After the installation of the PHP API you can use the simplest and most popular map/reduce API that Yothalot offers. You can use this API to write mapreduce jobs and send them to the Yothalot cluster. With this API you can also write jobs that process a lot of data without reducing it, so called race jobs, and send these to the Yothalot cluster.
If you want to use the PHP API you have to install besides Yothalot the Yothalot PHP extension.
To run a Yothalot mapreduce task you first have to write a class that implements the Yothalot\MapReduce interface. In this class you write your mapreduce algorithm. Besides implementing the mapreduce interface you have to set up a connection to Yothalot with the Yothalot\Connection class. In the connection you provide the credentials to set connect to Yothalot. With your mapreduce implementation and the connection you can create a Yothalot\Job. If you have this job you can run the job after having added the data on which the job should run. Optionally you can set some performance settings. That is it. You can wait for the job to be processed and retrieve some results about the behavior of the job, or you can detach your script and let Yothalot do its work while you do other work. If you want to process a lot of data and are not interested in the reduce step of the mapreduce algorithm you can implement a class using the Yothalot\Race interface instead of the Yothalot\MapReduce interface, with which you create a race job. The implementation of a race job is besides implementing a different interface similar to implementing a mapreduce job. Besides that it is also possible to run a single task on the Yothalot cluster. You do this by implementing the Yothalot\Task interface.
- Interface Yothalot\MapReduce
- Interface Yothalot\Race
- Interface Yothalot\Task
- Class Yothalot\Connection
- Class Yothalot\Job
- Class Yothalot\Pool
- Class Yothalot\Result
- Class Yothalot\RaceResult
- Class Yothalot\TaskResult
The classes and interface mentioned above are sufficient for writing and running mapreduce jobs. As an extra service however, the Yothalot API also comes with a couple of extra utility classes.
The Yothalot\Path class can be used for converting absolute pathnames into pathnames that are relative to the GlusterFS mount point, and the other way around. This is especially useful if you use different mount points on different servers.
The Yothalot\Input and Yothalot\Output classes allow you to read and write files in the same format used internally by the Yothalot framework for intermediate result files.
- Interface Yothalot\MapReduce
- Interface Yothalot\Race
- Interface Yothalot\Task
- Class Yothalot\Connection
- Class Yothalot\Input
- Class Yothalot\Job
- Class Yothalot\Pool
- Class Yothalot\Output
- Class Yothalot\Path
- Class Yothalot\RaceResult
- Class Yothalot\TaskResult
- Class Yothalot\Record
- Class Yothalot\Reducer
- Class Yothalot\Result
- Class Yothalot\Values
- Class Yothalot\Writer