You didn't tell us what you want to do with your table or your set of tables.
As tables are not only storage units but usually bear semantics - read: if data is stored in one table it means something else than the same data in a different table - partitioned tables cannot simply be substituted by multiple tables.
Looked at it on a storage technology level, table partitions are practically the same as tables. Each partition has got its own delta store & can be loaded and displaced to/from memory independent from the others.
Generally speaking there shouldn't be too many performance differences between a partitioned table and multiple tables.
However, when dealing with partitioned tables, the additional step of determining the partition to work on is always required. If computing the result of the partitioning function takes a major share in your total runtime (which is unlikely) then partitioned tables could have a negative performance impact.
Having said this: as with all performance related questions, to get a conclusive answer you need to measure the times required for both alternatives.
- Lars

Similar Messages

2GB to 4GB: Will there be improvement?

I have a 2GHz C2D Mac Mini with 2GB of RAM. It's been working fine but I've been using Aperture lately and it brings the machine to its knees. Barring getting a new Mac, if I went to 4GB, will there be a noticeable improvement?
I know about the issue of addressing 3GB, so I'm fine w/ the 'wasted' 1GB as long as I see a performance improvement. Just trying to figure out if this is worth the investment.

Oh, and IMHO, a nice big external desktop Firewire HDD, 7200RPM or higher, will trump any internal HDD you put in the Mini, no matter how big and fast. The power and speed of a desktop HDD will even cancel out the fact that the FW port is slower than the internal SATA. A friend of mine actually took the time to measure it, and found this to be true.
An internal SSD could be even faster, but they are simply not cost effective, plus you have to use an external for storage anyway, since a 500GB SSD will set you back at least a grand. If you have that kind of money to spend, you are better off with a Mac Pro, bro'!

How to get only column names from different tables as single table columns

Hi All,
I have one requirement in which we want only column names from different tables.
for example :
I have three tables T1 ,T2, T3 having
col1 clo2 clo3 --> T1 ,
col3 col5 ,clo6 --> T2 ,
Clo6 col8 col9 --> T3
columns i want to get only all Column names from all table as single Resultset not any data from that how can i get that empty resultset
because this empty result i want to bind in datagridview(front end) as Empty resultset
Please tell me anyways to do this
Niraj Sevalkar

If I understand you want an empty result set, just with metadata. SET FMTONLY do the trick:
SET FMTONLY ON
SELECT Col1, Col2, Col3, ....., Coln
FROM
T1 CROSS JOIN T2 CROSS JOIN T3
SET FMTONLY OFF
Another alternative is to include an imposible contition
SELECT Col1, Col2, Col3, ....., Coln
FROM
T1 CROSS JOIN T2 CROSS JOIN T3
WHERE 1 = 0
If you are using a SqlDataAdapter in your client application. You can use the FillSchema method. the select command may be any select statement that returns the columns you want. Under the covers FillSchema will call SET FMTONLY ON.
If you are using SqlCommand.ExecuteReader you can pass SchemaOnly to CommandBehavior argument. SET FMTONLY ON is called under the covers. Again the select command may be any select statement that returns the columns you want.
"No darás tropezón ni desatino que no te haga adelantar camino" Bernardo Balbuena

Table Control In Modulepool with Multiple Selection.

Hi Gurus,
I have 2 Screens. In First Screen Im entering Plant and Date.
In Second Screen I have 10 Radio Buttons along with Table Control.In table control i have 6 columns.I have to add multiple rows according to radio button selection and for every row it has to carry First Screens Plant and Date also.So, In table Controls table it has to store along with radio button,Plant and Date and the Table control Values.
Pls send me the solution asap.
I'l give good points.
Thanks in Advance,
regards,
kumar.ch

Ok,good

I am proposing to upgrade the 4GB RAM in my Mac Mini 2.5 GHz core i5, by replacing it with 2 x 8GB Crucial 'slots'. Will the performance be significantly better than if I upgrade with 2 x 4GB slots? Is it cost effective, is the question.

