How to launch a Helium Application in local deploy mode

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

How to launch a Helium Application in local deploy mode

fish fish
Hi Group,

I have written a Helium application and deploy in local mode. Also I enable it in 'Helium' page. However, I don't know how to launch the enabled application in notebook. I tried type "%helium run" as showed in this video: https://cwiki.apache.org/confluence/display/ZEPPELIN/Helium+proposal , but error occurs which indicate no helium interpreter. Could anyone kindly tell me how to launch Helium application in right way? Thanks!

Zeppelin version: 0.8.0-snapshot 

application.json:

{
"type" : "APPLICATION",
"name" : "test_app",
"description" : "test Helium App",
"license" : "Apache-2.0",
"artifact" : "./examples/zeppelin-mytest-0.0.1-SNAPSHOT.jar",
"className" : "com.test.Test",
"resources" : [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]],
"icon" : "<i class='fa fa-search-minus'></i>"
}
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: How to launch a Helium Application in local deploy mode

moon
Administrator
Hi,

Dev mode in actual implementation became '%dev run' instead of '%helium run' in proposal.

To use '%dev run', you need build Zeppelin with -Phelium-dev flag.

Hope this helps!

Best,
moon

On Fri, Apr 14, 2017 at 8:16 PM fish fish <[hidden email]> wrote:
Hi Group,

I have written a Helium application and deploy in local mode. Also I enable it in 'Helium' page. However, I don't know how to launch the enabled application in notebook. I tried type "%helium run" as showed in this video: https://cwiki.apache.org/confluence/display/ZEPPELIN/Helium+proposal , but error occurs which indicate no helium interpreter. Could anyone kindly tell me how to launch Helium application in right way? Thanks!

Zeppelin version: 0.8.0-snapshot 

application.json:

{
"type" : "APPLICATION",
"name" : "test_app",
"description" : "test Helium App",
"license" : "Apache-2.0",
"artifact" : "./examples/zeppelin-mytest-0.0.1-SNAPSHOT.jar",
"className" : "com.test.Test",
"resources" : [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]],
"icon" : "<i class='fa fa-search-minus'></i>"
}
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: How to launch a Helium Application in local deploy mode

fish fish
Hi Lee,

Thank you for your reply. Actually I already tried dev mode and succeeded. My question is how to launch Helium application not in development mode but in deploy mode like 'SPELL' described in https://zeppelin.apache.org/docs/snapshot/development/writingzeppelinspell.html#1-enabling. Could you please kindly guide more? Thanks!

Best,
Chen 

2017-04-15 17:13 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Dev mode in actual implementation became '%dev run' instead of '%helium run' in proposal.

To use '%dev run', you need build Zeppelin with -Phelium-dev flag.

Hope this helps!

Best,
moon

On Fri, Apr 14, 2017 at 8:16 PM fish fish <[hidden email]> wrote:
Hi Group,

I have written a Helium application and deploy in local mode. Also I enable it in 'Helium' page. However, I don't know how to launch the enabled application in notebook. I tried type "%helium run" as showed in this video: https://cwiki.apache.org/confluence/display/ZEPPELIN/Helium+proposal , but error occurs which indicate no helium interpreter. Could anyone kindly tell me how to launch Helium application in right way? Thanks!

Zeppelin version: 0.8.0-snapshot 

application.json:

{
"type" : "APPLICATION",
"name" : "test_app",
"description" : "test Helium App",
"license" : "Apache-2.0",
"artifact" : "./examples/zeppelin-mytest-0.0.1-SNAPSHOT.jar",
"className" : "com.test.Test",
"resources" : [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]],
"icon" : "<i class='fa fa-search-minus'></i>"
}

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: How to launch a Helium Application in local deploy mode

moon
Administrator
Hi,

Helium application launch will appear when ResourcePool has required resource [1] for an application.

Spark interpreter put last evaluated object in the ResourcePool.
For example, if you run "new java.util.Date()" in a paragraph, a date object will be created in the ResourcePool. And there will be a button displayed in the paragraph and list all applications that consumes java.util.Date type object.
 
if you build Zeppelin with -Pexamples flag, then you can run "new java.util.Date()" in spark interpreter and run example application.

Let me know if it helps.

