Hi David
I was thinking on these lines my self, but was unable to formulate it. I think
Ill elaborate on the actual problem as you've suggested.
There are a number of college students who I have gathered various information
about, example:
1. What their major is (4 options)
2. What year they
A nice clear post. The choice of "Router" is obviously key. For the given
routing examples based on user or geo it should be possible to map a request to
a server. For many other situations it may prove much harder to determine which
server has a warm cache because there is no user and there is
Agam,
Depending on the set of possible values, you could represent the properties
with relationships instead. A unique property value can then be represented
by a node, which would be linked to all nodes that have that value. The
relationship type could indicate the property. The "value" nodes wou
Hi Cedric, thanks for the quick reply!
Yes, it's not very 'graphy' but it is a viable solution. My only concern is,
what may not have been so clear earlier:
The properties have to be the same, i.e.,
If A, C, D, E, F, G are the nodes that have the 'x=4' properties in common, the
properties may be
Very nice post Marko, clean and to the point!
Cheers,
/peter neubauer
GTalk: neubauer.peter
Skype peter.neubauer
Phone +46 704 106975
LinkedIn http://www.linkedin.com/in/neubauer
Twitter http://twitter.com/peterneubauer
http://www.neo4j.org - Your high perf
Traverse the graph returning all nodes with at least 1 matching property.
For each node returned add up the matching properties to node A and
put that in a list.
Sort the list.
On Thu, Feb 24, 2011 at 1:49 AM, Cedric Hurst wrote:
> Not sure if its the best one, but one possible strategy would be
Brendan,
I saw something similar on Windows,I don't remember what was the
cause, but could you run mvn clean install in order to make sure
things are cleaned up properly?
Cheers,
/peter neubauer
GTalk: neubauer.peter
Skype peter.neubauer
Phone +46 704 106975
LinkedIn http://ww
Craig,
great work!
Tobias helped me to refactor the BerkeleyDB index to work with the
latest Indexing changes, so we should be able to upgrade even that
part to Neo4j 1.3.M03 next week and test. Together, we might be able
to get a better grip on insert performance, which is the main annoying
thing
Not sure if its the best one, but one possible strategy would be to
define a comparator with a constructor that takes an argument of your
comparison node (A), and then implement the compareTo() function to
retrieve the number of common properties for one node against the
number of common properties
Hey
I'm a graph database and Neo4j newbie and I'm in a bit of a fix:
*Problem Description*
Let's say I have 'n' nodes in the graph, representing the same type of
object. They have certain undirected links between them.
Now each of these 'n' nodes has the same 10 properties, the *values* of
which
Hi David and Michael,
Firstly, I knew that field argument had to be there for a reason, I
just couldn't figure it out from the examples. ;-) I think your
explanation is exactly what I was looking for, Michael. I'll give it
a shot tomorrow.
And David, you are correct, I'm looking to create two
Thx Mattias.
-
Regards,
Francois Kassis.
--
View this message in context:
http://neo4j-user-list.438527.n3.nabble.com/Neo4J-Rest-server-API-tp2559737p2565570.html
Sent from the Neo4J User List mailing list archive at Nabble.com.
___
Neo4j mailing
Hi,
I tried to build from the source code with Maven but failed with "The
source must not be a directory."
I did some search on internet found this is a unresolved bug.
I wonder how do you overcome it?
Cheers,
Brendan
--- maven-dependency-plugin:2.1:unpack-dependencies (get-sources) @ neo4j --
Cedric,
Thank you for your feedback! We value it highly.
I'm trying to understand your use case. You have entities of classes A, B,
and C. Your graph looks like this:
A --RELATED_TO--> B
A --RELATED_TO--> C
And you want to provide query methods for getting all B's and C's for a
given A. Correct
I do say, the good gentleman has produced a clearly written, appropriately
illustrated, and highly valuable post.
Cheers,
Andreas
ps. Excepting a few tiny errors in spelling and punctuation.
On Feb 24, 2011, at 1:17 AM, Jim Webber wrote:
> Hi folks,
>
> I've written up my thoughts on the cac
Hi folks,
I've written up my thoughts on the cache sharding pattern on my blog. See:
http://jim.webber.name/2011/02/23/abe72f61-27fb-4c1b-8ce1-d0db7583497b.aspx
Jim
___
Neo4j mailing list
User@lists.neo4j.org
https://lists.neo4j.org/mailman/listinfo/us
Hi,
I wrote a blog post regarding knowledge representation and reasoning using
graph databases:
http://markorodriguez.com/2011/02/23/knowledge-representation-and-reasoning-with-graph-databases/
Automated reasoning is currently dominated by RDF and its stack of
technologies. However, th
Cedric,
thanks so much for your feedback.
By now I kept the template and the NodeBacked kind of separate.
The Neo4jTemplate API has changed a bit :) see here:
https://github.com/SpringSource/spring-data-graph/blob/master/spring-data-neo4j/src/main/java/org/springframework/data/graph/neo4j/templa
Hi,
I just merged in David Winslows CQL support into Neo4j Spatial. Basically it
is an alternative to the JSON based DynamicLayers we provided for OSM. The
main advantages of CQL are:
- Works on the Geometries themselves, using the very extensive CQL
syntax, an SQL-like syntax for querying
Firstly, thanks for your great work on the Spring Data Graph API so far. I
took a look at the draft you had out here:
https://gist.github.com/835408
And it looks like an awesome start. I'm relatively new to Neo4J (saw the
SpringOne keynote and the roo talk), but one thing I think would be very
2011/2/22 Brendan Cheng
> Hi Mattias,
>
> Does the latest online backup work on backup to a running read only db?
