Approach for the removal of the physical dimension table

Similar Questions

• Increase the physical dimensions of the image in high resolution?

  I have an image I want to use in a video, if it is not physically large enough to fill the dimensions of 1920 x 1080 video, she has a resolution of 300ppi and video needs only 72 dpi. How to take advantage of the fact that it's an image high definition and increase the physical dimensions of the image? Is there a way to do this?

  Thank you.

  If you are told that you should print at 4 inches by 6 inches at 300 dpi, follow these steps:

  • Choose Image - Image size.
  • Check the box resampling .
  • In the resolution box, type 300 .
  • Put 4 to 6 inches in the height and Width boxes (or vice versa if it's a bitmap portrait).

  You will find yourself with an image 1200 x 1800 pixels that will be printed at 4 by 6 inches, your goal of printing at 300 ppi.

  Printing will be probably not as sharp and detailed as if you had printed at 4 by 6 inches without downsampling at 300 ppi.

  You will do well to try to begin to understand how the images are expressed in pixels, and why the above is true, while in the future, you will know when the issue of the requirements.  This thread you might also like:

  http://forums.Adobe.com/thread/975142?TSTART=30

  -Christmas

• How can I write the trigger for the global temporary Table

  Hi Grus,
  How can I write the trigger for the global temporary Table.
  
  I created the TWG with trigger using the script below.
  
  
  CREATE A GLOBAL_TEMP GLOBAL TEMPORARY TABLE
  (
  EMP_C_NAME VARCHAR2 (20 BYTE)
  )
  ON COMMIT PRESERVE ROWS;
  
  
  CREATE OR REPLACE TRIGGER TRI_GLOBAL_TEMP
  BEFORE DELETE, UPDATE OR INSERT
  ON GLOBAL_TEMP
  REFERRING AGAIN AS NINE OLD AND OLD
  FOR EACH LINE
  BEGIN
  INSERT INTO VALUES EMPNAME (: OLD.) EMP_C_NAME);
  END;
  /
  
  
  trigger was created successfully, but her would not insert EMPNAME Table...
  
  Please guide if mistaken or not? If not wanting to give a correct syntax with example
  
  
  Thanks in advance,
  Arun M M

  BEGIN
  INSERT INTO EMPNAME VALUES (:OLD.EMP_C_NAME);
  END;
  

  you are referencing old value in insert stmt.
  
  BEGIN
  INSERT INTO EMPNAME VALUES (:new.EMP_C_NAME);
  END;
  

  then run your app, it works very well...

  CREATE GLOBAL TEMPORARY TABLE GLOBAL_TEMP
  (
  EMP_C_NAME VARCHAR2(20 BYTE)
  )
  ON COMMIT PRESERVE ROWS;
  
  CREATE OR REPLACE TRIGGER TRI_GLOBAL_TEMP
  BEFORE DELETE OR UPDATE OR INSERT
  ON GLOBAL_TEMP
  REFERENCING NEW AS NEW OLD AS OLD
  FOR EACH ROW
  BEGIN
  dbms_output.put_line(:OLD.EMP_C_NAME||'yahoo');
  INSERT INTO EMPNAME VALUES (:new.EMP_C_NAME);
  dbms_output.put_line(:OLD.EMP_C_NAME);
  END;
  /
  
  create table EMPNAME as select * from GLOBAL_TEMP where 1=2
  
  insert into GLOBAL_TEMP values('fgfdgd');
  commit;
  select * from GLOBAL_TEMP;
  select * from EMPNAME;
  
  output:
  1 rows inserted
  commit succeeded.
  EMP_C_NAME
  --------------------
  fgfdgd               
  
  1 rows selected
  
  EMP_C_NAME
  --------------------
  fgfdgd               
  
  1 rows selected
  

  He got Arun

  Published by: OraclePLSQL on December 28, 2010 18:07

• FOR the UPDATE OF table.field

  Why the following does it
  

  create table division (code varchar2(2) primary key, div_desc varchar2(20));
  
  insert into division values ('01', 'Ninja assassins');
  insert into division values ('02', 'Working for the man');
  
  
  create table employees (tk number, first_name varchar2(10), last_name varchar2(10), code varchar2(2) references division(code));
  
  insert into employees values (1, 'Chuck', 'Smith', '01');
  insert into employees values (2, 'John',  'Smith', '02');
  
  DECLARE
  
    CURSOR my_csr IS
      Select e.tk, e.first_name, e.last_name
      From employees e, division d
      Where e.code = d.code
        and e.code = '01'
        and e.last_name = 'Smith'
      For update;
      
      cnt_updated NUMBER;
  
  BEGIN
  
    cnt_updated := 0; 
  
    FOR my_row IN my_csr
    LOOP
  
      Update employees
      Set last_name = 'Forbes'
      Where current of my_csr;
      
      cnt_updated := cnt_updated + SQL%ROWCOUNT;
      dbms_output.put_line('You updated '||cnt_updated||' records');
  
    END LOOP;
  END;

  but just add the clause "for update" allows to make the point?

