[GitHub] [shardingsphere] strongduanmu commented on pull request #5327: create RANGE sharding algorithm

[GitBox <gi...@apache.org>]

strongduanmu commented on pull request #5327:
URL: https://github.com/apache/shardingsphere/pull/5327#issuecomment-619721301


   > @strongduanmu
   > 
   > Hi, I am glad to read your organized and straightforward code. :)
   > There is my concern about `range.partition.split.value` which is the key point of this algorithm.
   > 
   > Users need to list all endpoints of each range partitions, i.e, 10,20,30,40...
   > Besides, it generates complicated code handling. Do you think it is better to prefer the partition volume than listing all endpoints? That is, in this case: instead of `range.partition.split.value=10,20,30,40`, maybe we can consider `partition.volume = 10`. I am not sure, so I'd like to listen to your voice. 😀
   
   @tristaZero Partition volume looks more concise, but it can only split all values into same interval partition except the first and last partition, this way may not meet the scene of custom partition range.
   
   Users use partition volume to split all values, and may encounter some scenarios where the amount of partition data is small. At this time, it may be more desirable to be able to merge multiple partition ranges, that is, custom partition ranges.
   
   Is it better to provide two Range algorithms? One meets the requirements of flexible scenarios, and the other is more suitable for general scenarios.
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org