Out of curiosity, if you had a DB where there are 100,000,000 nodes and
you can cut each node name from twenty characters to 2, would that not
save the persistence requirements by a factor, I considering the holistic
system?
With XML parsing, this did play a factor in large document instances
simply because the SAX parser could hand off SAX events quicker. The
factor was not huge but the in memory size of a resultant DOM and the DOM
based (akak XSLT, XPath, XQuery) queries did perform marginally better.
Just being curious and feel free to impose a beer fine at the next user
group meeting if I am off base.
;-)
Duane
***********************************
Technoracle Advanced Systems Inc.
Consulting and Contracting; Proven Results!
i. Neo4J, PDF, Java, LiveCycle ES, Flex, AIR, CQ5 & Mobile
b.
http://technoracle.blogspot.com
t. @duanechaos
"Don't fear the Graph! Embrace Neo4J"
On 2012-10-06 5:28 AM, "Michael Hunger" <
michael...@neotechnology.com>
wrote:
>--
>
>