Hi Marc, Just to be sure: the indexers itself are not limited in the number of CPU they can grab? The 60 indexers run on the same machine? Or in independent cloud containers? If the indexers are not

Hi Boxuan, I can definitively try from the 0.6.0 pre release or master version, that's a good idea, I'll come back with the result, Thanks! Cheers, Marc

Hi, I didn't check your metrics but my first impression was this might be related to the internal thread pool. Can you try out the 0.6.0 pre-release version or master version? Remember to set

Hi all, We've been trying to scale horizontally Janusgraph to keep up with the data throughput, but we're reaching some limit in scaling. We've been trying different things to pinpoint the bottleneck

FWIW I've tried exporting the graph in the example to JSON (GraphSON) and the metaproperty *is* preserved, however when I import the same graph from the json file the metaproperty is not created.

Thank you for this example. After running this, I can see that the property "metatest" has been ignored and is missing completely from the GraphML output. Another issue that I have with GraphML is

Hi Laura, No. As the TinkerPop docs say: "graphML is a lossy format". You can try for yourself with:Best wishes, Marc

Hi there, Recently I am trying to measure Janusgraph performance. When I was trying to run benchmark using Janusgraph 0.5.3, I found too low performance when testing pagerank and wcc. The code I used

HI Marc, I have tried it. And finally I got complete Graph500 vertices and edges loaded. But there is a still weird thing that I found the same exception reported in the log. Could you please explain

Hi Boxuan, Thank you for the response. I tried by forceClosing all management instances except the current one before setting "tx.log-tx". Management is getting updated with latest value, this is not

Hi Mark, Thanks for the suggestions. Narrowed down the issue to JanusGraph's support for Azul Prime JDK 8, which generates nanosecond precision for Instant.now() API, as compared to the millisecond

On Tue, Aug 24, 2021 at 06:20 AM, <hadoopmarc@...> wrote: withGot it. I will try it soon. Thanks, Marc! Shipeng

Hi Shipeng Qi, The system that you use might be too small for the number of threads in the loading code. You can try to decrease the number of threads from 8 to 4 with: private static ExecutorService

I suspect this is due to stale management instances. Check out https://developer.ibm.com/articles/janusgraph-tips-and-tricks-pt-2/#troubleshooting-indexes and see if it helps.

Hi there, I am new to Janusgraph. I have some problems in loading data to Janusgraph with Cassandra CQL as storage backend. When I tried to load Graph500 to Janusgraph, planning to run benchmark on

Hi Boxuan,For existing JanusGraph setup, I am updating tx.log-tx configuration by setting management system property as mentioned in

Thank you, this is exactly what I look for.

Hi Toom, Do you want to ALWAYS make sure the vertex is indexed? If so and if you happen to use Elasticsearch, you can set See

Hello, The vertex that has just been created are not immediately available on mixed index (documented here [1]). I'm looking for a way to make sure the vertex is indexed, by waiting the mixed index

Janus supports "properties of properties", ie. properties defined on other properties. How is this represented with graphml? Should I use nested elements like this <node> <data