Release Candidate Version: 0.10.0-incubating-rc1
Item | Status | Notes | |
---|---|---|---|
All Artifacts and Checksums Present | Fix | Not all artifacts have md5 checksums. See here. Also, the previous release has sha1 files in addition to asc and md5 files. | |
Release Candidate Build | Windows | ||
OS X | Pass | ||
Linux | |||
Test Suite Passes | Windows | ||
OS X | Pass | (DE will re-verify) | |
Linux | Pass | All daily tests from May 18 to May 24 have passed. See here. | |
All Binaries Execute | Pass | Verified on OS X. (DE will re-verify) | |
Check LICENSE and NOTICE Files | Pass | (DE will re-verify) | |
Src Artifact Builds and Tests Pass | |||
Single-Node Standalone | Windows | ||
OS X | Pass | ||
Linux | |||
Single-Node Spark | Pass | ||
Single-Node Hadoop | Pass | ||
Notebooks | Jupyter | ||
Zeppelin | |||
Performance Suite | Spark | Pass | Performance testsuite run on Spark 1.6.1 for data sizes {80MB, 800MB, 8GB, 80GB}, sparse/dense, intercept 0/1/2, and the algorithm classes binomial (Mlogreg, L2SVM, MSVM), multinomial (Mlogreg, MSVM, Naive Bayes), regression (LinregCG, LinregDS, GLM poisson-log, GLM gamma-log, GLM binomal-probit), clustering (Kmeans), and statistics (Univariate, Bivariate). The good news is that there are no compiler/runtime issues and performance is as expected. |
Hadoop |
Status Options | |||
---|---|---|---|
Pass | Fix | Fail |
Verify that each expected artifact is present at https://dist.apache.org/repos/dist/dev/incubator/systemml/ and that each artifact has accompanying checksums (such as .asc and .md5).
The release candidate should build on Windows, OS X, and Linux. To do this cleanly, the following procedure can be performed.
Clone the Apache SystemML GitHub repository to an empty location. Next, check out the release tag. Following this, build the distributions using Maven. This should be performed with an empty local Maven repository.
Here is an example:
$ git clone https://github.com/apache/incubator-systemml.git
$ cd incubator-systemml
$ git tag -l
$ git checkout tags/0.10.0-incubating-rc1 -b 0.10.0-incubating-rc1
$ mvn -Dmaven.repo.local=$HOME/.m2/temp-repo clean package -P distribution
The entire test suite should pass with no errors on Windows, OS X, and Linux. The test suite can be run using:
$ mvn clean verify
Validate that all of the binary artifacts can execute, including those artifacts packaged in other artifacts (in the tar.gz and zip artifacts). Here is an example of doing a basic sanity check on OS X.
# build distribution artifacts
mvn clean package -P distribution
cd target
# verify main jar works
java -cp ./lib/*:systemml-0.10.0-incubating.jar org.apache.sysml.api.DMLScript -s "print('hello world');"
# verify SystemML.jar works
java -cp ./lib/*:SystemML.jar org.apache.sysml.api.DMLScript -s "print('hello world');"
# verify standalone jar works
java -jar systemml-0.10.0-incubating-standalone.jar -s "print('hello world');"
# verify src works
tar -xvzf systemml-0.10.0-incubating-src.tar.gz
cd systemml-0.10.0-incubating-src
mvn clean package -P distribution
cd target/
java -cp ./lib/*:systemml-0.10.0-incubating.jar org.apache.sysml.api.DMLScript -s "print('hello world');"
java -cp ./lib/*:SystemML.jar org.apache.sysml.api.DMLScript -s "print('hello world');"
java -jar systemml-0.10.0-incubating-standalone.jar -s "print('hello world');"
cd ..
cd ..
# verify in-memory jar works
echo "import org.apache.sysml.api.jmlc.*;public class JMLCEx {public static void main(String[] args) throws Exception {Connection conn = new Connection();PreparedScript script = conn.prepareScript(\"print('hello world');\", new String[]{}, new String[]{}, false);script.executeScript();}}" > JMLCEx.java
javac -cp systemml-0.10.0-incubating-inmemory.jar JMLCEx.java
java -cp .:systemml-0.10.0-incubating-inmemory.jar JMLCEx
# verify standalone tar.gz works
tar -xvzf systemml-0.10.0-incubating-standalone.tar.gz
cd systemml-0.10.0-incubating-standalone
echo "print('hello world');" > hello.dml
./runStandaloneSystemML.sh hello.dml
cd ..
