Hello,
It's a good idea, machine learning in Apache Ignite is actively developing
right now and new features are always appreciated.
At the same time, let me point that deep learning is already supported in
Apache Ignite via integration with TensorFlow as the most popular deep
learning framework
It has been observed that:
1. No server nodes up, and started client node, it waits for server nodes to
start working.
2. Serer nodes are up, and started client node, it connect to Server nodes
and start working.
3. No server nodes up, but visor is up, and if we started client node, it
connect to v
We consume data from Kafka, process (look up on various cache & some business
logic) consumed data and add it to the IgniteDataStreamer. Which one should
be used to consume data from Kafka in Ignite application? And what is
difference between Ignite Kafka Streamer and Kafka consumer (2.0)?
--
Se
I want all nodes in a cluster to have equal number of cache partitions. With
default Affinity function it is not happening, why default does not use all
nodes in a cluster? Will it be okay to write custom Affinity function? And
what will we lose doing so? Did anyone write custom Affinity function?
i have an idea as a proposal for my thesis. in a nutshell i want to develop a
framework on top of Apache ignite for processing deep learning application.
as an engine on top of Apache ignite i have considered DL4J ( deep learning
for java). i send this message for you about express your opinion on
We can get the total number of gets/puts & average time for gets/puts on
Caches. But in my case I do not use get/put on cache, I use
IgniteDataStreamer to load data in cache. I need to know time taken by
IgniteDataStreamer.
Thanks
--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Hello,
After we upgrade to the 2.7 we meet a wired warn; basically all our ignite
cache running in LOCAL model in a internal network.
All the configuration are almost default. but we meet a ERROR logger of the
tcp-disco-msg-worker* but after that the the cluster still working, no crash
hap
Now the cluster have 100+ nodes, when 'Start check connection process'
happens,
Some node will throw oom with Direct buffer memory (java nio).
When check connections,Many nio socker will be create ,Then oom happens?
How to fix the oom except larger xmx?
Thanks.
--
Sent from: http://apache-igni
Hello,
I am currently trying to integrate Ignite 2.7 into an existing Web
Application built on top of Apache Tomcat. Therefore I use it in an embedded
mode.
While everything works fine so far, I came across an issue regarding the
idle CPU consumption. On my developing machine (Laptop, Inteil I7
unsubscribe
Thanks for the reply.
Everything related to the cache we understand the current architecture, in
our code base we'll probably treat everything cache related the same as
schema migrations in a DB (migration scripts, etc.).
Our real issue is with Ignite Runnables conflicting on the Binary
Marshaller
Hello! Thanks for your reply!
Unfortunately,the above query is a sub query of a big query.So I think I
can't overcome it.How about integration with spark?Will spark deal with such
queries in a smart way?
ilya.kasnacheev wrote
> Hello!
>
> The recommendation here is to use Lazy SQL Queries.
> h
Hello!
1. I guess you can do both. For long-lived processes the former is
preferred to avoid surprises.
2. No.
3. Yes, data is always converted on-heap before given away to any client
code.
Regards,
--
Ilya Kasnacheev
пн, 17 дек. 2018 г. в 11:19, Isaeed Mohanna :
> Hi
> 1. when u say that I s
Hello!
As far as my understanding goes:
You can't peer class load your Key/Value types.
You also can't redeploy your Key/Value types.
They even survive node restart via WORKDIR/marshaller directory, and come
back to haunt you.
There are plans to maybe ease some of those limitations in 3.0, but no
Hello!
I'm not completely sure but persistent REPLICATED cache is the same as
PARTITIONED with MAXINT backups.
It means that every node will have a copy of data, but it has to be in BLT
to be used.
Regards,
--
Ilya Kasnacheev
пн, 17 дек. 2018 г. в 14:46, kimec.ethome.sk :
> Hi all,
>
> Could
Hello!
You could be running Tasks from your service, and you can deploy Tasks via
Deployment SPI, e.g. UriDeploymentSpi.
https://apacheignite.readme.io/docs/deployment-spi
Here it will check if new version is available, redeploy it and start using
new JARs.
Regards,
--
Ilya Kasnacheev
пн, 17
Hello!
I understand your concern but REST does not support such use case out of
box.
