MAXDB : Data Distribution

Hi,
What is the counterpart for Histograms (in Oracle) in MaxDB? i.e How can we get the data distribution in MAXDB?

> I would like to know whether kernel trace provides the information about this optimization and sampling?? If not, how can this optimizer trace be accessed or found?
Yes, activating the kernel trace for the optimizer would create further informational output.
Anyhow, this output is intended for the developer of the optimizer and difficult to understand.
It's nothing like the CBO trace of oracle.
Really - if you want to understand better what the optimizer "thought" about your data, then use EXPLAIN JOIN and EXPLAIN SEQUENCE. These commands produce output that deliver some useful information.
Apart from that it should be clear, that the inner workings of the optimizer are not documented and changed all the time to improve performance and stability. In 7.6 alone there had been so many changes and enhancements.
As the general design goal of MaxDB is ease of use it shouldn't be necessary to know what the optimizer exactly does.
If you find your statement is badly optimized - post it here and we have a look.
Maybe it's a bug that should be fixed.
regards,
Lars

Similar Messages

  • ALE:Material Master Data Distribution

    Hi
    I am new to ALE Material Master Data Distribution and if the questions are incorrectly worded, please accept my apologies. It is due to lack of knowledge.
    I am doing Material Master Data distribution and I am using standard basic type MATMAS03. The standard TCode MM01, MM02 has SAP tables and 3 custom tables.
    I have turned the active on using BD61 and change pointers to message type.
    Q: Is the change pointer, applicable to entire material master? i.e. if I change a field in Z table or standard material tables, can an IDOC be created for the entire material record?
    Q: What is the function of BD52? Does this override the option Activate Change pointers for message type? Does it mean, if the changes are made to those fields in BD52 a change pointer is created?
    Q: If I change Z table only, will there be a change pointer created? If not, how could I accomplish this using BD52.
    Q When do we use the enhancement MGV00001?
    Also, If you could add anything in addition to above queries, it will be helpful to me and I appreciate it.
    With regards,
    William

    Thank you Sudhakar.
    Your tips are good, but, I was not able to get the problem resolved.
    Let me explain what I have done based on your tips.
    1.  MATMAS uses MATMAS03. The basic type has NOT been modified to have z fields yet.
    2.  Using BD52, i have created the following fields
         MATERIAL    ZEMM_MARA_TECH   IM_ADVISORY_CODE
         MATERIAL    ZEMM_MARA_TECH   KEY.
    3.  Modified the MMR for IM_ADVISORY_CODE.
    I did not see a change pointer for this change.
    Q:  When I created above entry using BD52, was asking the table name. The table name is stored in TCD0B. What TCode is available the entries in TCDOB. For the time being, i created using SQL.
    Q. Is there a follow up configuration required to capture the changes?
    I will award the points in a couple of days time, even if I do not get the solution, because you have spent your precious time for the community. I will wait for a couple of days for other suggestions and close the Problem.

  • Master Data Distribution !

    Hi!
       I want to know the purpose of master data distribution for the following between the vendor & the customer.
       1. Material Master
       2. Vendor Master & Customer Master.
      Whats the purpose of linking our system with our vendor or customer etc with <b>regard to master data</b>
      Pls explain in detail.
      Thanks
      Rahul.

    Hi Rahul,
    We dont do master data distribution with customer system or vendor system.
    Master data distribution is done between distributed systems of the same organization using ALE configuration. So we dont link to customer or vendor systems for transfering master data but for transfering transactional data like purchase orders or sales orders etc.
    Master Data Distribution
    Rather than distributing the complete master data information, views of the master data can be distributed (for example, material sales data, material purchase data). Each view of the master data is stored in a separate message type.
    Users can specify which data elements in a master record are to be distributed.
    Various distribution strategies are supported:
    ·        Cross-system master data can be maintained centrally and then distributed. The final values are assigned locally.
    ·        A view of the master data can be maintained locally. In this case there is always one maintenance system for each view. After the master data has been maintained it is transferred to a central SAP system and distributed from there.
    Types of Distribution
    ·        Active distribution (PUSH)
    If the master data is changed (for example, new data, changes or deletions), a master data IDoc is created in the original system and is distributed by class as specified in the distribution model.
    ·        Requests (PULL)
    A request occurs when a client system needs information about master data held in the system. You can select specific information about the master data, for example, the plant data for a material.
    If you want to be notified of all subsequent changes to the master data, this has to be set up “manually” between the two systems. It is not yet possible for this to be done automatically in the distribution mechanism in the original system.
    Transferring the Master Data
    A distinction is made between transferring the entire master data and transferring only changes to the master data.
    If the entire master data is transferred, a master IDoc is created for the object to be distributed in response to a direct request from a system. Only the data that was actually requested is read and then sent. The customer specifies the size of the object to be distributed in a request report.
    If only changes are transferred, the master IDoc is created based on the logged changes.
    Reward points for the useful answers,
    Aleem.