>
I think there may be problems with cache invalidation on that running
read-only database and I just tried and there were other problems as well...
so, no... not at the moment.
>
>
2011/2/23 francoisk6
>
> Hi michael, thx for the reply.
>
> ok, if this is the new way to index:
> graphdb.index().forNodes(indexName).add(node,field,value) .
> What about the lucene fulltext search and the query search.
> What i need at end of the day, is to index data using java. and then using
Hi,
I tried adding both jars you said at server-project level with and it
finally worked !
However when these two jars where included in libraries which where maven
dependencies used by the server project, it
was like they weren't there and I kept getting the same error.
I have to say that I don'
Hi Alfredas,
One thing I forgot to mention. You can express pattern match queries in Gremlin:
https://github.com/tinkerpop/gremlin/wiki/SPARQL-vs.-Gremlin
While not having the same syntax, you can still do pattern matching as a
traversal. But yes, I ultimately want to get an extension to
Ok it's out there on 0.0.11
Thing is when you add a node to an index (before creating the index),
Neo4j creates the index on its own.
@neo.create_node_index(name, type, provider) # creates
an index, defaults are "exact" and "lucene"
@neo.create_relationship_index(name, "fulltext")
>>Since I want to later search for nodes by their mysql_id, do I need to create
>>an index and add the
mysql_id to that index? It appears as if that is the only way to
efficiently find nodes by a key
other than their neo generated ID. Before I went down the path of
creating indexes, I wanted to
Very nice, thanks.
A follow-on question:
Since I want to later search for nodes by their mysql_id, do I need to create
an index and add the
mysql_id to that index? It appears as if that is the only way to efficiently
find nodes by a key
other than their neo generated ID. Before I went down
Hi michael, thx for the reply.
ok, if this is the new way to index:
graphdb.index().forNodes(indexName).add(node,field,value) .
What about the lucene fulltext search and the query search.
What i need at end of the day, is to index data using java. and then using
the rest api to search for nodes.
Thanks for the response.
Then my idea of a server plugin wasn't a bad idea, great.
My next question is then: how do I traverse only a part of the possible
sub-graph?
I mean: let's suppose I start traversing from node 'A' and want to get
all 2 length paths on relationships 'TYPE_X' and 'TYPE_Y'.
What was the actual index name you used for indexing?
So the part of your sample app which does:
graphdb.index().forNodes(indexName).add(node,field,value)
Just creating the index-service for the graph-db (which is imho the
old index API anyway) won't index anything.
Michael
On Wed, Feb 23, 201
First - you should perhaps write a Server-Plugin that does your heavy
lifting on the server and provides a REST endpoint to get the results.
Not sure if non-GET verbs are supported yet (otherwise you can always
go for an unmanaged extension defining your own resources).
You can do indexing for cer
Hi,
Thanks for the details. I did increase the test data and it does slow down a
lot. At the moment I have 3 nodes and it computes("touches") between 4 and
8
millions of nodes for each request :) But still does it in an honorable 5
seconds average.
I understand what you mean for the node
Dear All,
I a using neo4J 1.2, i filled a sample data and indexed each inserted new
node using java with:
IndexService index = LuceneIndexService(graphDb)
IndexService fulltextindex = LuceneFulltextIndexService(graphDb)
but in rest api when using:
curl -H Accept:application/json http://localhos
Hi all,
I'd like to get ideas on how to handle a (relatively) big graph. My
graph is stored in a neo4j server. The structure is simple but highly
interconnected:
- I have nodes containing longer texts
- and I have many nodes containing tokens of those texts.
Relationships connect tokens to texts
Another thing to consider is that your traversal is loading node properties,
which can slow it down when compared to a traversal based only on
relationship types. However, having said that, your graph is quite small,
and so the entire graph should fit in memory with the right cache settings,
and th
On Tue, Feb 22, 2011 at 4:26 PM, Marko Rodriguez wrote:
> Hey,
>
>> Thanks a lot. Updated to 0.8-SNAPSHOT. Strangely maven did not resolve
>> groovy dependence automatically. Had to add it by hand.
>
> Huh? Gremlin 0.8-SNAPSHOT uses Groovy 1.7.8 (which was released maybe 1 week
> ago). Gremlin 0.
Hi again,
Well well well, yep the cache did help, even if the queries were different ones
:) It's obviously not just caching the final list of paths, as I naively
assumed.
I got the requests down to about 150ms which looks much better. I suppose with
proper optimization I could even get better
Hi Peter,
and thank you for your time.
So I have 1100 nodes and 34616 relationships.
The test runs once and then shutdown, so yes I could try to ran it multiple
times before shutting down. However a particular request between nodes A and B
will probably not ran multiple times, so I am not sure
John,
how many nodes and relationships do you have in your graph? Are you
running only once without warming up the caches (you might want to run
the calculations say 10 times and take the mean)?
Also, it depends on how many nodes you are "touching" during your
traversals. The number of hops needed
Hello,
I am starting to evaluate neo4j and I wrote some basic test application. The
results I get are not as fast as I would have wished :) but of course it is
probably because it is not optimized or simply because my case do not fit that
well with graph DB.
If somebody would have few minutes
40 matches
Mail list logo