  DECLARE
  
    CURSOR my_csr IS
      Select e.tk, e.first_name, e.last_name
      From employees e, division d
      Where e.code = d.code
        and e.code = '01'
        and e.last_name = 'Smith'
      For update *of e.tk*;

  We discovered this recently, and fellow developers are wondering "why?" Is there a reason documented, or is - just how it is?
  
  Thank you
  -= Chuck

  of 'of the user to the database PL/SQL Oracle® reference Guide. "
  "10g Release 2 (10.2):
  "During the interrogation of several tables, you can use the FOR UPDATE clause to limit the line blocking to specific tables. Rows in a table are locked unless done FOR UPDATE OF clause refers to a column in the table. For example, the following query locks the rows in the employees table, but not in the departments table:

  DECLARE
  CURSOR c1 IS SELECT last_name, department_name FROM employees, departments
  WHERE employees.department_id = departments.department_id
  AND job_id = "SA_MAN."
  FOR the UPDATE OF treatment; »

  a little demo:

  SQL> select * from v$version;
  
  BANNER
  -------------------------------------------------------------------------
  
  Oracle Database 11g Enterprise Edition Release 11.1.0.6.0 - Production
  PL/SQL Release 11.1.0.6.0 - Production
  CORE    11.1.0.6.0      Production
  TNS for 32-bit Windows: Version 11.1.0.6.0 - Production
  NLSRTL Version 11.1.0.6.0 - Production
  
    1   DECLARE
    2     CURSOR my_csr IS
    3       Select e.tk, e.first_name, e.last_name, e.rowid
    4       From employees_t e, division d
    5       Where e.code = d.code
    6         and e.code = '01'
    7         and e.last_name = 'Smith'
    8       For update of e.tk;
    9       cnt_updated NUMBER;
   10   BEGIN
   11     cnt_updated := 0;
   12     FOR my_row IN my_csr
   13     LOOP
   14       Update employees_t
   15       Set last_name = 'Forbes'
   16       Where current of my_csr;
   17       cnt_updated := cnt_updated + SQL%ROWCOUNT;
   18       dbms_output.put_line('You updated '||cnt_updated||' records');
   19     END LOOP;
   20*  END;
  SQL> /
  You updated 1 records
  
  PL/SQL procedure successfully completed.
  
  SQL> ed
  Wrote file afiedt.buf
  
    1   DECLARE
    2     CURSOR my_csr IS
    3       Select e.tk, e.first_name, e.last_name, e.rowid
    4       From employees_t e, division d
    5       Where e.code = d.code
    6         and e.code = '01'
    7         and e.last_name = 'Smith'
    8       For update;-- of e.tk;
    9       cnt_updated NUMBER;
   10   BEGIN
   11     cnt_updated := 0;
   12     FOR my_row IN my_csr
   13     LOOP
   14       Update employees_t
   15       Set last_name = 'Forbes'
   16       Where current of my_csr;
   17       cnt_updated := cnt_updated + SQL%ROWCOUNT;
   18       dbms_output.put_line('You updated '||cnt_updated||' records');
   19     END LOOP;
   20*  END;
  SQL> /
  
  PL/SQL procedure successfully completed.
  

  Amiel

• Problem creating hierarchy based on 2 physical dimension tables

  I'm having a problem create 1 dimension logic with a hierarchy of exploration, based on two separate physical dimension tables. The errors I get when navigating in the exploring hierarchy is:
  "Cannot find coverage source logical table to the columns of the logical" &
  "Missing join between the logical tables.
  