  • Error during DG data distribution

    Hi SAP DG Experts,
    I am configuring and testing DG data distribution from central EHS system to multiple ERP systems.
    I am able to send the DG idoc and ERP system is receiving that.
    But the DG data creation is failing and when I check DGp7 I can get an error message "data records were not save as no selection date was available".
    Please let me know if you have come across this issue?
    Thanks
    PS

    Dear Pugazendhi
    to distrbute sucessfully DG master via ALE you need to disztrbute
    a.) the phrases
    b.) do a proper set up of receiver system
    Check e.g.
    Distribution (ALE) of Exceptions to Dangerous Goods Regulations - Dangerous Goods Management (EHS-DGP) - SAP Library
    This topic is asked for very rarely
    C.B:

  • Data distribution in distributed caching scheme

    When using the distributed ( partitioned ) scheme in coherence , how the data distribution is happening among the nodes in data-grid..? Is there any API to control it or are there some configurations to control it ?

    Hi 832093
    A distributed scheme works by allocating the data to partitions (by default there are 257 of these, but you can configure more for large clusters). The partitions are then allocated as evenly as possible to the nodes of the cluster, so each node owns a number of partitions. Partitions belong to a cache service so you might have a cache service that is responsible for a number of caches and a particular node will own the same partitions for all those caches. If you have a backup count > 0 then a backup of each partition is allocated to another node (on another machine if you have more than one). When you put a value into the cache Coherence will basically perform a hash function on your key which will allocate the key to a partition and therfore to the node that owns that partition. In effect a distibuted cache works like a Java HashMap which hashes keys and allocates them to buckets.
    You can have some control over which partition a key goes to if you use key association to co-locate entries into the same partition. You would normally do this to put related values into the same location to make processing them on the server side more efficient in use-cases where you might need to alter or query a number of related items. For example in finanial systems you might have a cache for Trades and a cache for TradeValuations in the same cache service. You can then use key association to allocate all the Valuations for a Trade to the same partition as the parent Trade. So if a Trade was mapped to partition 190 in the Trade cache then all of the Valuations for that Trade would map to partition 190 in the TradeValuations cache and hence be on the same node (in the same JVM process).
    You do not really want to have control over which nodes partitions are allocated to as this could alter Coherence ability to evenly distribute partitions and allocate backups properly.
    JK

  • WBS data distribution for IDOC-PI-JDBC with system status 'CRTD' not working.

    Hello Experts,
    I am busy with a project where the PS project creation and updates should be transfered to the external legacy system through IDOC-PI-JDBC scenario.
    I am using the BAPI_PROJECT_MAINTAIN for this process. This scenario is working fine with the projects system status released (REL).
    If the system status is Created (CRTD) then no idoc creation happening therefor no data distribution of work breakdown structure(WBS) to the legacy system.
    Can anybody give me some suggestion of make it happening.
    Thanks in advance.
    Regards,
    Antony.

    Hello all,
    I did it with the help of my abap developer.
    We implemented a BADI, and from the BADI we created the idoc.
    When the idoc is there the legacy system receive the data.
    Regards,
    Antony.

  • Data distribution scheme and database fragmentation

    Hi all,
    I'm working on a scenario (University) involving the fragmentation of a central database. A company has regional offices i.e. (England, Wales, Scotland) and each regional office has differing combinations of business areas. They currently have one central database in their head office and my task is to "design a data distribution scheme". By scheme does this mean something like horizontal / vertical fragmentation? Also can somebody point me to an Oracle specific example of creating a fragmented table? I've tried to search online and have found the "partition by" keyword but not much else except for database linking - but I'm thinking this is more concerned with querying than actually creating the fragments.
    Many thanks for your time