Thanks,
moon




On Mon, Apr 17, 2017 at 12:55 PM fish fish <[hidden email]> wrote:
Hi Lee,

Thank you for your reply. Actually I already tried dev mode and succeeded. My question is how to launch Helium application not in development mode but in deploy mode like 'SPELL' described in https://zeppelin.apache.org/docs/snapshot/development/writingzeppelinspell.html#1-enabling. Could you please kindly guide more? Thanks!

Best,
Chen 

2017-04-15 17:13 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Dev mode in actual implementation became '%dev run' instead of '%helium run' in proposal.

To use '%dev run', you need build Zeppelin with -Phelium-dev flag.

Hope this helps!

Best,
moon

On Fri, Apr 14, 2017 at 8:16 PM fish fish <[hidden email]> wrote:
Hi Group,

I have written a Helium application and deploy in local mode. Also I enable it in 'Helium' page. However, I don't know how to launch the enabled application in notebook. I tried type "%helium run" as showed in this video: https://cwiki.apache.org/confluence/display/ZEPPELIN/Helium+proposal , but error occurs which indicate no helium interpreter. Could anyone kindly tell me how to launch Helium application in right way? Thanks!

Zeppelin version: 0.8.0-snapshot 

application.json:

{
"type" : "APPLICATION",
"name" : "test_app",
"description" : "test Helium App",
"license" : "Apache-2.0",
"artifact" : "./examples/zeppelin-mytest-0.0.1-SNAPSHOT.jar",
"className" : "com.test.Test",
"resources" : [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]],
"icon" : "<i class='fa fa-search-minus'></i>"
}

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: How to launch a Helium Application in local deploy mode

fish fish
Thank you Lee! I can run the Clock example. However, in https://github.com/apache/zeppelin/blob/master/spark/src/main/java/org/apache/zeppelin/spark/SparkInterpreter.java#L1287 , SparkInterpreter only put last evaluated variable into resource pool. So if my helium app need multiple resources like [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]], how should I run in one paragraph? I run following code in one paragraph but seems only the last sqlContext be out in resource pool. 
%spark
sc
spark
sqlContext

Best,
Chen

2017-04-17 13:27 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Helium application launch will appear when ResourcePool has required resource [1] for an application.

Spark interpreter put last evaluated object in the ResourcePool.
For example, if you run "new java.util.Date()" in a paragraph, a date object will be created in the ResourcePool. And there will be a button displayed in the paragraph and list all applications that consumes java.util.Date type object.
 
if you build Zeppelin with -Pexamples flag, then you can run "new java.util.Date()" in spark interpreter and run example application.

Let me know if it helps.

Thanks,
moon




On Mon, Apr 17, 2017 at 12:55 PM fish fish <[hidden email]> wrote:
Hi Lee,

Thank you for your reply. Actually I already tried dev mode and succeeded. My question is how to launch Helium application not in development mode but in deploy mode like 'SPELL' described in https://zeppelin.apache.org/docs/snapshot/development/writingzeppelinspell.html#1-enabling. Could you please kindly guide more? Thanks!

Best,
Chen 

2017-04-15 17:13 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Dev mode in actual implementation became '%dev run' instead of '%helium run' in proposal.

To use '%dev run', you need build Zeppelin with -Phelium-dev flag.

Hope this helps!

Best,
moon

On Fri, Apr 14, 2017 at 8:16 PM fish fish <[hidden email]> wrote:
Hi Group,

I have written a Helium application and deploy in local mode. Also I enable it in 'Helium' page. However, I don't know how to launch the enabled application in notebook. I tried type "%helium run" as showed in this video: https://cwiki.apache.org/confluence/display/ZEPPELIN/Helium+proposal , but error occurs which indicate no helium interpreter. Could anyone kindly tell me how to launch Helium application in right way? Thanks!

Zeppelin version: 0.8.0-snapshot 

application.json:

{
"type" : "APPLICATION",
"name" : "test_app",
"description" : "test Helium App",
"license" : "Apache-2.0",
"artifact" : "./examples/zeppelin-mytest-0.0.1-SNAPSHOT.jar",
"className" : "com.test.Test",
"resources" : [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]],
"icon" : "<i class='fa fa-search-minus'></i>"
}


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: How to launch a Helium Application in local deploy mode

fish fish
Hi

Just wonder any updates on this thread? Thanks!