  I use OBIEE 10.1.3.4
  Here are the details of what I have set up as:
  
  Physical layer:
  Table DIM_ORG with the dimension columns:
  -dimension_key
  -org_total_code
  -org_total_description
  -org_detail_code
  -org_detail_description
  
  Dimension table DIM_DEPT with columns:
  -dimension_key
  -dept_total_code
  -dept_total_description
  -dept_detail_code
  -dept_detail_description
  
  FACT_SALES table with columns of facts:
  -fk_org
  -fk_dept
  -sum_sales
  
  Physical joins:
  FACT_SALES.fk_org = DIM_ORG_dimension_key
  FACT_SALES.fl_dept = DIM_DEPT.dimension_key
  
  Business model and the mapping of layer:
  I created a logical dimension ORG_DEPT. It contains two sources of logic table (DIM_ORG & DIM_DEPT) and the following logical columns:
  -All departments (mapped to dept_total_code)
  -Organization (mapped to org_detail_description)
  -Organisation number (mapped to org_detail_code)
  -Department (mapped to dept_detail_description)
  -Department Code (mapped to dept_detail_code)
  The logical key of the company is based on the combination of number of organization & Department Code
  
  The hierarchy, I need is: all departments-> organization-> Department so I created the following hierarchy for ORG_DEPT:
  -Total level containing: all departments
  -Organization level containing: company (defined as logical level key) number & Organisation (als defined the key level drill)
  / Department detail level containing: Department (defined as logical level key) Code and the Department (defined as key level drill).
  
  In the LTS of the ORG_DEPT dimension, I've set levels of content for sources:
  DIM_ORG: Level of organization
  DIM_DEPT: Level of detail Department
  The LTS non - joins inner - came against the associated physical tables.
  
  I created a logical fact table (based on the physical fact table) SALES and joined him on the ORG_DEPT logical dimension table.
  In the LTS, level of content for ORG_DEPT is part of the level of retail service. Non - joins inner were aded against related physical tables.
  
  When I create a report in response to test the hierarchy and select only "all departments' I get the value of the returned correct size. When I try to break through to the next level, I get the following ODBC error:
  "Could not find coverage source logical table for logical columns: [all of them]." "Please check more detailed level keys are correctly mapped.
  
  When I create a report in responses and select "Every department" and "Sales", I get the correct result. When I try to break through to the next level, I get a different ODBC error:
  "Lack of join between the logical tables DIM_DEPT and DIM_DEPT: there must be at least a physical link to join between the underlying physical tables.
  
  All suggestions are welcome!
  
  Thank you!

  You have no relationship defined between the Department and org.

  You can either:
  a. create a table of physical dimension with a composite key of org_id & dept_id if there is a relationship defined somewhere, then to obiee model as you have already done, but in the physical layer on the composite key join
  b. model your exisitng as two-dimensional and two hierarchies tables and set up a path of forest favorite between them according to your needs.

  Kind regards

  Robert

• Best approach for the design of the Radial Image - sector scanning Sonar

  This is not a problem, but you are asked to think about the best approach.

  I need a sector sonar image data.  This device runs on 360 (degree sub-segments) and generates a set of data from the origin - radially outwards at an angle of no - to an endpoint.  Therefore, for each 'ping' sonar I need data points in the center of a circle graph, a radial angle on the outside.

  I don't know if I should use LabView vi for the drawing tools, or go to ActiveX or?

  Anyone has a suggestion on the best approach to this?

  Thank you
  Peter

  Peter,

  You can use a chart of the LabVIEW intensities for this kind of display.

  Maybe not the best way, but it is possible.

  As an attachment, I did an example with your specifications.

  datapoints 1024 by degree

  See example image:

  

  Basic principle:

  -Create the matrix of 2048 x 2048 points database

  -For each 'line of angle' calculate which pixel to replace each data point in the line

  -Because "datapoints" becoming bigger, far from the Center, repeat the action for several "sup angles. for example, to 85 deg also Calc. 84.6, 84.8, 85, 85.2 85.4

  It gives you an idea of the possibilities, but performance may be necessary.

  -Only external Calc x pixels more degrees

  -Do some kind of anti-aliasing to avoid the pixelated lines.

  See attached example. (LV 8.6)

• How to get the MAC address for the physical NETWORK adapter of the HOST ESX (vmnic0/vmnic1)

  I can recover physical NIC details for the server HOST ESX, but MAC is empty... How can I get?

