• State Suggested Answer
  • Locked Locked
  • Replies 9 replies
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 13760 views
  • Users 0 members are here
Options
Related
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

SharePlex vs RMAN

babudba
over 6 years ago

Hello All,

I have been working on Oracle Golden Gate quote long time but I was asked to do some POC in Shareplex for upcoming migration. 

From yesterday I'm reading Shareplex 9 with lots of documents but found lots interesting. Specifically, The below links has lots of details. 

https://gcn.com/~/media/GIG/GCN/Whitepapers/Quest_Top%205%20Reasons%20to%20Choose%20SharePlex.ashx

Unfortunately, I'm not able to find what I'm looking.

1. Is RMAN fully integrated with Shareplex ? Let say, I have multiple terabyte database which I'm looking for application migration / and database migration from one version to another version. In this case, I would like have to start RMAN backup and start with SCN like Golden Gate . 

Is it possible in shareplex ? Or only option RMAN Image backup ? If RMAN image backup then it may require downtime in source database. Is that correct ?

2.  In Golden Gate, We have option to do one time data copy. I mean initial data sync using DATA PUMP.. I can use this feature for smaller database. Is it possible from Shareplex ?

Thank you.  Babu

 

 

 

Parents
  • 0 over 6 years ago
    For your first question, recover database using sequence#. First we need to recover full log, no partial log recovery. In your example, you are using log 1234, then the reconcile is 1234.

    For SCN base, if you recover database until SCN then use the same number. SCN + 1 is for data pump, flashback_scn (flashback usage).

    SharePlex has the ability to column selection for replication. You can elect to replication only the columns that you want on the source. There are some items that you have to taken care of like PK/UK, constraint on the target. There might be a few more but you get the gist of it. The missing columns should not caused issue on the target. Also, column mapping is available so the column name on the source and target doesn't need to match. This tool is great for application upgrade.

    Typically, the target should have the same or more data (column/row) than the source and this is fine with SharePlex since we are name mapping everything so the sql will work fine. As I mentioned above is that the SQL that we apply should not caused any constraint error. For example, you have an extra column on the target but you have a non-null constraint on it. As we insert the data, we don't have that column since it doesn't exist on the source, the sql will failed. The solution either have a default value for that column or remove the constraint. If you follow this rule then replication will work beautifully.
Reply
  • 0 over 6 years ago
    For your first question, recover database using sequence#. First we need to recover full log, no partial log recovery. In your example, you are using log 1234, then the reconcile is 1234.

    For SCN base, if you recover database until SCN then use the same number. SCN + 1 is for data pump, flashback_scn (flashback usage).

    SharePlex has the ability to column selection for replication. You can elect to replication only the columns that you want on the source. There are some items that you have to taken care of like PK/UK, constraint on the target. There might be a few more but you get the gist of it. The missing columns should not caused issue on the target. Also, column mapping is available so the column name on the source and target doesn't need to match. This tool is great for application upgrade.

    Typically, the target should have the same or more data (column/row) than the source and this is fine with SharePlex since we are name mapping everything so the sql will work fine. As I mentioned above is that the SQL that we apply should not caused any constraint error. For example, you have an extra column on the target but you have a non-null constraint on it. As we insert the data, we don't have that column since it doesn't exist on the source, the sql will failed. The solution either have a default value for that column or remove the constraint. If you follow this rule then replication will work beautifully.
Children
No Data