on 04-30-2014 2:53 PM
Hello experts ,
My DB's 1 table has 4 billion records , so I'm using range partition for this , I devided 10 partitions.
I want to change this range partition to hash partition.
In this time , I think I should do following .
1. execute "alter table bigtable merge partitions";
2. execute "alter table bigtable partition by hash (column1) partitons 10"
I think it will be correspond to 2 billion row limit in sequence 1.
How to execute re-partitioning in this case ? Should I unload current data to file or sometiong and load from this
Regards,
Jim
You can change the partitioning type without merging the partitions, and as you said you would hit the limit even if you did that. I would recommend stopping the SLT master job (if this is fed from SLT), then performing this operation. It will clearly take a good chunk of time to do so.
Also to note, hash will simply work on the column values that you partition on, you can't really guarantee that there will equal distributions across the 10 partitions.
Regards,
Justin
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
89 | |
10 | |
10 | |
9 | |
6 | |
6 | |
6 | |
5 | |
4 | |
3 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.