[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Knowledgegraph] Fwd: Re: CS6480 course project assignment - Network Management System - KnowOps

Very cool!


-------- Forwarded Message --------
Subject: Re: CS6480 course project assignment - Network Management System - KnowOps
Date: Wed, 21 Oct 2015 03:53:13 -0600
From: Aniraj Kesavan <aniraj@cs.utah.edu>
To: Jacobus (Kobus) Van der Merwe <kobus@cs.utah.edu>
CC: Mike Hibler <mike@flux.utah.edu>, Josh Kunz <joshkunz@me.com>, PRAVEEN THIRAVIYARATHINAM <u0992684@utah.edu>, u0545548 <renquinn@cs.utah.edu>


So, for the past week we were trying out titan and tinkerpop on various backends and we didn't get anywhere. Ren had suggested using bulbflow and neo4j, they have gremlin and titan integration but all the titan and tinkerpop problems were still coming up.

So we took a step back and just tried to realise a graph on neo4jserver using bulbflow's api, when we have something like that they should be pluggable to titan/tinkerpop once we get that working since tinkerpop can talk neo4j.

The graph is getting build from the parser script I wrote for the logs that Mike had given us (it's still getting built as I'm sending the e-mail).

The parser script is available here at /proj/utahstud-PG0/groups/knowops/parser_bulbflow.py (I didn't keep the logs there since it's publicly accessible). The script is super hacky and is basically trying to model the graph according to how I parse it.


They have this cool data browser tab where we can specify 'Cypher' queries. This link shows some sample queries that you could run.

For example,
start n=node(*)
match n-[:was_caused_by]-()
where has(n.type) and n.type='action'
return n
​
Once you get results for a query, you can pick a node from the results, try to show it's relationships and switch the visualisation mode. It gives it's best effort visualisation with limits on how much can be displayed.


Cheers,
Aniraj



On Fri, Oct 9, 2015 at 12:12 PM, Aniraj Kesavan <aniraj@cs.utah.edu> wrote:
Report Attached.


Thanks,
Aniraj

On Thu, Oct 1, 2015 at 9:20 PM, Aniraj Kesavan <aniraj@cs.utah.edu> wrote:
Report for the week.

Thanks,
Aniraj

On Thu, Sep 24, 2015 at 7:47 PM, Aniraj Kesavan <aniraj@cs.utah.edu> wrote:
Report for the week.

Thanks,
Aniraj

On Fri, Sep 18, 2015 at 10:24 AM, Aniraj Kesavan <aniraj@cs.utah.edu> wrote:
Report for this week.


Thanks,
Aniraj

On Wed, Sep 16, 2015 at 5:43 PM, Aniraj Kesavan <aniraj@cs.utah.edu> wrote:
Fine with me too

On Wed, Sep 16, 2015 at 4:29 PM, Jacobus (Kobus) Van der Merwe <kobus@cs.utah.edu> wrote:
On 9/16/15 3:35 PM, Mike Hibler wrote:
If Kobus doesn't mind, we will go ahead and meet in his office (I can get in)
since I don't have enough chairs in mine!
 Sure.




--
Aniraj Kesavan
MS Computer Science '15-'17,
School Of Computing
University Of Utah



--
Aniraj Kesavan
MS Computer Science '15-'17,
School Of Computing
University Of Utah



--
Aniraj Kesavan
MS Computer Science '15-'17,
School Of Computing
University Of Utah



--
Aniraj Kesavan
MS Computer Science '15-'17,
School Of Computing
University Of Utah



--
Aniraj Kesavan
MS Computer Science '15-'17,
School Of Computing
University Of Utah



--
Aniraj Kesavan
MS Computer Science '15-'17,
School Of Computing
University Of Utah