    >
    Partitioning is what the tutor meant by "fragmentation". So if there is a current central database and I have created new databases for each regional office I could run something like the below statement on the regional databases to create a bespoke version of the employee table filtered by data relevant to them? This is all theoretical and we don't have to develop the database, I just want to get the syntax correct - Thanks!
    >
    There you go talking about 'new databases' again. You said your original task was this
    >
    my task is to "design a data distribution scheme".
    >
    Is the task to give the regions access to their own data in the ONE central DB? Or to actually create a new DB for each region that contains ONLY that regions data?
    So are we talking ACCESS to a central DB by region? Or are we talking replication of the entire central DB to multiple regions?
    Your example table is partitioned by region. But if each region has their own DB why would you put data for other regions in it?
    If you are wanting each region to have access to their own data in the central DB then you could partition the central DB tables like your example:
    CREATE TABLE employees (
    id INT NOT NULL,
    fname VARCHAR(30),
    lname VARCHAR(30),
    hired DATE NOT NULL DEFAULT '1970-01-01',
    separated DATE NOT NULL DEFAULT '9999-12-31',
    job_code INT,
    store_id INT
    PARTITION BY LIST(region_id) (
    PARTITION Wales VALUES IN (2)
    ); But if you are creating a regional DB that includes data only for that region there is no need to partition it.

  • Full Master Data Distribution in ALE - Vendor Master CREMAS(Full Distributi

    Hi Experts,
       I have to do a Full Master Data Distribution in ALE - Vendor Master CREMAS(Full Distribution) using BD21.
       I have already read the blog of Mickel and incorporated the code in Change_Pointer_Read.
       It is still not working. Should I include the code at the end of the Function Module. If yes, how to code at the last as
       I can include my code at the SAP suggest Enhancement and End Enhancement.
       Is there any way I can create the implicit Enhancement at the end of the Function Module.
       I will appreciate any help to make the Full Distribution work.
    Thanks,
    Mich

    Hi,
       Here is the link.
       /people/michal.krawczyk2/blog/2009/06/04/distribution-of-full-master-data-objects-from-change-pointers
       My requirement is to send the whole Vendor Master in a IDoc even though for eg there is only one field change in address field of Vendor Master. Full Master Data distribution. But the program BD21 creates only IDocs with segments where the changes are made in the particular tab like address.
       Is there any Enhancement or way to enforce the whole Master Data(CREMAS) to be distributed.
       I will really appreciate any helpful answer .
    Thanks,
    Mich

  • MDG-F data distribution in case of add on deployment

    Hi Expert
    Can any one describes the procedure for MDG-F data distribution for add-on deployment scenario?
    Is it required to release edition each time? Then what is meaning of check box- Immediately Distribute Change Requests:in MDG 6.1  if you have to replicate edition manually each time? Is data replication model is required for data distribution as well?

    Hello Sanjay
    MDG-F is a Flex model. You have to configure replication model in any case.
    There are 2 ways to replicate data from MDG.
    1. Auto Replication - Here you have to mark  immediately distribute change request while creating edition. With this tick mark, data will get replicated automatically. No need to release any edition.
    2. Manual replication - When you want to replicate the data manually, do not tick the auto distribution option. The data will get stored in the edition. You can release the edition any time. Only thing you need to make sure that there is no open change request in the edition. Once you release the edition, you can view the edition in manual replication option.
    Kiran

  • I would like to know what data represents the yellow "Other" in the Data Distribution Window on the Summary page of the iPad. It varies between a low of 1.50 GB to as much 5 GB. It seems to me that this is wasted space. Any ideas?

    I would like to know what data represents the yellow "Other" in the Data Distribution Window on the Summary page of the iPad. It varies between a low of 1.50 GB to as much 5 GB. It seems to me that this is wasted space. Any ideas?

    I have observed my "Other" data increases in size after I deleted a video. My Other data once get to 18GB, yes eighteen.
    It seems like the only way to control this mysterious "Other" data is restoring the iPad to its manufacture configuration. Restore it to a previous back up doesn't help either.  Once in a while is ok, but several times a month is a pain.
    After the last restore (to manufacture), my other immediately jump to 0.4GB and seems to hold at 0.65GB.
    Looks more and more like a bug in iOs.

