OLAP Cache

how does OLAP Cache work?

Logical File Names
Use
To save data in a file in the cache modes Main Memory Cache with Swapping, Flat File Cache per Application Server, or Cross-Application Server (see Cache Mode), the following specifications must first be set:
●      Definition of the logical file path and assignment to a platform-specific physical path
●      Definition of the logical file name
From the Cache Monitor screen, using  Logical File Names, you can access the complex table maintenance for logical file paths and names. The table maintenance consists of several subdialogs (views). SAP has delivered a proportion of the required settings.
You need to check these settings and make any necessary adjustments for your file system before you can use the persistent OLAP cache with a flat file as a background store.
Note that these assignments are subject to settings made for the transport connection (You need a request that can be transported).
Features
The following section gives an overview of the subdialog in the table maintenance using an example configuration. For more information about the procedure, see Defining Logical Path and File Names.
You can choose the following subdialogs using the dialog structure:
Logical File Path Definition
The (platform-independent) logical file path is evaluated at runtime to generate the (platform-dependent) physical path (see Assignment of Physical Paths to Logical Path).
Two examples of this are:
●        Logical file path: BW_OLAP_CACHE
Name: BI OLAP Cache
●        Logical file path: BW_OLAP_CACHE_SPAN
Name: BI OLAP Cache Cross-Application Server
Assignment of Physical Paths to Logical Path
At runtime, the physical path assigned to the corresponding operating system is generated from the logical file path. 
Since the physical path is platform-dependent, the customer has to specify it. The closer the file is to the application server or, with the cross-application server cache mode, to the application server, the faster it can be found and read. 
Log(ical) path: BW_OLAP_CACHE
Name: BI OLAP Cache
Syntax group: UNIX (all Unix platforms)
Phys(ical) path: /usr//global/ (You can choose the path up to <FILENAME> according to your own configuration.)
If you cannot find the relevant syntax group for file names and paths for your operating system, you can define your own (see Syntax Group Definition and Assignment of Operating System to Syntax Group).
The <FILENAME> parameter must always be set for the physical file name (see Logical File Name Definition, Cross-Client). At runtime, the complete platform-specific file name is automatically formed from the physical path and the physical file name.
Logical File Name Definition, Cross-Client
At runtime, a complete platform-specific file name including the path is generated for a (platform-independent) logical file name.
In the following example, cache data is stored on an application server:
Logical file: BW_OLAP_CACHE (we recommend you use the delivered attributes for the logical file).
Name: BI OLAP Cache Files 
Phys(ical) file: CACHE___
Data format: BIN (binary)
Application area: BW (Business Information Warehouse)
Logical path: BW_OLAP_CACHE (see Logical File Path Definition). Input help is available.
In the following example, cache data is stored cross-application server:
Log(ical) file: BW_OLAP_CACHE_SPAN
Name: BI OLAP Cache Cross-Application Server
Phys(ical) file: CACHE__
Data format: BIN
Application area: BW
Log(ical) path: BW_OLAP_CACHE_SPAN
Note that a particular schema is prescribed for the physical file name:
&#9679;      For application server storage (see Global Cache Parameters and Cache Mode Main Memory Cache with Swapping and Cluster / Flat File Cache per Application Server):
CACHE___
&#9679;      For cross application server storage (see Global Cache Parameters and Cache Mode Cluster / Flat file Cache Cross-application Server):
CACHE__
The following table offers an overview of the variables that are only replaced with actual values during runtime:
Variables in Physical File Names
Variable
Meaning
<HOST>
Computer/server name (for SY-HOST). 
If the data is stored cross-application server, this variable is not required. 
<SYSID>
System ID
For example, if two different BI systems are running on one application server, this variable serves to differentiate the systems so that they do not overwrite each other.
<PARAM_1>
External parameter 1 (sequential number in HEX)
Definition of Variables
If you want to use variables in physical file names and paths, you can define these by choosing New Entries.
Syntax Group Definition
You can define a new syntax group for operating systems with the same syntax for file names and paths by choosing New Entries.
Syntax grp: UNIX
Name: All Unix platforms
Length (of file name)
Extension: Choose this option if the data format for the logical data name is allowed as the file name extension for this syntax group.
Active: Choose this option if the operative system for this syntax group is used in the available system configuration.
Assignment of Operating System to Syntax Group
You can assign your operating system to a syntax group by choosing New Entries if necessary.
OP system: SunOS
Name: SUN Unix
Regards,
SUBHA

