Why is there increased overhead(performance) in NAV than DISPLAY attribute?

Hi,
I understand that navigational attibutes lead to increased overhead because of the additional joins: The join between Fact table and dimension table ( I guess on DIM ID) and then the additional join between the Dimension and the Master Data tables (I guess on the SID).
1.
What happens in the case where the attribute is NOT a Nav but DISPLAY? Can you modify the above paragraph for me if it were a DISPLAY attribute? Isn't the display attribute also in the Master Data table and won't there also be an additional join?
2.
So when a query runs, which of the joins takes place first? I am trying to get the order whether it is from inside to the outside or outside to the inside? When you execute a query does the join between the master table and dimension table take place first then the join between the dimension and fact table next? Or the reverse?
Explanation with examples will be very help.
Thanks
Edited by: Amanda Baah on May 10, 2011 10:42 AM

hi,
I understand that navigational attibutes lead to increased overhead because of the additional joins: The join between Fact table and dimension table ( I guess on DIM ID) and then the additional join between the Dimension and the Master Data tables (I guess on the SID).
The above understanding is correct
1.
What happens in the case where the attribute is NOT a Nav but DISPLAY? Can you modify the above paragraph for me if it were a DISPLAY attribute? Isn't the display attribute also in the Master Data table and won't there also be an additional join?
In both the cases the joins formed will be as per the above statement, however in case of navigational attribute the query fetching the data stores the data independent of master data such that you are able to do navigation/drill down independently on the values.
2.
So when a query runs, which of the joins takes place first? I am trying to get the order whether it is from inside to the outside or outside to the inside? When you execute a query does the join between the master table and dimension table take place first then the join between the dimension and fact table next? Or the reverse?
the schema structure is fact table at center then the dimension tables and then the SID tables. when query is executed it first fetches data from Fact table then retrieves entries corresponding to those in DIM tables and after that SID tables.
regards,
Arvind.

Similar Messages

Why is there increase in monthly billing

Why does current bill reflect an increase ? It has been $101.49 and this month's billing is for $120.14, an increase of $18.65. I haven't made any changes to my account.
Thank you for any assistance you can provide.

It seems like you made a change to your account. This may be a prorated month. Only you can check your bill details to tell.

Navi attri & display attribute

Hi experts
Could any one explain clearly about the function and use of -
Navigational Attributes
Display Attributes.
I am very thankful if any one explain about...
Thanks
Deepali

Hi,
You can go through the below documents.
http://help.sap.com/saphelp_scm70/helpdata/EN/58/2b8e39e1ba3541e10000000a11402f/frameset.htm
http://help.sap.com/saphelp_scm70/helpdata/EN/43/d7419fb96c0236e10000000a11466f/frameset.htm
http://help.sap.com/saphelp_scm70/helpdata/EN/07/9a7bce1d4811d2b42e006094b981ac/frameset.htm
Display and navigational Attributes
Hope this can help.
regards,
kaushik

Is there a real increase of performance using a 802.11ac router with a MacBook Pro 13 Retina?

Hi guys!
First of all, sorry for my english. I'm trying to improve myself!
I would like to know if could be a smart choice to change my router with a "802.11ac" router. Is there a real increase of performance?
Right now i'm using a 10mb in download and 10mb in upload connection (fibre optic).
Thank you!
Rocco

Your download speed from your ISP is not changed 10 MB download/upload speed.
http://www.speedtest.net/
Your transmit rate from router to computer will increase with simultaneous dual band 802.11ac
hold the option key and click on your WiFI in the upper right screen to see a breakdown.
802.11n will be held back

Report burst:To increase query performance in xcelsius

Is there anyway to increase query performance in xcelsius by using report bursting

Fremlin,
Report bursting is only for distributing your reports to your end users.
You can improve performance only by following the [Best practices|https://www.sdn.sap.com/irj/boc/index?rid=/library/uuid/a084a11c-6564-2b10-79ac-cc1eb3f017ac] in xcelsius.
-Anil

After installing Mountain Lion, why is there Yahoo access for one user but not another?--both are administrators.