  Thank you/HKS

  $net = get-VMHostNetwork - VMHost myServer.domain.net

  $net | FC

  class VMHostNetworkInfoImpl

  {

  VMKernelGateway = 166.12.113.1

  ConsoleGateway = 166.12.113.1

  ConsoleGatewayDevice = vswif0

  DnsAddress =

  [

  169.40.0.2

  169.40.0.1

  ]

  DnsFromDhcp = False

  DnsDhcpDevice =

  DomainName = domaine.net

  Host name = hostname1

  SearchDomain =

  [

  EU.hedani.NET

  ]

  PhysicalNic =

  [

  class PhysicalNicImpl

  {

  BitRatePerSec = 1000

  FullDuplex = True

  WakeOnLanSupported =

  ID = key - vim.host.PhysicalNic - vmnic0

  DeviceName = vmnic0

  Mac =

  DhcpEnabled = False

  IP =

  SubnetMask =

  }

  class PhysicalNicImpl

  {

  BitRatePerSec = 0

  FullDuplex = False

  WakeOnLanSupported =

  ID = key - vim.host.PhysicalNic - vmnic1

  DeviceName = vmnic1

  Mac =

  DhcpEnabled = False

  IP =

  SubnetMask =

  }

  ]

  VirtualSwitch =

  [

  class VirtualSwitchImpl

  {

  Key = key - vim.host.VirtualSwitch - vSwitch0

  Name = vSwitch0

  NumPorts = 128

  NumPortsAvailable = 105

  NIC =

  [

  vmnic1

  vmnic0

  ]

  MTU =

  }

  class VirtualSwitchImpl

  {

  Key = key - vim.host.VirtualSwitch - vSwitch1

  Name = vSwitch1

  NumPorts = 64

  NumPortsAvailable = 56

  NIC =

  MTU =

  }

  class VirtualSwitchImpl

  {

  Key = key - vim.host.VirtualSwitch - vSwitch2

  Name = vSwitch2

  NumPorts = 64

  NumPortsAvailable = 64

  NIC =

  MTU =

  }

  ]

  ConsoleNic =

  [

  class HostConsoleVirtualNicImpl

  {

  PortGroupName = Service_Console_VLAN1

  ID = key - vim.host.VirtualNic - vswif0

  DeviceName = vswif0

  Mac = 00:50:56:4 c: 1 d: 3B

  DhcpEnabled = False

  IP = 111.11.111.11

  Subnet mask = 255.255.255.0

  }

  class HostConsoleVirtualNicImpl

  {

  PortGroupName = Service_Console_ISOL

  ID = key - vim.host.VirtualNic - vswif1

  DeviceName = vswif1

  Mac = 00:50:56:4 d: 88:87

  DhcpEnabled = False

  IP = 10.0.0.241

  Subnet mask = 255.255.255.0

  }

  ]

  VirtualNic =

  [

  class HostVMKernelVirtualNicImpl

  {

  VMotionEnabled = True

  PortGroupName = Vmotion_VLAN1

  ID = key - vim.host.VirtualNic - portgroup3

  DeviceName = portgroup3

  Mac = 00:50:56:65:04:56

  DhcpEnabled = False

  IP = 166.12.113.72

  Subnet mask = 255.255.255.0

  }

  ]

  }

  get-vmhost | % {(get-view $_.id).config.network.pnic} | select Device, Mac
  

  http://www.vmwarescripting.com

• Search for the nearest value table

  IM wondering if its possible to find a 1 d table worth, if it is not found, then find the nearest match. IM using the 'double' data type and I need at least an accuracy of 4 decimal point in the research.

  For now, im rounding decimal values to the nearest integers from you, but it is not effective at all. Im getting a lot of unwanted split in the plot. I need to be able to search for the decimal points. I found this function "About Equals.VI" starting from a previous thread. Even that uses the concept of the rounding.

  I enclose 3 screws "0.8 & 0.4.VI" are that I created, rounding decimals nearest digit.

  The "approximately equal" is the one I found online.

  The 3rd is I try to mess with what I can find as close a match.

  Y at - it a trick to work around this problem?

  Thank you

  Eureka

  This VI uses the first type of step 3 in your data.  Since you said that you can get the data, you can drop the in VI.

  At the end where I calculate the slope, you can use a linear adjustment if you have full or professional LabVIEW.

  

  

• Help for the reconciliation of table. Average cost

  

  So this is my attempt to imitate a function block that we use in our standard converter software - "Decimation filter" which is nothing more than a running average / mobile. The sample size is adjustable to execution of 2 to 64 samples (decimation factor). I saw many topics on this and used to average around 4 large samples shift registers - but I wanted to be able to change the sample size without recompiling. I'm new to LV, there is likely a lot of better ways to do this.

