The Fuseki Basic server would be good for running on a Pi which isn't
really that small - it is a complete linux system, runs JavaSE and can
have a large "disk" (SD card), not some embedded controller with a small
amount of NV RAM.
java -jar jena-fuseki-basic-VER-server.jar ...args...
No on-disk state.
Andy
On 03/07/17 07:51, Osma Suominen wrote:
[email protected] kirjoitti 01.07.2017 klo 17:23:
Now _that_ is the benchmarking project I would love to start-- forget
the Billion Triples Challenge, let's do the "how small a system can
you put a complete SPARQL server on". :grin:
If you're willing to relax "complete SPARQL server" a bit, it's been
done already. There was a demo at ISWC2014 showing a Raspberry Pi
running a Triple Pattern Fragments server with DBpedia data - but in
this case, the SPARQL engine was on the client side, not on the Pi. It
got the Best Demo Award.
Some more information here:
https://www.quora.com/Is-there-a-SPARQL-endpoint-or-triple-store-running-on-a-raspberry-pi
-Osma
