GPU use in improving the calculating speed of algorithms

[shubham]

Hii Everyone,
is there any way one can use/implement GPU in neo4j to improve the query performance(especially graph algorithm for similarity measures).
if there is any way please mention the steps for same.
thankyou

[poonsfci]

Hi, I couldn't find any official documentation about GPU support in neo4j except the bloom for visualization. When you look at the Graph Data Science Library Installation (https://neo4j.com/docs/graph-data-science/current/installation/), it only talks about CPU, not GPU. Perhaps, additional tools/framework may be required.

[shubham]

Thanks for the reply @poonsfci,
is there anyway you are aware of using additional tool/framework?
please let me know the steps for the same .
thankyou.

[poonsfci]

Perhaps, the following article may help you in this area:

ZDNet

Nvidia Rapids cuGraph: Making graph analysis ubiquitous | ZDNet

A new open-source library by Nvidia could be the secret ingredient to advancing analytics and making graph databases faster. The key: parallel processing on Nvidia GPUs.

Noted that Networkx can access Neo4j data.

Hope that this can help.