I'm not entirely convinced default JVM choices matter significantly to people on this list. However, defaults or more precisely their impact heavily contributes to the experience of majority of other JVM users.
I hope it doesn't spoil the the fun for the common programmer.
Having said that, my own experience is G1 may outperform other algos when dealing with larger heaps, but for up to 3-4 gb, UseParallelGC + UseNUMA + numactl still leaves it in the dust if you're after latency in high nines.
--
You received this message because you are subscribed to the Google Groups "mechanical-sympathy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-symp...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-sympathy+unsub...@googlegroups.com.
Lucene apparently has trouble with G1. So it will affect Solr as well.
--
Regards,
Janmejay
PS: Please blame the typos in this mail on my phone's uncivilized soft keyboard sporting it's not-so-smart-assist technology.
To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-symp...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups "mechanical-sympathy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-symp...@googlegroups.com.
To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-sympathy+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups "mechanical-sympathy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-sympathy+unsub...@googlegroups.com.
To unsubscribe from this group and stop receiving emails from it, send an email to mechanical-symp...@googlegroups.com.