You could try implementing your own *ConnectorMessageInterceptor* to read
additional params from request, put them to thread-local, get them from
thread-local in constructor.
But I think you will be served better
Hello!
PUBLIC.EVENTTHETA_PARENTS2CELLID_ASC_S2CELLID_ASC_EVENTDATE_ASC_EVENTHOUR_ASC_IDX
Looks like this index is not selective enough for efficient filtering by
EVENTDATE. So you have to scan a lot of entries regardless of date range.
You can set local=true on SQL query to only return results fr
Hello!
The recommendation here is to use Lazy SQL Queries.
https://apacheignite-sql.readme.io/docs/performance-and-debugging#section-result-set-lazy-loading
Then, be careful to process data in small batches and avoid holding on to
large subsets of result set.
Regards,
--
Ilya Kasnacheev
пн, 17
Hello.
I do a simple sql join operation on a ignite node which is version 2.6 and
has 22GB heap memory and 22 GB off-heap memory.I also enable persistence.The
query exhausted ignite's heap memory quickly.How can I overcome this
situation.
Here is my query:
explain
SELECT
t2.*,
t1.
Zaheer,
Currently services are loaded using that class loader, that was used to
load Ignite.
So, you need to restart a node to add something to its classpath.
It's planned to add a hot redeployment feature, that will allow providing
and changing
service implementation in runtime.
Here is the desi
The issue is still present in 2.7.
I added a ticket on Jira with sample code that reproduces the issue.
https://issues.apache.org/jira/browse/IGNITE-10717
For now I think we can work around the issue by overriding the default
BinaryNameMapper, but this feels rather hacky to me.
On Mon, Dec 17, 2
Hi,
I have a use case, where a node is already started and I want to copy some
jars into that node's libs folder.
Visor has the deploy command. But without using visor can we do it somehow?
I know peer class loading happens in data and compute. But for services the
documentation says no peer cla
Hello Dennis,
Thank you for the response.
Yes, the Objects need to have same schema.
With regards to Converting to BinaryObjects, please let me know whether
Nested BinaryObject fields more than one level is supported. My
understanding is NO.
My requirement is,
1. Write Simple and Nested JSON in
Hi all,
Could somebody confirm my conclusion below?
It seems it is possible to declare a REPLICATED cache configuration for
caches that are mapped to a data region backed by the native persistence
layer.
Ignite does not complain about this configuration and boots happily.
Yet, after cluster r
Hareesh,
Are the object supposed to have the same schema?
If yes, then JSONs may be converted to BinaryObjects.
If no, then you'll have to store them as blobs or text.
If you want to access such data from SQL, your tables
should have a flat structure. You may have nested binary
objects, but in SQ
Gert,
Could you check if the problem with a deployment mode reproduces on Ignite
2.7?
If it does, please file a ticket with an explanation and a reproducer to
https://issues.apache.org/jira/
Thanks!
Denis
пн, 17 дек. 2018 г. в 12:12, Gert Dubois :
> I investigated the issue further and narrowe
I’m not sure there are any — and you’re right, there probably should.
Having said that, integration is very straight-forward.
You run pyspark (or spark-submit), passing in the Ignite jar files (using the
—jars parameter). For example:
$SPARK_HOME/bin/spark-submit --jars
$IGNITE_HOME/libs/ignit
Hello,
Note that in Ignite 2.7 only OpenSSL 1.0 is supported.
Now, for your issue. It's weird and I have never faced that one by myself,
but according to MSDN [1] you need to link to shell32.dll.
[1] -
https://docs.microsoft.com/en-us/windows/desktop/api/shellapi/nf-shellapi-shfileoperationa
Be
I investigated the issue further and narrowed the issue down to the Binary
Marshaller not working as expected given the configured Deployment Mode.
When forcing my clients to use unique class names in the metadata of the
Binary Marshaller (I forced this by overriding the global BinaryNameMapper
and
Hi
1. when u say that I should have a constant amount allocated for heap do u
mean initially allocated min:4GB, Max 4GB or having a settings of min:1GB
, Max: 4GB is allright as well?
2. If i do not configure any on-heap caches does ignite automatically
allocate on-heap caches frequent requests or
31 matches
Mail list logo