Similar Messages

  • Pre-fill the OLAP cache for a query on Data change event  of infoprovider

    Hi Gurus,
    I have to pre-fill the OLAP cache for a query,which has bad performance.
    I read a doc 'Periodic Jobs and Tasks in SAP BW'
    which suggested sum steps to do this
    i hav created the setting for Bex broadcasting for scheduling job Execution with data change in info provider
    thereafter doc says  "an event has to be raised in the process chain which loads the data to this InfoProvider.When the process chain executes the process u201CTrigger Event Data Change (for Broadcaster)u201D, an event is raised to inform the Broadcaster that the query can be filled in the OLAP cache."
    how can this b done please provide with sum proper steps
    Answers are always appreciated.
    Thanks.

    Hi
    U need to create a process chain or use the existing process chain which you are using to load your current solution, just add event change process type in the process chian  and inside it add the info provider which are going to be affected.
    Once you are done with this go to the broadcaster  and  create new setting for that query...you will see the option for event data chainge in infoprovider just choose that  and create the settings.
    hope it helps

  • Impact of real time cube on query performance and OLAP cache

    Hi:
    We have actual and plan cubes both setup as real time cubes (only plan cube is being planned against, not actual cube) and both cubes are compressed once a day.
    We are planning on implementing BIA accelerator and have questions related to query performance optimization:
    1/ Is there are any query performance benefits in changing the actual cube to be a basic cube (using program SAP_CONVERT_NORMAL_TRANS) if the F table is fully compressed
    2/ Can OLAP cache be leveraged for the queries run against the real time cubes e.g. actual cubes
    3/ What is the impact on BIA of having the actual cube as real time (whetehr or not there is data being loaded/planned during the day in that cube)
    Thank you in advance,
    Catherine

    1) Is there are any query performance benefits in changing the actual cube to be a basic cube (using program SAP_CONVERT_NORMAL_TRANS) if the F table is fully compressed
    From the performance point of view, tha actual cubes i.e std cubes are relatively better.
    2) Yes OLAP cache can be leveraged for bringing up the plan query but all the calculations are done in the planning buffer.
    3) Not sure.

  • Query views are not using OLAP cache

    Hi,
    I am trying to pre-fill the OLAP cache with data from a query so as to improve the performance of query views. 
    I have read several documents on the topic, such as “How to… Performance Tuning with the OLAP Cache” (http://www.sapadvisors.com/resources/Howto...PerformanceTuningwiththeOLAPCache$28pdf$29.pdf)
    As far as I can see, I have followed the instructions and guidelines in detail on how to set up the cache and pre-fill it with data. However, when I run the query views they never use the cache. For example, point 3.4 in the abovementioned document does not correspond with my results.
    I would like some input on what I am doing wrong:
    1. In RSRT I have Cache mode = 1 for the specific query.
    2. The query has no variables, but the following restrictions (in filter): 0CALMONTH = 09.2007, 10.2007, 11.2008 and 12.2007.
    3. I have one query view with the restriction 0CALMONTH = 10.2007, 11.2008 and 12.2007.
    4. I have a second query view, which builds on the same query as the first query view. This second query view has the restriction 0CALMONTH = 11.2008 and 12.2007.
    5. There are no variables in the query.
    6. I run the query. 
    7. I run the first query view, and the second query view immediately after.
    8. I check ST03 and RSRT and see that cache has not been used for either of the query views.
    Looking at point 3.4 in the abovementioned document, I argue that the three criteria have been fulfilled:
    1. Same query ID
    2. The first query view is a superset of the second query view
    3. 0CALMONTH is a part of the drill-down of the first query view.
    Can someone tell me what is wrong with my set-up?
    Kind regards,
    Thor