  I would like to have answered is linked to clear the table, if the decimation factor is set to a lower number than the last time that the loop executed. (The uppercase - false statement is wired directly by)

  The math in the shift register: creates an array index that cycles from 0 to (decimation Factor-1). The index is then used to fill elements in the table (the rest being zeros). When the decimation factor decreases, I need zero external element in (former decimation factor - 1) (new decimation Factor-1) positions. So I tried various things, but the only thing that seems to work, it's the re - initialize the array. I think it's less than optimal.

  I tried:

  (1) leaving the tunnel of continuous wire output for the real deal and selecting the option 'use default if unwired' - thinking I'd get a table of 64 elements, of zeros. Doesn't seem to work.

  (2) a constant matrix cable tunnel exit if this value is true. When I followed him, after a decline in decimation factor - the probe seems to indicate an array of elements, not 64 No. And I do not see how to specify the size of the array of constant matrix.

  If I use this in my application it will run on a target of cRIO.

  Any help much appreciated.

  You already have your initialized table, why not use it?  Wire your table initialized via if this value is TRUE.  Or better yet, use Select? function.

  

• Solution for the mutation of table of error when using triggers

  Hello

  Could you please give the solution for the error table mutation during the use of triggers. I'll give you the simple scenario here,

  I created a trigger on the employees table, whenever all DML operations takes place in the employees table, it must run the trigger body. I intentionally used also of the employees in the body of the trigger table. Please give me the solution.

  Triggering factor:

  create or replace trigger test_trigger

  before you insert or update or delete employees

  for each line

  declare

  an employees.first_name%type;

  Start

  Select first_name from people where job_id = "AD_PRES";

  end;

  DML statement:

  Update employees set salary = 20000 where job_id = "AD_PRES" (I run this query)

  Error message:

  ORA-04091: table HR. Is the transfer of EMPLOYEES, the function of triggering/can not see

  ORA-06512: at "HR. TEST_TRIGGER', line 4

  ORA-04088: error during execution of trigger ' HR. TEST_TRIGGER'

  Can anyone tell, in what other scenarios, we get this error or recursive table mutation?

  Thanks in advance

  Hello

  When you create a trigger on a table, you cannot edit/query this table the trigger is not completed.

  You can use

  (1) transaction of Pragma autonomous

  (2) instead of row-level trigger, you have statement-level trigger

  (3) to 11 g, try with COMPOSITE trigger

  Try below to avoid it (a solution):

  create or replace trigger test_trigger

  before you insert or update or delete employees

  for each line

  declare

  PRAGMA AUTONOMOUS_TRANSACTION;

  an employees.first_name%type;

  Start

  Select first_name from people where job_id = "AD_PRES";

  COMMIT;

  end;

• Format mask for the size of table cells

  Hi all

  I'm working on a catalog with a lot of products, and each product has a table for its specifications.

  The catalogue is ready for printing, but now the customer asks me to enlarge certain parts of the text in the table.

  This means that I have to give some cells more space. The cells have different sizes, and that's the problem:

  Is there a way to define the new cell widths and heights as a sort of "master format" so that I can change the tables
  for the design of new without marking of each column and adjusting by hand?

  Unfortunately, the layout is too complex to have tables on a master page, as they are not on a particular coordinate

  throughout the catalog.

  I hope I described my problem quite clearly I'm no native speaker

  I thank very you much in advance!

  Max

  The script linked in the post one of this thread will allow you to adjust automatically to match - table column widths it allows to save the widths of a table in the style of the table and then apply them to other tables of this style.  I don't know about heights, but if all the heights will be the same within a given table, you can select all the lines and enter the exact height of the cell.

  Re: width of column in the specs of the cell

• ESX host with local storage recommended for the host partition table?