After installing Mountain Lion, why is there Yahoo access for one user but not another?--both are administrators.

We've had several instances where we have had to run chkdsk on arrays with over 1m files. Average completion time is approximately 72 hours. The maximum downtime window they have available is the 64 hour weekend window. File sizes and number of files were
much smaller then than they are now.
The idea, in theory, was to use VHDs to compartmentalize the data into smaller volumes which could be more easily managed. It would also improve performance when transferring these compartments of data as they would use sequential read/write rather than
fragmented/random. This idea was never fleshed out in entirety, they don't split data up into little containers, but simply into big ones per project. Hence the 11m files in one container that I am currently trying to diagnose.
Some other important facts: The VHD in question is mounted in B:/project/ as this server also allows remote workers to log in, but they are restricted to see only data in E:. Disks A-D are hidden via group policy.
Update: icacls is failing on a large number of files within this dataset. I counted the path characters to ensure it wasn't the 255 character limit I was encountering and verified that the paths being blocked are only about 150 characters long. Once it finishes,
I'll have to try taking ownership and then re-running it. At this point I still have no idea how long to expect. I'm running out of time as the environment will be in use again at 9AM tomorrow morning.

How to increase the performance in server 2008 R2 for RDP users

Hi,
My application take to much time to load. If anyone double click on mail client the exe file will appear in task manager but it will open after 5 mins. how to increase the performance.
My sever configuration is as below,
SC2600 Intel motherboard with total 24 core processors and 32 GB RAM and 8 TB Hard Disk. RAID 5 is configured which has two lungs one is 167 GB for C drive and other is 4.5 TB for D drive.
There are 28 Thin-clients connected to server through L300 N computing Thin-clients.
Thin-clients connect to V-space server installed in server for RDP users to get connected.
we have installed around 20 applications including printer and scanner driver. And apps are has below,
Firefox browser, windows mail, Adobe acrobat XI, canon printer and scanner drivers, Epson printer and scanner driver, E scan anti-virus, office 2007, v space, power ISO, win-rar,Tally and e token drivers and some backup software's.
Below are the services and features enabled,
AD, File services, RDP, web server, Hyper-v, .net frame work.
Is there a way to increase the performance .
Very slow performance.

Hi,
what would you suggest on hardware configuration must be for above mentioned applications and services with those many users.
how many cores and ram size is required.

Why is there only one Operating Concern defined in GBI?

Hi all,
I have the following question: when looking at the CO structure of the GBI case study, why is there only one, global operating concern defined? According to the definition of an OC, it is the level whereupon profitability analysis is performed. So why is in GBI that has two very distinct business areas not at least two OCs defined (one per business area), if not more (per product)?
As a follow-on, what is then the relation between OC and Controlling Area? I am looking for a plain English explanation that I can also use for teaching purposes.
Thank you in advance for reading!
Regards,
Gerhard

Hi Gerhard,
I hope my reply is not too late.
I agree with you. Profitability analysis is typically done at a detailed level - by market segments, which may be defined by geographical segments, customer groupings, product groupings, salesforce grouping, etc. CO-PA does support profitability analysis at such detailed levels.
After defining an operating concern and activating it, the system generates a series of database tables for the operating concern. I took a screenshot of the (partial) structure of the "master" table for the global GBI operating concern, GL00, from the GBI 2.2 system I am using, below:
As you can see, with such a table structure, CO-PA is capable of capturing profitability information at a very detailed level for reporting and analysis.
You may also notice that operating concern is not a field in the table. The reason is, the entire table (as well as its related tables in the same series) is used only for operating concern GL00. Its name is even hard-coded in the last 4 characters of the names of these tables. When CO-PA reports are executed in SAP ECC, they retrieve data from only one series of tables, i.e., only data of only one operating concern can be reported.
Back to your question: Why is there only one operating concern and not two? The wisdom of defining one global operating concern is that profitability data of both controlling areas (Europe and North America) will be updated into the same series of database tables and reported together. If there had been two separate operating concerns for the two controlling areas, there would have been two series of database tables, making profitability analysis across the two areas difficult, if not impossible.
Let me know if you need further clarifications.