    You need to use following process of process chain: "Attribute change run (ATTRIBCHAN)". This process needs to be incorporated into your process chains which loads data into provider on top of which your query is based.
    See following links on topic how to build it:
    https://help.sap.com/saphelp_nw73/helpdata/en/4a/5da82c7df51cece10000000a42189b/frameset.htm
    https://help.sap.com/saphelp_nw70ehp1/helpdata/en/9a/33853bbc188f2be10000000a114084/content.htm
    cheers
    m./

  • Olap cache not storing result sets for certain queries

    Hi - another cache question
    my result set for a query is not getting set in OLAP CACHE - the query variable level is stored ok but not the result set.
    different queries store the result sets ok.
    there is nothing in the query or cube properties that should disable the caching of the result set.
    there is plenty of space available and there has not been any changes to the query or data loaded or aggregates changed.
    any ideas?
    cheers - Neil

    I would do a performance check in RSRT2.  This will usually give you some additional information related to olap cache. 
    Hope that helps.

  • Automatic filling OLAP Cache

    Hi @ll,
    i am using bw 3.5....
    I`ve a question as to OLAP Cache. Is there any possibility/method to fill the Cache automatically every day?
    Id like to use the performance enhancement of caching queries....but i would like to avoid it that the first user who execute a query have a long waiting time....
    Exists there any options to handle this?
    thx in advance for all answers and helpful advices
    Cheers,
    Marian

    HEllo,
    Check the foll link
    https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/9f4a452b-0301-0010-8ca6-ef25a095834a
    Regds,
    Shashank

  • How to empty out Olap Cache for a query?

    Hi experts,
    I think the olap cache for a query should be resetted to zero after a data load (right?), if i have understood well how it works. Where can i check the olap cache for the query?  I'd like to ckeck the change of the Olap Cache for a specific query.
    I checkt in the Cache Monitor but I haven't found any data.
    Francesco

    Hi
    You can use the Transaction code RSRCACHE
    Regards
    N Ganesh

  • Fill OLAP cache via broadcaster

    Hi,
    I'm trying to use the broadcaster with distribution type 'Fill OLAP cache' and schedule it using " Including Event Data Changes in Process Chains ". The process chain has been adapted with the 'Event data change' process and scheduled. Process chain runs fine.  All was set up according to the link below.
    http://help.sap.com/saphelp_nw70/helpdata/EN/a5/359840dfa5a160e10000000a1550b0/frameset.htm
    However, when I try to schedule the broadcaster setting, I only get the options of "Execution at Predefined Time"  or direct scheduling.
    Should I not be able to see "Execution with Data Change in the InfoProvider " as an option? Is there some other setting to be made?
    Very grateful for any ideas.
    Thanx

    Check if you have S_RS_BCS role authorization.
    You have to use process type DATACHANGE in the process chain for data load.
    Check this http://help.sap.com/saphelp_nw04/Helpdata/EN/ec/0d0e405c538f5ce10000000a155106/content.htm

  • Fill Olap cache distribution type

    Hi,
    I'm trying to fill the olap cache with information broadcasting, but I cannot see option "Fill olap cache" in the distribution type.
    Only distribution type "broadcast E-Mail" and "Broadcast to portal" are present.
    I reached the broadcasting screen from Tcode RSRT -> ABAP Web.
    Is a Precalc. server required to get access to "fill olap cache" distribution type ?
    How can I get access to distribution type "fill olap cache"  ?
    System: BW 7.01 SP06