# verify distrib tar.gz works
tar -xvzf systemml-0.10.0-incubating.tar.gz
cd systemml-0.10.0-incubating
java -cp ../lib/*:SystemML.jar org.apache.sysml.api.DMLScript -s "print('hello world');"
# verify spark batch mode
export SPARK_HOME=/Users/deroneriksson/spark-1.5.1-bin-hadoop2.6
$SPARK_HOME/bin/spark-submit SystemML.jar -s "print('hello world');" -exec hybrid_spark
# verify hadoop batch mode
hadoop jar SystemML.jar -s "print('hello world');"
Each artifact must contain LICENSE and NOTICE files. These files must reflect the contents of the artifacts. If the project dependencies (ie, libraries) have changed since the last release, the LICENSE and NOTICE files must be updated to reflect these changes.
Each artifact should contain a DISCLAIMER file.
For more information, see:
- http://incubator.apache.org/guides/releasemanagement.html
- http://www.apache.org/dev/licensing-howto.html
The project should be built using the src
(tar.gz and zip) artifacts.
In addition, the test suite should be run using an src
artifact and
all tests should pass.
The standalone tar.gz and zip artifacts contain runStandaloneSystemML.sh
and runStandaloneSystemML.bat
files. Verify that one or more algorithms can be run on a single node using these
standalone distributions.
Here is an example based on the Quick Start Guide demonstrating the execution of an algorithm (on OS X).
$ tar -xvzf systemml-0.10.0-incubating-standalone.tar.gz
$ cd systemml-0.10.0-incubating-standalone
$ wget -P data/ http://archive.ics.uci.edu/ml/machine-learning-databases/haberman/haberman.data
$ echo '{"rows": 306, "cols": 4, "format": "csv"}' > data/haberman.data.mtd
$ echo '1,1,1,2' > data/types.csv
$ echo '{"rows": 1, "cols": 4, "format": "csv"}' > data/types.csv.mtd
$ ./runStandaloneSystemML.sh scripts/algorithms/Univar-Stats.dml -nvargs X=data/haberman.data TYPES=data/types.csv STATS=data/univarOut.mtx CONSOLE_OUTPUT=TRUE
Verify that SystemML runs algorithms on Spark locally.
Here is an example of running the Univar-Stats.dml
algorithm on random generated data.
$ tar -xvzf systemml-0.10.0-incubating.tar.gz
$ cd systemml-0.10.0-incubating
$ export SPARK_HOME=/Users/deroneriksson/spark-1.5.1-bin-hadoop2.6
$ $SPARK_HOME/bin/spark-submit SystemML.jar -f scripts/datagen/genRandData4Univariate.dml -exec hybrid_spark -args 1000000 100 10 1 2 3 4 uni.mtx
$ echo '1' > uni-types.csv
$ echo '{"rows": 1, "cols": 1, "format": "csv"}' > uni-types.csv.mtd
$ $SPARK_HOME/bin/spark-submit SystemML.jar -f scripts/algorithms/Univar-Stats.dml -exec hybrid_spark -nvargs X=uni.mtx TYPES=uni-types.csv STATS=uni-stats.txt CONSOLE_OUTPUT=TRUE
Verify that SystemML runs algorithms on Hadoop locally.
Based on the "Single-Node Spark" setup above, the Univar-Stats.dml
algorithm could be run as follows:
$ hadoop jar SystemML.jar -f scripts/algorithms/Univar-Stats.dml -nvargs X=uni.mtx TYPES=uni-types.csv STATS=uni-stats.txt CONSOLE_OUTPUT=TRUE
Verify that SystemML can be executed from Jupyter and Zeppelin notebooks. For examples, see the Spark MLContext Programming Guide.
Verify that the performance suite located at scripts/perftest/ executes on Spark and Hadoop. Testing should include 80MB, 800MB, 8GB, and 80GB data sizes.