My Mac Mini (2.5GHz Core i5) came with 4GB RAM, and I now wish to upgrade the RAM to either 8GB or 16GB in total (Crucial). Clearly, there is a price difference separating the two, but would this be justified by the extent of the improvement in performance, if I went for the more expensive option of 16GB RAM?
Secondly, if I am allowed to pose two questions: what seems to be the most compliant, reliable and cost effective external hard drive for this or Mac Mini's more generally? I was thinking of purchasing a 3TB HD for storing my rather large collection of photographs and Logic Pro 9 musical endeavours.
Your experience and advice would be appreciated, my having only recently made the switch from a Windows system to Mac.

I 150% agree with dwb and woodmeister50 about WD hard drives. They are NOWHERE near the quality they should be. When you can get a Seagate for just a few dollars more, and they last years longer, NO ONE should buy a WD. I lost $25,000 worth of music and audiobooks on a six-week-old WD in 2007 and that was the last I'd ever buy from them. They REFUSED to warranty it with the ORIGINAL sales receipt from Bestbuy.
I have 16Gb of Crucial RAM in my Mini and here's what I've discovered.
You WON'T see "lightning fast" performance increases, becasue the Mini is pretty darned fast to begin with. What you WILL see is that when you're running RAM intensive apps like Photoshop or Parallels, other apps won't "bog down".
Quite frequently I find myself with less than half a gig of free RAM out of that 16. When it was only 8 I was at 100% nearly ALL THE TIME. When I only had the 4 it came with, it was always maxed out.

Creating external table - from a file with multiple field separators

I need to create an external table from a flat file containing multiple field separators (",", ";" and "|").
Is there any way to specifiy this in the CREATE TABLE (external) statement?
FIELDS TERMINATED BY "," -- Somehow list more than just comma here?
We receive the file from a vendor every week. I am trying to set up a process for some non-technical users, and I want to keep this process transparent to them and not require them to load the data into Oracle.
I'd appreciate your help!

scott@ORA92> CREATE OR REPLACE DIRECTORY my_dir AS 'c:\oracle'
2 /
Directory created.
scott@ORA92> CREATE TABLE external_table
2    (COL1 NUMBER,
3      COL2 VARCHAR2(6),
4      COL3 VARCHAR2(6),
5      COL4 VARCHAR2(6),
6      COL5 VARCHAR2(6))
7 ORGANIZATION external
8    (TYPE oracle_loader
9      DEFAULT DIRECTORY my_dir
10    ACCESS PARAMETERS
11    (FIELDS
12        (COL1 CHAR(255)
13           TERMINATED BY "|",
14         COL2 CHAR(255)
15           TERMINATED BY ",",
16         COL3 CHAR(255)
17           TERMINATED BY ";",
18         COL4 CHAR(255)
19           TERMINATED BY ",",
20         COL5 CHAR(255)
21           TERMINATED BY ","))
22    location ('flat_file.txt'))
23 /
Table created.
scott@ORA92> select * from external_table
2 /
      COL1 COL2   COL3   COL4   COL5
         1 Field1 Field2 Field3 Field4
         2 Field1 Field2 Field3 Field4
scott@ORA92>

Although the proper options are set, firefox will not notify me "first" when I shut down firefox with multiple tabs open.

Firefox should warn me before shutting down when I have more than one tab open. Although the proper options are set, it just shuts down. VERY PAINFUL.

Would you check what you have with I've listed in item #31 at
* http://dmcritchie.mvps.org/firefox/firefox-problems.htm#tabslost
Whether you do it the old way or the new way you can always pick up your last session from the History menu. So even if choose not to restore you tabs at the end of session or are not asked you still have the option to work from the last session through the History menu.
You can make '''Firefox 8.0''' look like Firefox 3.6.*, see numbered '''items 1-10''' in the following topic [http://dmcritchie.mvps.org/firefox/firefox-problems.htm#fx4interface Fix Firefox 4.0 toolbar user interface, problems (Make Firefox 4.0 thru 8.0, look like 3.6)]. ''Whether or not you make changes, you should be aware of what has changed and what you have to do to use changed or missing features.''
* http://dmcritchie.mvps.org/firefox/firefox-problems.htm#fx4interface
There is a lot more beyond those first 10 steps listed, if you want to make Firefox more functional.
Please mark "Solved" one answer that will best help others with a similar problem -- hope this was it.

