HBASE-23779 Up the default fork count to make builds complete faster; make count relative to CPU count

Halve the build speed by doubling the -C count from 0.25C to 0.5C.
Pass mvn a -T of 0.5C too.

Author: saintstack

dev-support/hbase-personality.sh

@@ -140,7 +140,15 @@ function personality_modules
 
   clear_personality_queue
 
-  extra="-DHBasePatchProcess"
+  # Set a fork count based off the host cpu count.  Pass maven a -T argument.
+  # Default -T is one thread. 0.5C on an apache box of 16 cores and 2 jenkins
+  # 'executors' per host should make for 8 threads. Setting this here for yetus
+  # to pick up. See
+  # https://yetus.apache.org/documentation/0.11.1/precommit-advanced/#global-definitions
+  # See below for more on -T:
+  # https://cwiki.apache.org/confluence/display/MAVEN/Parallel+builds+in+Maven+3
+  forkcount="0.5C"
+  extra="-T${forkcount}  -Dsurefire.firstPartForkCount=${forkcount} -Dsurefire.secondPartForkCount=${forkcount} -DHBasePatchProcess"
   if [[ "${PATCH_BRANCH}" = branch-1* ]]; then
     extra="${extra} -Dhttps.protocols=TLSv1.2"
   fi

pom.xml

@@ -1413,13 +1413,15 @@
         by what apache jenkins nightly builds will tolerate (See HBASE-24072). Up this
       value is you want to burn through tests faster (could make for more failures
       if more contention around resources). There is a matching MAVEN_ARG
-      in our yetus personality where we set the maven -T command to 0.25C too.
-      For example, to run at a rate that is more furious than our 0.25C, do
+      in our yetus personality where we set the maven -T command to 0.5C too.
+      For example, to run at a rate that is more furious than our 0.5C, do
       something like this:
-         f="0.5C" ;  mvn -T$f -Dsurefire.firstPartForkCount=$f -Dsurefire.secondPartForkCount=$f test -PrunAllTests
+         f="0.75C" ;  mvn -T$f -Dsurefire.firstPartForkCount=$f -Dsurefire.secondPartForkCount=$f test -PrunAllTests
+      Note, tests seem to crash or fail nonsensically if the fork count is set too
+      high. On high-CPU boxes, a fork count of 1.0 doesn't always 'work'.
     -->
-    <surefire.firstPartForkCount>0.25C</surefire.firstPartForkCount>
-    <surefire.secondPartForkCount>0.25C</surefire.secondPartForkCount>
+    <surefire.firstPartForkCount>0.5C</surefire.firstPartForkCount>
+    <surefire.secondPartForkCount>0.5C</surefire.secondPartForkCount>
     <surefire.firstPartGroups>org.apache.hadoop.hbase.testclassification.SmallTests</surefire.firstPartGroups>
     <surefire.secondPartGroups>org.apache.hadoop.hbase.testclassification.MediumTests</surefire.secondPartGroups>
     <surefire.testFailureIgnore>false</surefire.testFailureIgnore>