Missing documentation for DSE or Neo4j depoyments

[raj-bains]

I'm working on my SaaS startup on top of Apache Spark, but want to use a graph database for metadata management.

I've spent 2 days trying to figure out how to use this and I can't. I love the gremlin-scala DSL (so elegant!!!!) and want to put it in my production application (ok, we're a startup - on path to production), but I want my developers to use it.

So, either I install Neo4j on serverA or I have DSE on serverA (DSE seems to be better since they distribute). How do I use gremlin-scala from my Spark code on serverB and connect to ServerA?

NOT ONE article in the community about this. What are the deployment options, what are the tradeoffs and how do I go about using it? Everyone jumps into the queries - I first need to install it you know and, no - I don't care about embedded on my local machine.

DSE java driver wants me to pass gremlin as strings - that sucks! But they'll be around and scalable since they have good sales.

Neo4j community forum doesn't want to help with Gremlin and want to push my to use Cypher and their unreadable scala DSLs

May I talk to mpollmeier for 15 minutes?

[mpollmeier]

There are potentially lot's of things to consider, and it all depends on your scenario. Most simplistic use case: start with the https://github.com/mpollmeier/gremlin-scala-examples repo and connect to a remote db rather than a local one.

If you want to discuss your specific scenario, flick me an email: michael@michaelpollmeier.com and I can provide some consulting.

[ksilin]

@raj-bains @mpollmeier did you come up with anything on this issue? I'm currently a bit disgusted about the prospect of using the all-String DSE Graph API. IUC, the primary issue with adapting gremlin.scala to DSE is establishing a gremlin connection with the server and mapping from a DseSession to a ScalaGraph. I dont have a clear idea how this could be done, help would be very much appreciated, would contribute an example should I succeed.

[raj-bains]

We never got this resolved.
For now we've built a class per graph node type and added DSL like
functions to have the code be readable
Also, there is a simple query builder that actually builds the strings
Not happy with it, but ugliness is contained.

On Fri, Oct 28, 2016 at 9:52 AM, Konstantin Silin notifications@github.com
wrote:

@raj-bains https://github.com/raj-bains @mpollmeier
https://github.com/mpollmeier did you come up with anything on this
issue? I'm currently a bit disgusted about the prospect of using the
all-String DSE Graph API. IUC, the primary issue with adapting
gremlin.scala to DSE is establishing a gremlin connection with the server
and mapping from a DseSession to a ScalaGraph. I dont have a clear idea
how this could be done, help would be very much appreciated, would
contribute an example should I succeed.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#168 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AEbfrMr1JMMHECPSbzKrD1bfkRdVqUUeks5q4ihGgaJpZM4JuzjL
.

[ksilin]

@raj-bains Thanks for the answer. Constraining and facading the stringyness is the approximate direction I have been heading as well :) Still very interested in adding a bit more safety though.

[ndelaforge]

+1
I would appreciate to have an example on how to connect gremlin-scala to DSE Graph.

[mpollmeier]

see #118 (comment)