  • Problem with data distribution from HCM to e-Recruiting

    Hello masters:
    I have a problem when I try to distribute data from HCM to e-Recruiting. I´m getting code 51 just for the PA infotypes segments (E1P0000, E1P0001, etc.). The error message tha I am receiving is:
    "Error in subroutine read_namtb for structure of infotype".
    My e-Recruiting system is SAP EHP 1 for SAP NetWeaver 7.0, so I'm using Business Partners for storing data from PA infotypes.
    I'm using message type HRMD_ABA, basic type HRMD_ABA05.
    I think it could be the implementation HRRCF00_INBD_NEWMOD from the badi HRALE00INBOUND_IDOC, because it usually get inactive even though I go to activate it.
    Does anybody knows what can I do?
    Thanks in advance!
    Edited by: Rodrigo Arenas Arriola on Jun 25, 2009 2:45 AM

    Hello Rodrigo,
    Ifyou are using EhP1 for NW 7.0 you should have E-Recruiting 600 EhP 4 on your system.
    I never encountered this special error message so far, so i have to "guess". The message indicates that the system had trouble reading some infotype information.
    The ALE distribution for EhP4 is storing the PA data in new infotypes 558*. The error you get points to a missing BAdI activation. If the BAdIs are not activated correctly, ALE will try to write the information from the PA infotypes 1 to 1 into the E-Recruiting system. But E-Recruiting does not know the PA infotypes which should exactly lead to the error you get.
    Could you check the following settings:
    BAdI                     Implementation 
    HRSYNC_P                 CONV_HR_DATA_TO_EREC -> active
    HRALE00INBOUND_IDOC      HRRCF00_INBD_NEWMOD -> active
    HRALE00SPLIT_INBOUND     HR_INB_PROCESS_IDOC -> inactive
    HRALE00INBOUND_IDOC      HRRCF00_DELETE_SPREL -> inactive
    Kind Regards
    Roman

  • Use of SQL Profiles where each schema has very different data distribution and volumes

    Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production
    PL/SQL Release 11.2.0.3.0 - Production
    CORE    11.2.0.3.0      Production
    TNS for Linux: Version 11.2.0.3.0 - Production
    NLSRTL Version 11.2.0.3.0 - Production
    Our architecture has multiple client schemas in the same database. Each schema has the same code base (tables, procedures etc) but each client's data is very different in terms of volumes and skew/distribution per table. This architecture was done based on cost - I know it's not ideal but it can't change.....
    I am fairly seasoned with performance management and so know the usual tricks of when to eat up the table using parallel full table scans etc. I couldn't further optimise a given stmt for our largest table. I'll call it TSPCI and it has monthly partitions (2 years) and totals about 35Gb in the largest client schema.
    Anyway, I was surprised when ADDM suggested that I could achieve 98% improvement if I were to use a given SQL Profile. Great?
    So, here's my issue - I've found that the same SQL_ID is shared across all those different client schemas: I can't see how to get it to pick/use the SQL Profile in only a particular client schema - let's call it NEX - and not in another (lets call it COL).
    If I generate a SQL Profile as NEX, has it analysed and built the SQL Profile based on the NEX schema and is it therefore invalid/undesirable to have that SQL Profile used in the COL schema??
    I suppose that I could add a small change (say /*+ NEX */) to the SQL in the NEX schema to make the given sql unique there and then generate a SQL Profile for that..........
    What am I missing here?

    Well, I can confirm the behaviour: accept a SQL Profile for a given SQL in one schema and verified that it is used in another schema (where the data volume and distribution is very different).
    I can also confirm the workaround - simply add a hint to the SQL to make it unique such that I could use different sql profiles for the otherwise exact same sql in different schemas.
    I'm happy enough with this workaround but I'll leave this thread unanswered in case someone can suggest a better approach.

  • How to read a byte data from maxdb data base

    Dear All,
    I have a issue in reading the data from database table.
    I have a column named as templateData which contains the byte data (biometric template data, which comes from fingerprint device) which is DataType of LONG and CODE of BYTE.
    I am not using the below to get the template data
    Connection con = null;
      Statement stmt = null;
      ResultSet resultSet = null;
    byte[] DbBioData = new byte[1024];
    InitialContext ctx = new InitialContext();
       if(ctx == null)
         throw new Exception("Boom - No Context");
       DataSource ds = (DataSource)ctx.lookup(db_drvstr);
       con = ds.getConnection();
       stmt = con.createStatement();
       resultSet  = stmt.executeQuery(db_query + " where SUBJECT_ID='"+ username +"'");
       if(resultSet.next())
        DbBioData = resultSet.getBytes(1);
        _loc.infoT("verify", "verify::Got BioData From MAXDB" +DbBioData );
        loc.infoT("verify", "verify::Query is: " +dbquery + " where SUBJECT_ID='"+ username +"'" );
    But I am not getting the proper data, could anyone please tell me the way to read the biometric data from data base table.

