[HADOOP-12065] Using more meaningful keys in EC schema

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Component/s: io
• Hadoop Flags: Reviewed
• Assignee: Kai Zheng
• Reporter: Kai Zheng
• Created: 04/Jun/15 23:42
• Updated: 30/Sep/15 19:38
• Resolved: 08/Jun/15 04:27

Description

Commented by szetszwo in HADOOP-12029:

We should also change "k" and "m" in the following to something like "numDataUnits" and "numParityUnits".

   public static final String NUM_DATA_UNITS_KEY = "k";
   public static final String NUM_PARITY_UNITS_KEY = "m";

This sounds good to me, as in fact in many erasure coding related papers, discussions, and implementations, though k and m are broadly used, the meanings are not consistent. For example, m can also mean numDataUnits + numParityUnits, all the rows in the coding matrix. The suggested names are more clear and explicit.