We very much worked with them, even repoduced their dataset and setup and
pointed out the incorrect use and configuration of Neo4j as well as a
shortcoming of their data-model.

They happily accepted the information but didn't bother to update their
*original* article.

After our last post
<http://istc-bigdata.org/index.php/benchmarking-graph-databases/> about
benchmarking graph databases, Neo Technology
<http://www.neotechnology.com/> representatives
contacted us and said that they repeated the shortest path queries on Neo4j
<http://www.neotechnology.com/neo4j-graph-database/> with the Facebook
dataset and they observed 2-4 orders of magnitude better performance. This
was an intriguing comment and we were very interested in finding out why
this could be the case.


On Mon, Nov 3, 2014 at 7:25 PM, Ton Akveld <[email protected]> wrote:

> Could be wrong, but I read the article end "We highly appreciate the
> feedback from Neo Technology representatives!" as a wish, not a comment on.
> :-)
>
>
> On Monday, November 3, 2014 6:58:18 PM UTC+1, Mark Findlater wrote:
>>
>> Yes, I believe istc-bigdata really appreciated it.
>>
>> Article begins - "Neo Technology representatives contacted us and said..."
>>
>> Article ends - "We highly appreciate the feedback from Neo Technology
>> representatives!"
>>
>> I think there is still a wait for Kamilos to share further information..
>>
>> M
>>
>> On Monday, 3 November 2014 17:50:55 UTC, Ton Akveld wrote:
>>>
>>> Interesting to know if Neo Technology gave feedback as requested: *We
>>> highly appreciate the feedback from Neo Technology representatives!*
>>> From: http://istc-bigdata.org/index.php/benchmarking-graph-
>>> databases-updates/
>>>
>>> On Monday, November 3, 2014 4:43:31 PM UTC+1, Ton Akveld wrote:
>>>>
>>>> Hi group,
>>>>
>>>> I just saw this question, and neo4j alleged performance worries me too.
>>>> Especially when remarks like  this: <Neo4j's main bottleneck is the
>>>> memory it consumes. An OSM file up to one
>>>> gigabyte seems to be the limit for Neo4j; importing larger datasets
>>>> takes a long time and queries become
>>>> slow.>
>>>> What worries me even more is the absence of reactions of Neo4j's TLM.
>>>> How to convince the world of positive use of Neo4j's graph database
>>>> when these 'rumors'  are not disproved?
>>>>
>>>> Kind regards,
>>>>
>>>> Ton
>>>>
>>>>
>>>> On Friday, October 31, 2014 9:10:10 PM UTC+1, gg4u wrote:
>>>>>
>>>>> hi,
>>>>> also interested in this.
>>>>> i d love to benchmark similar queries against NoSql (dynamo on aws)
>>>>> could you please share a foo table publicly?
>>>>> my model on nosql is
>>>>> node A as index, all neighbors of A as value (a whole string).
>>>>
>>>>  --
> You received this message because you are subscribed to the Google Groups
> "Neo4j" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Neo4j" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to