pirk-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From jacobwil...@apache.org
Subject [1/2] incubator-pirk git commit: Initial instructions for IBM Bluemix.
Date Tue, 18 Oct 2016 17:37:50 GMT
Repository: incubator-pirk
Updated Branches:
  refs/heads/gh-pages fc19ff322 -> 6e50c8308

Initial instructions for IBM Bluemix.

Project: http://git-wip-us.apache.org/repos/asf/incubator-pirk/repo
Commit: http://git-wip-us.apache.org/repos/asf/incubator-pirk/commit/9d531ca2
Tree: http://git-wip-us.apache.org/repos/asf/incubator-pirk/tree/9d531ca2
Diff: http://git-wip-us.apache.org/repos/asf/incubator-pirk/diff/9d531ca2

Branch: refs/heads/gh-pages
Commit: 9d531ca28dfdc3495cecd20cd731c3f1e18a20b1
Parents: fc19ff3
Author: Tim Ellison <t.p.ellison@gmail.com>
Authored: Thu Oct 13 14:10:45 2016 +0100
Committer: Tim Ellison <t.p.ellison@gmail.com>
Committed: Thu Oct 13 14:10:45 2016 +0100

 cloud_instructions.md | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/cloud_instructions.md b/cloud_instructions.md
index 00340d4..df00966 100644
--- a/cloud_instructions.md
+++ b/cloud_instructions.md
@@ -140,4 +140,27 @@ If you want to SSH in and set up a SOCKS proxy to access the [web interfaces](ht
 11. Now on the cluster, you can run the distributed tests:  
 `hadoop jar apache-pirk-0.2.0-incubating-SNAPSHOT-exe.jar org.apache.pirk.test.distributed.DistributedTestDriver
-j $PWD/apache-pirk-0.2.0-incubating-SNAPSHOT-exe.jar -t 1:J`
 12. When you are done working with your cluster, terminate it:  
-`aws emr terminate-clusters --cluster-ids `**`$cid`**
\ No newline at end of file
+`aws emr terminate-clusters --cluster-ids `**`$cid`**
+## IBM Bluemix
+1 .[Sign-up](https://console.ng.bluemix.net/registration/) for a free Bluemix account.
+2. From the Bluemix [catalog](https://console.ng.bluemix.net/catalog/) open the "Big Insights
for Apache Hadoop" service, found in the Data and Analytics section, and click "Create". 
The basic service plan is free during beta, but will need to be recreated every two weeks.
+3. Click "Open" to see the cluster list, and from there create a new cluster,
+  cluster name: test-cluster
+  user name: pirk
+  password: <password>.
+In Configuration:
+  Increase the number of data nodes to 5 (the maximum number available on the basic plan)
+  Scroll down and select "Spark" as an optional component.
+Click "Create".
+4. Select the test cluster from the Cluster List and take note of the SSH host name, e.g.
+5. Now you can run the distributed tests by copying the Pirk jar file and executing it in
Bluemix., i.e.
+`$ scp target/apache-pirk-0.2.0-incubating-SNAPSHOT-exe.jar pirk@bi-hadoop-prod-4174.bi.services.us-south.bluemix.net:
+pirk@bi-hadoop-prod-4174.bi.services.us-south.bluemix.net's password: 
           100%  145MB  10.3MB/s   00:14 
+$ ssh pirk@bi-hadoop-prod-4174.bi.services.us-south.bluemix.net
+pirk@bi-hadoop-prod-4174.bi.services.us-south.bluemix.net's password: 
+-bash-4.1$ hadoop jar apache-pirk-0.2.0-incubating-SNAPSHOT-exe.jar org.apache.pirk.test.distributed.DistributedTestDriver
-j apache-pirk-0.2.0-incubating-SNAPSHOT-exe.jar`

View raw message