Is there any information about using iCloud on a Mac with multiple user accounts (Mountain Lion)

When upgrading to Mountain Lion, you are asked to sign in using your iCloud ID. I don't understand how to set up the proper synching of "stuff" if your Mac has multiple user accounts, and they each have their own iCloud ID.
The whole issue of supporting multiple users on a Mac and making everything easy to understand with regard to iTunes and iCloud is, in my opinion, currently in a state of disaster. Apple needs to address this. They have conveniently made it a non-issue on iPads by not allowing multiple users. I can tell you that as one of the resident iOS and Apple experts in my office, most of the questions I get from others (not work-related) are on this topic. It needs to be easier to understand and accomplish.
Thanks in advance for any information you have on this subject.

CS7981 wrote:
is "Is there any benefit to using my Airport Express with my wireless router
no.

Pls help me to modify the query for performance improvement

Hi,
I have the below initialization
DECLARE @Active bit =1 ;
Declare @id int
SELECT @Active=CASE WHEN id=@id and [Rank] ='Good' then 0 else 1 END FROM dbo.Students
I have to change this query in such a way that the conditions id=@id and [Rank] ='Good' should go to the where condition of the query. In that case, how can i use Case statement to retrieve 1 or 0? Can you please help me to modify this initialization?

I dont understand your query...May be below? or provide us sample data and your output...
SELECT * FROM dbo.students
where @Active=CASE
WHEN id=@id and rank ='Good' then 0 else 1 END
But, I doubt you will have performance improvement here?
Do you have index on id?
If you are looking for getting the data for @ID with rank ='Good' then use the below:Make sure, you have index on id,rank combination.
SELECT * FROM dbo.students
where id=@id
and rank ='Good'

HT4325 Will there be a point in time when I can multiple speaker play to my Air devices, and my AppleTvs? Or do I need redundant Air @ the tv too?

Background
I drank the coolaid and have a MacMini command center along with all my PC's and Wifi and this and that.
I have a few Airport expresses in random locations.. bathroom.. kitchen... I also have an Apple Tv in the fam room, and bedroom.
From the Mac I see I can play songs in iTunes, click Multiple speakers, and choose all my AirExpress devices. However I cannot also mutiple to the AppleTvs. Of course I can AirPlay to a single tv, or a single AirExpress.
Question
Will there be a point in time where I can choose 'Multiple Speaker', and select all my home AirExpress class devices, and the AppleTv devices?
Current Infrastucture --- Will easily support my ambient musical desires.
     Mac Mini 2012, running 10.8.
     AirPort Expresses, newest generation.
     AppleTv 3's, 5.1 not the developer 5.1+ yet.
     Whole house each floor and exterior, one SSID wifi network, multiple bands, (cisco e3000)'s running dd-wrt with a backend WDU.
     Multiple Linksys (SRW208P)'s wired switching fabric, 802.1q and all that jazz.

Ok looks like the latest updates made this possible.

Will there be a 4h mtp fw

ok one simple question as already mentioned in the subject.
i'm wondering if there will be a mtp/p4s firmware, that includes all the updates/fixes of the new v firmware.
i'd like to use mtp with the 4 hour support and everything i read so far was "hopefully" or "i think so"
are there any official statements?!?
i'm looking forward for answers and i'm sorry for my bad english

ok are there any news after two months of waiting now?
or will there be ne new v2 firmware for the micro with the 4 hour standby fix ?
looking formward to some answers

Will there be Mission Control for Java 7?

Will there be a release of Mission Control that works with Java 7? If so, when?
I understand from this page, http://blogs.oracle.com/henrik/entry/java_7_questions_answers , that there will be no JRockit for Java 7 since it is being merged with HotSpot but I see no indication when Mission Control will support the new merged JDK.