Best,
Chen

2017-04-18 16:10 GMT+08:00 fish fish <[hidden email]>:
Thank you Lee! I can run the Clock example. However, in https://github.com/apache/zeppelin/blob/master/spark/src/main/java/org/apache/zeppelin/spark/SparkInterpreter.java#L1287 , SparkInterpreter only put last evaluated variable into resource pool. So if my helium app need multiple resources like [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]], how should I run in one paragraph? I run following code in one paragraph but seems only the last sqlContext be out in resource pool. 
%spark
sc
spark
sqlContext

Best,
Chen

2017-04-17 13:27 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Helium application launch will appear when ResourcePool has required resource [1] for an application.

Spark interpreter put last evaluated object in the ResourcePool.
For example, if you run "new java.util.Date()" in a paragraph, a date object will be created in the ResourcePool. And there will be a button displayed in the paragraph and list all applications that consumes java.util.Date type object.
 
if you build Zeppelin with -Pexamples flag, then you can run "new java.util.Date()" in spark interpreter and run example application.

Let me know if it helps.

Thanks,
moon




On Mon, Apr 17, 2017 at 12:55 PM fish fish <[hidden email]> wrote:
Hi Lee,

Thank you for your reply. Actually I already tried dev mode and succeeded. My question is how to launch Helium application not in development mode but in deploy mode like 'SPELL' described in https://zeppelin.apache.org/docs/snapshot/development/writingzeppelinspell.html#1-enabling. Could you please kindly guide more? Thanks!

Best,
Chen 

2017-04-15 17:13 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Dev mode in actual implementation became '%dev run' instead of '%helium run' in proposal.

To use '%dev run', you need build Zeppelin with -Phelium-dev flag.

Hope this helps!

Best,
moon

On Fri, Apr 14, 2017 at 8:16 PM fish fish <[hidden email]> wrote:
Hi Group,

I have written a Helium application and deploy in local mode. Also I enable it in 'Helium' page. However, I don't know how to launch the enabled application in notebook. I tried type "%helium run" as showed in this video: https://cwiki.apache.org/confluence/display/ZEPPELIN/Helium+proposal , but error occurs which indicate no helium interpreter. Could anyone kindly tell me how to launch Helium application in right way? Thanks!

Zeppelin version: 0.8.0-snapshot 

application.json:

{
"type" : "APPLICATION",
"name" : "test_app",
"description" : "test Helium App",
"license" : "Apache-2.0",
"artifact" : "./examples/zeppelin-mytest-0.0.1-SNAPSHOT.jar",
"className" : "com.test.Test",
"resources" : [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]],
"icon" : "<i class='fa fa-search-minus'></i>"
}



Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: How to launch a Helium Application in local deploy mode

moon
Administrator
Hi,

You can get reference to ResourcePool by running

val resourcePool = z.getInterpreterContext().getResourcePool() // [1]

Then you can invoke methods to put resource [2] into it.
Hope this works.

Thanks,
moon





On Tue, Apr 18, 2017 at 8:11 PM fish fish <[hidden email]> wrote:
Hi

Just wonder any updates on this thread? Thanks!

Best,
Chen

2017-04-18 16:10 GMT+08:00 fish fish <[hidden email]>:
Thank you Lee! I can run the Clock example. However, in https://github.com/apache/zeppelin/blob/master/spark/src/main/java/org/apache/zeppelin/spark/SparkInterpreter.java#L1287 , SparkInterpreter only put last evaluated variable into resource pool. So if my helium app need multiple resources like [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]], how should I run in one paragraph? I run following code in one paragraph but seems only the last sqlContext be out in resource pool. 
%spark
sc
spark
sqlContext

Best,
Chen

2017-04-17 13:27 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Helium application launch will appear when ResourcePool has required resource [1] for an application.

Spark interpreter put last evaluated object in the ResourcePool.
For example, if you run "new java.util.Date()" in a paragraph, a date object will be created in the ResourcePool. And there will be a button displayed in the paragraph and list all applications that consumes java.util.Date type object.
 
if you build Zeppelin with -Pexamples flag, then you can run "new java.util.Date()" in spark interpreter and run example application.

Let me know if it helps.

Thanks,
moon




On Mon, Apr 17, 2017 at 12:55 PM fish fish <[hidden email]> wrote:
Hi Lee,

Thank you for your reply. Actually I already tried dev mode and succeeded. My question is how to launch Helium application not in development mode but in deploy mode like 'SPELL' described in https://zeppelin.apache.org/docs/snapshot/development/writingzeppelinspell.html#1-enabling. Could you please kindly guide more? Thanks!

Best,
Chen 

2017-04-15 17:13 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Dev mode in actual implementation became '%dev run' instead of '%helium run' in proposal.