  Hi all

  I know that the best practical partition table should look like this:

  Mount Point	Partition	Size	Description
  		/ dev/sda (primary)
  / Boot	ext3	250 MB	Change of extra space for upgrades
  N/A	SWAP	1 600 MB	Change the maximum size of Exchange service console
  /	ext3	5120 MB	Change of extra space in the root
  		/ dev/sda (Extended)
  / var	ext3	4096 MB	Create partition to avoid an overflow of root with log files
  / tmp	ext3	1024 MB	Create partition to avoid an overflow of root with temporary files
  / opt	ext3	2048 MB	Create partition to avoid overfilling root with the log files of VMware HA
  / Home	ext3	1024 MB	Create partition to avoid an overflow of root with agent / user files
  	vmkcore	100 MB	Pre-configured

  However with a system with local storage would you set it up the same way, and whatever it is must be amended to change to get ESX to write log files to the different partitions?  I would like to configure it for future expansion for when the enterprise package is configured.

  Yes, I would set up the same way with an ESX with local storage server and nothing needs to be changed for the vmekernel right at the partition/var - unless of course you want to write to a log server that you need to change a configuration file to point to the server logs.  If you talk to the ESX Enterprise license you willnot have to change partiition table-

  Also on something else if you don't have a SAN/NAS storage you will also need to create a VMFS partition on the local data store to host your virtual machines

  If you find this or any other answer useful please consider awarding points marking the answer correct or useful

• Could not import a custom in the physical layer table

  Background
  **********
  We are implementing the OBI Applications
  The RPD is OBI (analytical) applications, a. We created custom tables and are now building the logic layer.
  
  Problem
  *******
  Created a new custom table in the physical OBAW using the username of the OBI (analytical objects owner)
  
  Connected to the BI-Admin offline using the user/pwd administrator name
  
  Utility of used import, the appropriate custom table table (you can see in the dialog box), taken from the warehouse. It brings up the dialog box, and it looks like it is to import the table (by displaying the file flying over).
  
  Updating of the RPD. Even tried to connect and again.
  
  When you check the physical layer you do not see the custom table. Y at - it a cache that needs to be eliminated?

  have you checked if it creates a new connection pool and the table under this new connection pool?

• SELECT... FOR the vs LOCK TABLE of UPDATE... MODE OF LINE SHARING IN

  Hello
  In practice, what is the difference between the two versions of line locking methods...?
  In my case, I need to block only the lines that will be extracted by a cursor... but because the select stmt using the sum aggregate function... I can't use the format:

  cursor c1 
  
   is
  
    select x , y ,sum(a)
  
      from d
  
     group by x,y
  
      for update;

  So, the only alternative I have (I think) is the "LOCK TABLE d PART of the LINE in MODE.
  Is the above correct... to obtain row locks the same as the "select" for the update?
  
  Note: I use Db10g v.2
  Thank you
  SIM

  When you draw a select... Updated, it's immediately another session in standby mode until you release the lock.
  TABLE LOCK is a bit different, the other session will run in standby mode after an UPDATE statement. If the other session, run an UPDATE statement before your own session, you will wait for the release lock.
  So, it's different behaviors. Conclusion: they are not same.

  Nicolas.

• Performance problem with SQL for recovering data in FACT and Dimension Table

  Hello

  We have a query that performs very slow when extracting data in FACT and the DIMENSION Table.

  SQL and Plan are in the attachment.

  Please suggest how this can be improved.

  The version of DB is 11.2.0.4

  Kind regards

  VN

  Slow is a very relative term, for about 1 hour is slow for some 1s is slow. So please always quantify your settings.

  The first thing caught my eye is done 4 times ia_oasis_tot_prov_cur_d access. Try to isolate which and change it like this and see if it returns the same result.

  select case when a.clm_capitn_ia_bil_prov_edw_sk = b.prov_edw_sk then b.prov_id end ben_billing_prov
       , case when a.clm_capitn_attnd_prov_edw_sk  = b.prov_edw_sk then b.prov_id end ben_attending_prov
       , case when a.clm_capitn_ia_pcp_edw_sk      = b.prov_edw_sk then b.prov_id end ben_pcp_number
       , case when a.clm_capitn_ia_ipa_prov_edw_sk = b.prov_edw_sk then b.prov_id end ben_ipa_number
    from edw.clm_capitn_f partition (jan2015) a
    left join
         edw.ia_oasis_tot_prov_cur_d b
      on (
             a.clm_capitn_ia_bil_prov_edw_sk = b.prov_edw_sk
          or a.clm_capitn_attnd_prov_edw_sk  = b.prov_edw_sk
          or a.clm_capitn_ia_pcp_edw_sk      = b.prov_edw_sk
          or a.clm_capitn_ia_ipa_prov_edw_sk = b.prov_edw_sk
         )
  

  If it works this means allows you to reduce certain I/O.