    Hi Kishore,
    is it me or is there no query definition in that code?
    I see that you concatenate a "db_query" with a string to make up a WHERE clause, but the db_query is nowhere defined before.
    So at least you should provide something like
    stmt = con.createStatement("SELECT templateDate FROM <tablename> ");
    before you do anything with the query.
    Besides this: have you ever heard of SQL injections? Try to use BIND-variables instead of concatenating strings. Otherwise your application will spend much time just with parsing your queries...
    Hmm... possibly the best thing you could do about this is to read the JAVA manual for MaxDB:
    <a href="http://maxdb.sap.com/currentdoc/ef/2de883d47a3840ac4ebb0b65a599e5/content.htm">Java Manual (SAP Library - Interfaces)</a>
    Best regards,
    Lars
    Edited by: Lars Breddemann on Dec 17, 2007 1:12 PM - corrected link

  • Data Distribution in the Data Volumes

    Hello
    Is it important that data be uniformly distributed in the data volumes?
    Is it possible to redistribute the data after adding some more file?
    Thank you & regards,
    T.C.

    As of MaxDB Version 7.7.06.09, such a mechanism can be activated using parameter EnableDataVolumeBalancing.
    If the parameter EnableDataVolumeBalancing (deviating from the default) is set to value YES, all data is implicitly distributed evenly to all data volumes after you add a new data volume or delete a data volume.
    https://service.sap.com/sap/support/notes/1173395

  • Material Master data distribution using the Classifications

    Hi All,
    I am attempting to cater for a business scenario where we wish to distribute material master records via IDOCS by assigning a class to each material we wish to send out. 
    We have a new classtype (Z01) which is setup for distribution purposes (the status for this class type has been maintained).  I have defined a filter using classes in BD60.
    The next step was to setup the distribution model for my message type and partner.  I added the filter group in the distribution model for the message type (filter group "dependant on class membership").  The final step after generating the partner profile was to assign a new class that I created to the material master.
    We are using change pointers and BD21 to create the IDOCs.  I am managing to create the outbound IDOC but I am getting a 29 Status with an error message "Class type Z01 , receiving system ZFR001 : no list maintained".  ZFR001 is the name of the external partner.  The name of the class that has been assigned to the material master is also ZFR001.  There are no characteristics assigned to the class.
    Has anyone setup a similar scenario?  I'd appreciate any guidance as I have hit a bit of a wall.
    Best regards,
    James.

    Hi,
    The message type to send HR data is HRMD_A, i don't think that filtering this only message you achieve what you describe.
    In this message type, yo can define filters to send only the relevant infotypes that you need, but all of them will be send in the same idoc, for one employee.
    So I think to achieve what you want, you should create customer Message types (Z's) as reductions of HRMD_A standard message type and then in BD64 create one different Model for each one. So in each model you can put only the infotypes you want. You should very carefull with the updating in your destination system, because if you process all the message type at the same time, there will be a lot of blocking in employee objects.
    It is weird what you want to do, but this could be the solution.
    Regards,

Maybe you are looking for

  • Is it possible to add a Mini Display Port converter to my MBP's thunderbolt to add a display ?

    I use MBP (late 2011). I find the hunderbolt port to be similar to the mini-display port of older mac. I wish to have another display, but not ready to spend $999 over a thunderbolt display from mac. So would like to know if a Mini-diaplay port conve

  • RegEx: How to find out which part of the pattern failed?

    Hi there, I was wondering: is there any way to find out where the pattern matching failed? Say I got the string "John Paul Yoko Ringo", and I want to match it against the pattern /John Paul George Ringo/. I would like to know something like "pattern

  • Entering special characters at the command line.

    I'm writing a simple program in java that automates certain processes in learning a foreign language. This includes holding a database of vocabulary within the context of a sentence and testing a student on words they have encountered. Nothing too tr

  • SAP Records Locking for preceeding documents in MM

    *Hi Experts,* Our client have a current requirement. They want that all the preceding documents to be locked once the subsequent documents are converted and posted. Eg- when PR converted to PO, PR to be locked or disable, PO converted to MIGO, PO to

  • BOXI Rel 3 - Data Tracking

    Hi All, I am trying to create a data tracking report which highlights the change of difference (value = current month value - previous month value). I will have a prompt which prompts user to select different year-month... Currently I can see the dat