    Hi Brian,
    Please check below:
    Broadcasting in BI 7.x - Do I need Portal, Missing Distribution Types
    Also check the expert for
    In SAP NetWeaver BI 7.0 (2004s), which BEx Web functions require SAP NetWeaver 7.0 Portal?
    As soon as you use any of the new BI capabilities for SAP NetWeaver 7.0 (2004s) BEx tools (such as BEx Analyzer & Broadcasting), you require usage types BI Java and Portal. The only exception is if you do not use SAP NetWeaver 7.0 BEx tools and ONLY use the old 3.x BEx tools.
    https://www.sdn.sap.com/irj/sdn/nw-datamanagement?rid=/library/uuid/a5067965-0901-0010-6f8a-bbf0b7424283#q-6
    -Vikram

  • Warming up the OLAP cache

    Hi,
    I would like to schedule execution of some queries in order to put the results in the OLAP cache for fast use.
    Each user run the query with very restricted selection: one period, one node in the CostCenter hierarchy and one currency type! All the selections are obligatory and for single values.
    I created a "super" query with the same characteristics, all in the rows or columns, with not mandatory variables, ...
    I scheduled the "super" query and it create an entry in the OLAP cache.
    When I now run the production query (other query on the same cube, for one node, one period, one currency) the system doesn't use the OLAP cache, but create a new line in the OLAP cache.
    It is impossible to schedule every combination.
    Is there a way to worm up the cache with a different Q?
    Should the seedling be done with Broadcaster or Reporting Agent (in NW2004s)? I think this should be the same ?!
    Any suggestion to fill the cache for my situation?
    Thanks, Tom

    The OLAP cache is at a query level, so to warm up the cache, you must run the actual query that the users will run, not just a similar or "super" query that is similar.
    SDN has some doc on effectively using Global cache which would be good to review.
    https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/9f4a452b-0301-0010-8ca6-ef25a095834a
    The other key thing to understand is how the setting on variable "Can be changed during query navigation" works.  If this setting is NOT set on your variables, then in order for a query to access the data in the cache, the query must have been run previously with the exact same variable values, as the varaiable values get saved as part of the cached info.
    Selecting the "Can be changed during navigation" variable setting changes the way the system considers the variable - it behaves as a filter.  This has some implications with way prompts for variables appear to the user, so there are user impacts.  In BEx, a variable is normally presented for input when you first run the query and then whenver you refresh it, but when you change the variable setting to Can be changed..., the variable prompt will be presented teh frist time the query is run, but NOT when a refresh is done, it now behaces as if you added a filter to the query.
    So if you change all the variables (or create new ones) used bythe query to Can be changed..., then run your query thru reporting agent or info broadcasting wide open, or with restrictions that encompass all your user's query executions, the subsequent user executions will use the global Olap cache.

  • Can a query view be broadcast to olap cache??

    Hi, 
    I am trying to pre-calculate a query view to the olap cache.  I do not see this option in drop down.  Is it possible
    Thanks

    Hi,
    Well the few setups of ASA + Ironport that I have seen uses an WCCP ACL on the ASA which already can define which traffic is forwarded to Ironport and which simply bypasses it and goes straight out of the ASA.
    Those that need to bypass Ironport are specified in the top of the ACL with "deny" statements.
    access-list WCCP line 1 deny ip host host
    or
    access-list WCCP line 1 deny ip host any
    I am not sure if this was what you were looking to do.
    - Jouni

  • OLAP Caching question

    Hi
    I'm not sure if this is the right forum, but I figured I would ask anyways. Using the broadcaster, I have configured a couple variable settings on a query to fill the OLAP Cache.
    Is there a way to update the parameter settings by an api instead of going in manually to each query and updating the parameter values?
    For example, say I want to cache only two months worth of data. I set the query settings for Jan and Feb. Now I want to configure to Mar and Apr. Instead of manually opening up the settings and entering and saving I want to do this using code if possible.
    The reason why is if I have 100 queries to set this, it would be much easier to run this through code than manually updating the values.
    Thanks in advance,
    Daniel

    check the table RSRPARAMETRIZA, field-content which stores the variant values. You can update the variant values dynamically using ABAP.
    This articles shows you how to do that.
    http://www.sdn.sap.com/irj/scn/index?rid=/library/uuid/f048c590-31a4-2c10-8599-bd01fabb93d4