Increase the Performance of a package while execution

Hi,
i am using oracle 9i and linux..
in my pl/sql if i executed a particular package it will take 5 min.
please explain how to increase the performance of the package
how to extract the query from the package and
how to build an explain plan..
and how to tune the query
Regards

If you wanna test pl/sql, sql profiler is a good way to test.
You can always use 10046 event to trace the queries in the package. Take their explain plan and see if there is any room for tune.
affar

Increase the performance of fileadapter

Hi,
can any body tell me how to increase the performance of the FILEADAPTER & DBADAPTER?
Regards,
Krishna.

As Anirudh has correctly stated you should upgrade if possible. In 10.1.3.1 the DB adapter had many issues, plus 10.1.3.1 was released 3 years ago so very old version.
I would recommend upgrading or starting on 11.1.1.2 as there have been many improvement in the file and db adapters to manage large payloads. In 10.1.3.x you have limitations of 1MB in size before you need to start tuning or doing some advanced configuration.
11g is the way forward, implemented using Oracle strategic application.
cheers
James

Why does this forum perform much slower than form forum in metalink?

I feel strongly that this forum performs much slower than the form forum in metalink where there are even more active and more issues created there.
I don't know why Oracle creates two form forum, which one is faster and another one is slower.
What is the difference b/w them besides here is jsp pages and over there is plsql pages?

Oracle certainly allows you to have users that do not have roles. Or users that don't have any system privileges. Or users that don't have any object privileges.
If you want the query to return a row for every row in DBA_USERS, you would need to outer join all the other tables to DBA_USERS.
Justin

How to increase the performance of an Application

Hi,
I am using jdeveloper11.1.1.2.0 and weblogic server10.3.2.I have lot of txn in the database.when click the txn menu then page landing is very slow.My Txn data is 24thousands records.
My RAM is 8GB and 64bit machine .So how to fast my application?Is any idea then Please help me.
Thanks
Anup

Hi Frank,
I have follow the below link but there is no performance effect of my application.
http://www.gebs.ro/blog/oracle/adf-view-object-performance-tuning-analysis/
On VO level ,In Tuning
I have modified In batch Of 26 ,As needed ,and Query Optimizer hints FIRST_ROWS_26
AND ON the page level ,Table Column Property,Fetch size i have modified 26 and page definition level ExchangeTradedPlacementHeadVOIterator default range size is 25.
So i have above change but there is no effect on the performance of my application.
Please give me the Solution what will i do to increase the performance of an application.
Thanks
Anup

Re: Why are there numbers not showing up on my verizon online call and texts log?

I am having the same problem with texts not appearing on my online text log and have also noticed a huge increase in data since my two sons have had their iPhone 5s. I have the iPhone 4. Two problems here
iPhone users DO appear. In the posted questions and response below, you indicate that if we are all using iPhones then we are not using the Verizon network but using data. HOW DO I STOP THIS DATA USAGE AND RETURN TO USING THE NETWORK? I can't afford the data charges that I am getting.. HELP!! and Thank you.
1.   Re: Why are there numbers not showing up on my verizon online call and texts log?
          SuzyQ   May 11, 2013 8:03 AM (in response to aheath17)
If you are all using iPhones, then it's likely the texts are going through iMessages and therefore use data. Since the texts are not using the Verizon network, but the app, they do not show up in the logs. The app is not specifically blocking the texted numbers from showing up, but the information comes through Verizon as data usage, not numbers texted.

    Sparkyjames81,
Lets figure out just where those texts are going! There are many apps that can be downloaded from the Play Store for Android phones that will allow messages to be sent over the internet, and therefore they are not actually text messages and will not show up in our records. Even Facebook has their own messaging app available that does this.
If you have an iPhone, they come with a service called iMessage that allows for messages to other Apple devices be sent via the internet, and this will also not show up on our records. These are all ways to bypass our actual texting services, and since these messages are not being sent as texts they do not show up on our records.
SarahO_VZW
Follow us on Twitter @VZWSupport

How to increase the performance of a domain index?

