codingjaguar / spark

Mirror of Apache Spark

Apache Spark

Spark is a fast and general cluster computing system for Big Data. It provides high-level APIs in Scala, Java, Python, and R, and an optimized engine that supports general computation graphs for data analysis. It also supports a rich set of higher-level tools including Spark SQL for SQL and DataFrames, MLlib for machine learning, GraphX for graph processing, and Spark Streaming for stream processing.

http://spark.apache.org/

Online Documentation

You can find the latest Spark documentation, including a programming guide, on the project web page and project wiki. This README file only contains basic setup instructions.

Building Spark

Spark is built using Apache Maven. To build Spark and its example programs, run:

build/mvn -DskipTests clean package

(You do not need to do this if you downloaded a pre-built package.) More detailed documentation is available from the project site, at "Building Spark". For developing Spark using an IDE, see Eclipse and IntelliJ.

Interactive Scala Shell

The easiest way to start using Spark is through the Scala shell:

./bin/spark-shell

Try the following command, which should return 1000:

scala> sc.parallelize(1 to 1000).count()

Interactive Python Shell

Alternatively, if you prefer Python, you can use the Python shell:

./bin/pyspark

And run the following command, which should also return 1000:

>>> sc.parallelize(range(1000)).count()

Example Programs

Spark also comes with several sample programs in the examples directory. To run one of them, use ./bin/run-example <class> [params]. For example:

./bin/run-example SparkPi

will run the Pi example locally.

You can set the MASTER environment variable when running examples to submit examples to a cluster. This can be a mesos:// or spark:// URL, "yarn" to run on YARN, and "local" to run locally with one thread, or "local[N]" to run locally with N threads. You can also use an abbreviated class name if the class is in the examples package. For instance:

MASTER=spark://host:7077 ./bin/run-example SparkPi

Many of the example programs print usage help if no params are given.

Running Tests

Testing first requires building Spark. Once Spark is built, tests can be run using:

./dev/run-tests

Please see the guidance on how to run tests for a module, or individual tests.

A Note About Hadoop Versions

Spark uses the Hadoop core library to talk to HDFS and other Hadoop-supported storage systems. Because the protocols have changed in different versions of Hadoop, you must build Spark against the same version that your cluster runs.

Please refer to the build documentation at "Specifying the Hadoop Version" for detailed guidance on building for a particular distribution of Hadoop, including building for particular Hive and Hive Thriftserver distributions.

Configuration

Please refer to the Configuration Guide in the online documentation for an overview on how to configure Spark.

spark's Issues

Dec 4 Cache Plan

Set level matching for Filter

Match the filter expression without effect of its orders

• Code changes
• Add test
• Style/Test cases check
• Jira
• Pul request
• Contact committer for auto test

final stage

1. Poster
2. Reporter
3. Baidu data evaluation
4. Synthetic data evaluation
5. Distributed data evaluation
6. Set of filter expression: bug fix
7. PR: set
8. Projection set
9. Filter merge
10. PR: projection + filter

Meeting note 10/18

1. generate a synthetic workload
2. initial implementation:
   1. how to index cache? Serialize the list of tuples [(column, predict), …]
   2. how to store the cache items? use the cache() provided by sparksql
   3. when executing a query, the cache planner asks logical planner for all the tables, columns and the predicates applied on them, then pass a list of key value pairs to cache manager. cache manager is responsible for inserting callbacks to spark so that the internal results will be materialized.
3. next implementation:
   1. cache the joined tables
4. Workflow analyze
   1. Use cache planner to analyze all the workflow at first
   2. Cache planner then start the sparkSQL plans sequentially.
5. RDD collector:
   1. Cache manager need to collect data, maybe from spark

phase 1 plan (Oct.18 - Oct 28)

Test preparation

Generate the test cases for future usage, and overcome difficulties on DevOps.

• Finish generate sample data
• Investigate on TPC-C and try to use test case on that
• Write sample iterative test cases
• benchmark on origin sparkSQL

Developing first prototype

• Read codes on SparkSQL, learn how to add Cache planter
• Read codes on Spark, learn how to collect data from Spark.
• Design the interface of CachePlaner and CacheManager.
• Code the prototype
• UnitTest
• Test cases on the benchmark

Yang's Plan:

• Learn the scala in one hour
• Finish the Test preparation steps

Jiang's Plan

• Read Spark & SparkSQL code

Cache replace strategy

1
For this dilemma:

1. if we cache table a, table b, table a.join.b
2. a and b are much less than the a.join.b, what cache replacement strategy is going to take?

2
For this dilemma:

1. cache table a.join.b, table a and table b are in disk
2. table a and table b can be filtered a lot from the predicate. a.join.b is very large
   Choose which?

A solution for both of these dilemma is a sophisticated cost estimator.

Proposal of the project

• Finish the draft version
• Discussion with professor on 9/30 12:40 - 13:00
• After meeting discussion
  • Time line
  • Steps
  • Working hours
• Finish the final version