To use '%dev run', you need build Zeppelin with -Phelium-dev flag.

Hope this helps!

Best,
moon

On Fri, Apr 14, 2017 at 8:16 PM fish fish <[hidden email]> wrote:
Hi Group,

I have written a Helium application and deploy in local mode. Also I enable it in 'Helium' page. However, I don't know how to launch the enabled application in notebook. I tried type "%helium run" as showed in this video: https://cwiki.apache.org/confluence/display/ZEPPELIN/Helium+proposal , but error occurs which indicate no helium interpreter. Could anyone kindly tell me how to launch Helium application in right way? Thanks!

Zeppelin version: 0.8.0-snapshot 

application.json:

{
"type" : "APPLICATION",
"name" : "test_app",
"description" : "test Helium App",
"license" : "Apache-2.0",
"artifact" : "./examples/zeppelin-mytest-0.0.1-SNAPSHOT.jar",
"className" : "com.test.Test",
"resources" : [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]],
"icon" : "<i class='fa fa-search-minus'></i>"
}



Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: How to launch a Helium Application in local deploy mode

fish fish
Thank you Lee! 

2017-04-19 11:55 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

You can get reference to ResourcePool by running

val resourcePool = z.getInterpreterContext().getResourcePool() // [1]

Then you can invoke methods to put resource [2] into it.
Hope this works.

Thanks,
moon





On Tue, Apr 18, 2017 at 8:11 PM fish fish <[hidden email]> wrote:
Hi

Just wonder any updates on this thread? Thanks!

Best,
Chen

2017-04-18 16:10 GMT+08:00 fish fish <[hidden email]>:
Thank you Lee! I can run the Clock example. However, in https://github.com/apache/zeppelin/blob/master/spark/src/main/java/org/apache/zeppelin/spark/SparkInterpreter.java#L1287 , SparkInterpreter only put last evaluated variable into resource pool. So if my helium app need multiple resources like [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]], how should I run in one paragraph? I run following code in one paragraph but seems only the last sqlContext be out in resource pool. 
%spark
sc
spark
sqlContext

Best,
Chen

2017-04-17 13:27 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Helium application launch will appear when ResourcePool has required resource [1] for an application.

Spark interpreter put last evaluated object in the ResourcePool.
For example, if you run "new java.util.Date()" in a paragraph, a date object will be created in the ResourcePool. And there will be a button displayed in the paragraph and list all applications that consumes java.util.Date type object.
 
if you build Zeppelin with -Pexamples flag, then you can run "new java.util.Date()" in spark interpreter and run example application.

Let me know if it helps.

Thanks,
moon




On Mon, Apr 17, 2017 at 12:55 PM fish fish <[hidden email]> wrote:
Hi Lee,

Thank you for your reply. Actually I already tried dev mode and succeeded. My question is how to launch Helium application not in development mode but in deploy mode like 'SPELL' described in https://zeppelin.apache.org/docs/snapshot/development/writingzeppelinspell.html#1-enabling. Could you please kindly guide more? Thanks!

Best,
Chen 

2017-04-15 17:13 GMT+08:00 moon soo Lee <[hidden email]>:
Hi,

Dev mode in actual implementation became '%dev run' instead of '%helium run' in proposal.

To use '%dev run', you need build Zeppelin with -Phelium-dev flag.

Hope this helps!

Best,
moon

On Fri, Apr 14, 2017 at 8:16 PM fish fish <[hidden email]> wrote:
Hi Group,

I have written a Helium application and deploy in local mode. Also I enable it in 'Helium' page. However, I don't know how to launch the enabled application in notebook. I tried type "%helium run" as showed in this video: https://cwiki.apache.org/confluence/display/ZEPPELIN/Helium+proposal , but error occurs which indicate no helium interpreter. Could anyone kindly tell me how to launch Helium application in right way? Thanks!

Zeppelin version: 0.8.0-snapshot 

application.json:

{
"type" : "APPLICATION",
"name" : "test_app",
"description" : "test Helium App",
"license" : "Apache-2.0",
"artifact" : "./examples/zeppelin-mytest-0.0.1-SNAPSHOT.jar",
"className" : "com.test.Test",
"resources" : [[":org.apache.spark.sql.SparkSession", ":org.apache.spark.SparkContext", ":org.apache.spark.sql.SQLContext"]],
"icon" : "<i class='fa fa-search-minus'></i>"
}




Loading...