We're working on merging JRockit features into HotSpot, and supporting these with a new version of Mission Control.
Flight Recorder will come in a JDK 7 update.
JavaOne 2011 pres regarding this:
https://oracleus.wingateweb.com/scheduler/eventcatalog/eventCatalogJavaOne.do , search for "HotRockit"
Klara, Mission Control QA

Warehouse partitioning - performance of queries across multiple partitions?

Hi,
We are using Oracle 11.2.0.3 and have a large central fact table with several surrogate ids which have bitmap indexes on them and have fks looking at dimension tables + several measures
(PRODUCT_ID,
CUSTOMER_ID,
DAY_ID,
TRANS_TYPE_ID,
REGION_ID,
QTY
VALUE)
We have 2 distinct sets of queries users look to run for most part, ones accessing all transactions for products regradless of the time those transactions happened (i.e. non-financial queries - about 70%,
queries determining what happened in a particular week - 20% of queries.
Table will have approx 4bn rows in eventually.
Considering adding extra column to this DATE and range partition this to allow us to drop old partitions every year - however this data wouldn't be joined to any other table.
Then considering sub-partitioning by hash of product_id which is surrogate key for product dimension.
Thoughts on performance?
Queries by their nature would hit several sub-partitions.
Thoughts on query performance of queries which access several sub-partitions/partitions versus queries running aganist a single table.
Any other thoughts on partitioning strategy in our situation much apprecaited.
Thanks

>
Thoughts on query performance of queries which access several sub-partitions/partitions versus queries running aganist a single table.
>
Queries that access multiple partitions can improve performance for two use cases: 1) only a subset of the entire table is needed and 2) if the access is done in parallel.
Even if 9 of 10 partitions are needed that can still be better than scanning a single table containing all of the data. And when there is a logical partitioning key (transaction date) that matches typical query predicate conditions then you can get guaranteed benefits by limiting a query to only 1 (or a small number) partition when an index on a single table might not get used at all.
Conversely, if all table data is needed (perhaps there is no good partition key) and parallel option is not available then I wouldn't expect any performance benefit for either single table or partitioning.
You don't mention if you have licensed the parallel option.
>
Any other thoughts on partitioning strategy in our situation much apprecaited.
>
You provide some confusing information. On the one hand you say that 70% of your queries are
>
ones accessing all transactions for products regradless of the time those transactions happened
>
But then you add that you are
>
Considering adding extra column to this DATE and range partition this to allow us to drop old partitions every year
>
How can you drop old partitions every year if 70% of the queries need product data 'regardless of the time those transactions happened'?
What is the actual 'datetime' requirement'? And what is your definition of 'a particular week'? Does a week cross Month and Year boundaries? Does the requirement include MONTHLY, QUARTERLY or ANNUAL reporting?
Those 'boundary' requirements (and the online/offline need) are critical inputs to the best partitioning strategy. A MONTHLY partitioning strategy means that for some weeks two partitions are needed. A weekly partitioning strategy means that for some months two partitions are needed. Which queries are run more frequently weekly or monthly?
Why did you mention sub-partitioning? What benefit do you expect or what problem are you trying to address? And why hash? Hash partitioning guarantees that ALL partitions will be needed for predicate-based queries since Oracle can't prune partitions when it evaluates execution plans.
The biggest performance benefit of partitioning is when the partition keys used have a high correspondence with the filter predicates used in the queries that you run.
Contrarily the biggest management benefit of partitioning is when you can use interval partitioning to automate the creation of new partitions (and subpartitions if used) based solely on the data.
The other big consideration for partitioning, for both performance and management, is the use of global versus local indexes. WIth global indexes (e.g. a global primary key) you can't just drop a partition in isolation; the global primary key needs to be maintained by deleting the corresponding index entries.
On the other hand if your partition key includes the primary key column(s) then you can use a local index for the primary key. Then partition maintenance (drop, exchange) is very efficient.