Oracle Database 10g Enterprise Edition Release 10.2.0.3.0 - 64bit Production
I am using a dynamic query which uses a domain index for a clob column. The table has 1021487 records. How can i increase the performance of a select query based on this table? This is the only part where my application gets slowed down.
The query is
SELECT :search_id, 100, :item_type, c.search_college, c.search_colname, TRUNC(dbms_random.value(900000000000, 999999999999)) rand, COUNT(1), c.college_rating, min(DECODE(c.search_postcode, NULL, 99999, NULL)) search_distance FROM w_search_text c WHERE c.search_course_type is not null
and (contains(c.search_text,:search_text,2) > 0)
AND c.search_item_type = :item_bind
GROUP BY c.search_college, c.search_colname, c.college_rating
ORDER BY c.search_college
here c.search_text is the domain indexed column?
Is there a way to tune it?
Ram

Here c.search_text is the clob column.
i am using 4 bind variables and they are search id : 123456
item_type : O
search_text : French
item_bind : Z
I tried explain plan and here are the results
Plan
SELECT STATEMENT ALL_ROWSCost: 5 Bytes: 38,280 Cardinality: 264
     3 SORT GROUP BY Cost: 5 Bytes: 38,280 Cardinality: 264
          2 TABLE ACCESS BY INDEX ROWID TABLE HOT_ADMIN.W_SEARCH_TEXT_LIVE2 Cost: 4 Bytes: 38,280 Cardinality: 264
               1 DOMAIN INDEX INDEX (DOMAIN) HOT_ADMIN.W_SRCH_TXT_IDX2 Cost: 4
Its using the domain index, But the column is highly fed up with data...so ithink that may be the reason.Is there any HINTS in particular to tune domain indexes?
Ram

Would upgrading my MBP hardware increase FCPX performance?

Hi -
I have a mid-2012 13 inch MacBook Pro (non retina) and I plan on getting into video editing for school next year. My current system specs are:
2.9 GHz Dual-core Intel Core i7 (i'm pretty sure it turbo boosts up to 3.6 GHz)
8 GB 1600 MHz DDR3
Intel HD 4000 1024 MB
750 GB 5200 RPM hard drive
Running 10.9.2 Mavericks
The computer runs great for things I use it for now such as Microsoft Office 2011 and media consumption, but I'll be doing more video editing next year. I currently own Motion 5 on it and it runs fine at the beginning, but when I add more layers it really starts to slow down and the computer gets very hot. I also notice that there is only like 25 MB RAM free (not sure if thats a mavericks thing or if motion is using all 8 gigs.) I will most likely be getting Final Cut Pro X next year, so I was wondering what upgrades I should make to this computer to increase the performance.
As far as I know, you can only upgrade the RAM and Hard Drive in this model Mac, so I would be planning to upgrade to 16 GB ram and an SSD (not sure what size yet.) I know FCPX is a real RAM hog, so I figured throwing 16 gigs at it should increase its performance. However, I'm not so sure how replacing the HD with an SSD would increase performance in FCPX and Motion, besides how long it takes programs to open and the computer to boot.
So my question is: would it be worth it to make these upgrades and how much would it improve my FCPX and Motion performance?
If anyone could post some links to where I could find these that would be great too.
-Thanks
Message was edited by: Mr A Man2

In Mavericks always ram is "used" completely, unless you have way to much ram.
What you look for is the "memory pressure" to see the ram load.
8GB is more than enough normally, but FCPX is a real ram cannibal. So may be it makes sense to upgrade to 16GB. But know that the macs are extremely sensitive for ram being the exact specifications and good qualtiy. In my experience the best is Crucial or OWC, use the configurators on their sites.
Lex

Why is there increased overhead(performance) in NAV than DISPLAY attribute?

Similar Messages

Maybe you are looking for