  • Query mass Broadcasting to fill OLAP cache

    Hi,
    I have +- 50 relative small queries that I'd like to precalculate each night (push them to the OLAP cache) as these queries are displayed together in one KPI dashboard.
    I know the Bex Broadcaster is the tool that should be used but at this moment I don't know any way to schedule the precalculation for ALL queries in mass. Is there a way to Broadcast queries in mass via a certain program or via any other means?
    Putting them in one workbook is an option (and having this workbook precalculated) but I was wondering if there are any other solutions ?

    I have now been investigating mass broadcasting via :
    1 one workbook with several queries integrated
    1 webtemplate with several queries integrated
    But both alternatives don't offer the option "Fill OLAP Cache" in the Bex Broadcaster. It seems this is only selectable when you start the Broadcaster for one individual query ?  So my conclusion is that you need to make a settings for all queries individually ? Or is there a way to change the distribution method for webtemplates ? Quite a big dissapointment for me anyway...

  • InfoProvider OLAP Cache settings

    The BI nw2004s 7.0 version is support stack 12 SAP_BW level 14.  How do you set the olap cache mode to flat file at an InfoProvider level.  Via rsrt, each query can be set, but the defaults are currently 'Cache without swapping'.  Want to change that to 'persistence flat file', but not have to manually do it for every query.  For BW 3.5, the process was: Use Transaction SPRO and
    menu path SAP Customizing Implementation Guide • SAP NetWeaver •
    SAP Business Information Warehouse • Reporting-relevant Settings •
    General Reporting Settings in Business Explorer • InfoProvider Properties
    However, this path does not exist in 7.0
    Any feedback on how to change the cache mode for an infoprovider so that it doesn't have to be manually done for each query.

    Hi Dan,
    Please go to the infocube itself on RSA1 in display or change mode.
    Through the menu - enviroment>infoprovider properties>change.
    On the first tab u can change the settings of the cache.
    Tomer.

  • Do Query Views use existing OLAP Cache or create their own

    Hi, I'm looking to find out if a Query View will use the Parent Query OLAP Cache or will it create it's own? 
    I am imagining that it would use the query view, but create additional cache if it drilled down on a free characteristic that wasn't in the original query output.  Any comments

    If there are any entries in the CACHE which meets the drilldown and filter criteria for the view it will be able to make use of existing Cache.
    As you mentioned for any additional drilldowns if there are not entries in the cache it will have to read from DB.

Maybe you are looking for

  • Not receiving AOL emails

    A friend and I just discovered that I am only receiving some of the emails he is sending me. He has an AOL account. After checking, I found that the only emails I am receiving are being sent from his Verizon phone account. The ones he is sending dire

  • 1099 reporting issue

    We created a KR document in 2006.  The check was cut in 2006.  A 1099 was generated and sent to the recipient. But, in February 2007, the check was voided and other check was cut (dated February 2007). Now, the numbers are off between invoices and pa

  • More problems on installing 9i on RedHat 7.0

    I have problems installing 9i at RH70 also. But in my case I can't even start Oracle Installer, only generate a segmentation fault. (I can't reproduce this behaviour installing 9iAS release 1, by the way) Two doubts: - Can somebody confirm whether in

  • 2960 - HULC LED PROCESS

    Dear All, I have cisco WS-C2960S-48FPS-L stacked. Weekly twice, my PoE connections are dropped and when the device is restarted, everything starts working normal. This issue happens weekly once or twice. Verified the power input and it's absolutely f

  • What the Point of SharedInputStream? I can share any InputStream!

    I made an InputStream called: MIMEInputStream which is a subclass of FilterInputStream. It allows me to read by line, simply by saving what it reads in a local buffer and looking for every occurrence of \n or \r\n. then returning what it found. I did