Flashback a single table

Hi there,
Is it possible to flashback a single table in Oracle, I don't want to have to flashback the whole database to restore a single table, in the event that the data in it deleted.
Thanks...

user9101605 wrote:
in the case of flashback table case (only DML activties),data will be taken from undo tbs right?
No, flashback logs will be used.
Sorry, my mistake.
Flashback Table uses information in the undo tablespace to restore the table
http://download.oracle.com/docs/cd/B19306_01/backup.102/b14192/flashptr003.htm
Edited by: Nikolay Ivankin on 07.06.2011 17:19

5 tables Used - Performance Improvement Help Req...

Hello Experts,
I am using 5 table joins to fetch data from query; does this will degrade performance?
Please find below the query used; in this WHERE Clause, 1) le.locker_entry_id is PRIMARY Key column
2) dc.object_type - created a index on this column 3) Created a function based index on this column upper(dt.partner_device_type)
Can any other code improvement be done or view be created ?
Please suggest.
SELECT count(*) FROM locker_entry le;
COUNT(*)
1762
SELECT count(*) FROM digital_compatibility dc;
COUNT(*)
227757
SELECT count(*) FROM digital_encode_profile dep;
COUNT(*)
48
SELECT count(*) FROM device_type dt;
COUNT(*)
421
SELECT count(*) FROM digital_sku dsku;
COUNT(*)
26037
EXPLAIN PLAN FOR
SELECT
/*+ INDEX(dep, DIGITAL_ENCODE_PROFILE_ID_PK) */
DISTINCT le.locker_entry_id AS locker_entry_id,
dep.delivery_type
FROM locker_entry le,
digital_compatibility dc,
digital_encode_profile dep,
device_type dt,
digital_sku dsku
WHERE le.sku_id                   = dsku.legacy_id
AND le.digital_package_id         = dsku.digital_package_id
AND dsku.digital_sku_id           = dc.object_id
AND dc.encode_profile_id          =dep.digital_encode_profile_id
AND dt.capability_set_id          =dc.capability_set_id
AND le.locker_entry_id           IN (:1, :2, :3, :4, :5, :6, :7, :8, :9, :10, :11, :12, :13, :14, :15, :16, :17, :18, :19, :20, :21, :22, :23, :24, :25, :26, :27, :28, :29, :30, :31, :32)
AND dc.object_type                =:"SYS_B_0"
AND upper(dt.partner_device_type) =:33;
SELECT * FROM TABLE(DBMS_XPLAN.DISPLAY);
| Id | Operation                             | Name                         | Rows | Bytes | Cost (%CPU)|
|   0 | SELECT STATEMENT                      |                              |     1 |   370 |   481   (3)|
|   1 | HASH UNIQUE                          |                              |     1 |   370 |   481   (3)|
|   2 |   NESTED LOOPS                        |                              |     1 |   370 |   480   (3)|
|   3 |    NESTED LOOPS                       |                              |     1 |   336 |   479   (3)|
|* 4 |     HASH JOIN                         |                              |     1 |   193 |   478   (3)|
|   5 |      MAT_VIEW ACCESS BY INDEX ROWID   | DIGITAL_SKU                  |     1 |    48 |     5   (0)|
|   6 |       NESTED LOOPS                    |                              |    16 | 1392 |     5   (0)|
|   7 |        INLIST ITERATOR                |                              |       |       |            |
|   8 |         TABLE ACCESS BY INDEX ROWID   | LOCKER_ENTRY                 |    32 | 1248 |     1   (0)|
|* 9 |          INDEX RANGE SCAN             | LOCKER_ENTRY_ID_PK           |    32 |       |     1   (0)|
| 10 |        BITMAP CONVERSION TO ROWIDS    |                              |       |       |            |
| 11 |         BITMAP AND                    |                              |       |       |            |
| 12 |          BITMAP CONVERSION FROM ROWIDS|                              |       |       |            |
|* 13 |           INDEX RANGE SCAN            | IDX_DIGITAL_SKU_LEGACY_ID    |     1 |       |     1   (0)|
| 14 |          BITMAP CONVERSION FROM ROWIDS|                              |       |       |            |
|* 15 |           INDEX RANGE SCAN            | IDX_DIGITAL_PACKAGE_ID       |     1 |       |     1   (0)|
|* 16 |      MAT_VIEW ACCESS FULL             | DIGITAL_COMPATIBILITY        | 2098 |   217K|   472   (3)|
|* 17 |     INDEX RANGE SCAN                  | DEVICE_TYPE_IDX              |     1 |   143 |     1   (0)|
| 18 |    MAT_VIEW ACCESS BY INDEX ROWID     | DIGITAL_ENCODE_PROFILE       |     1 |    34 |     1   (0)|
|* 19 |     INDEX UNIQUE SCAN                 | DIGITAL_ENCODE_PROFILE_ID_PK |     1 |       |     1   (0)|
Predicate Information (identified by operation id):
   4 - access("DSKU"."DIGITAL_SKU_ID"=TO_NUMBER("DC"."OBJECT_ID"))
   9 - access("LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:1) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:2) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:3) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:4) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:5) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:6) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:7) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:8) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:9) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:10) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:11) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:12) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:13) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:14) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:15) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:16) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:17) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:18) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:19) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:20) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:21) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:22) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:23) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:24) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:25) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:26) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:27) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:28) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:29) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:30) OR
              "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:31) OR "LE"."LOCKER_ENTRY_ID"=TO_NUMBER(:32))
13 - access("LE"."SKU_ID"="DSKU"."LEGACY_ID")
15 - access("LE"."DIGITAL_PACKAGE_ID"="DSKU"."DIGITAL_PACKAGE_ID")
16 - filter("DC"."OBJECT_TYPE"=:SYS_B_0)
17 - access("DT"."CAPABILITY_SET_ID"="DC"."CAPABILITY_SET_ID" AND
              UPPER("PARTNER_DEVICE_TYPE")=:33)
19 - access("DC"."ENCODE_PROFILE_ID"="DEP"."DIGITAL_ENCODE_PROFILE_ID")
Note
   - 'PLAN_TABLE' is old version
Trace information
=================
recursive calls     17
db block gets     0
consistent gets     239
physical reads     61
redo size     0
bytes sent via SQL*Net to client     742
bytes received via SQL*Net from client     1361
SQL*Net roundtrips to/from client     2
sorts (memory)     5
sorts (disk)     0
   Edited by: Linus on Oct 10, 2011 2:44 AM

Linus wrote:
Yes Bravid i got regarding the point to remove the hint from the query; unless there is a change in execution plan with index hint.
My concern is; is there any issues by keeping 5 tables on single query and will this degrade performance?
Because as of during the certification practise; i used to overcome some where ,"NOT to use many table joins".
So that is reason i am asking you; sorry if am wrong in any ways.
Thanks...There's nothing inherently wrong with joining lots of tables and there isn't one specific thing that will degrade performance. You could have a query that joins 2 tables performing very badly and a query with 10 tables that performs very well. A lot of it is down to the quality of the decisions the optimiser can make. To get the best decisions out of it you need to make sure it had enough information to work with. So that means making sure stats are available and relevant for the data you're querying and removing any hints unless you have a specific reason to use them.
Earlier this year I had the "joy" of working with a query that had 65 tables in it. It was an INSERT INTO...SELECT FROM and it was a clear example of where a single statement really should have been about 6 or 7 separate statements. The reason in this case was because there were 6 or 7 different sections to the query that had essentially no relationship with each other and they were all outer joined and used lots of analytic functions and case statements to categorise each row and populate the same columns with different values depending on which query block it had originated from.
This is an extreme example but the point is you have to look at the statement and decide whether it does it's job well. My personal preference is to try to avoid big "generic" SQL statements that cater for lots of different scenarios because they can easily become over complicated and difficult to maintain and tune.
HTH
David
Edited by: Bravid on Oct 10, 2011 1:53 PM

Will there performance improvement over separate tables vs single table with multiple partitions?

Similar Messages

Maybe you are looking for