Iced Cakes Pipeline

Motivation

Pipelines can increase throughput when processing a stream of data. We will gain additional experience with Phasers by building a software pipeline.

Code To Use

Mixer

class Mixer

mix(int cakeIndex)

Oven

class Oven

bake(int cakeIndex, MixedIngredients mixedIngredients)

Icer

class Icer

ice(int cakeIndex, BakedCake bakedCake)

Looping

Our reference page's parallel loop section

Java For Loop

Phasers

Our reference page's phasers section

class Phaser (Guide to the Java Phaser)

register

arrive

awaitAdvance use via PhaserUtils.awaitAdvanceForPhase(phaser, phase)

PhaserUtils

awaitAdvanceForPhase

If you are on or past the phase you want to await, then calling phaser.awaitAdvance() directly is fine. If you might be on a prior phase, then invoke PhaserUtils.awaitAdvanceForPhase(phaser, phase). If in doubt, invoking awaitAdvanceForPhase is safer.

	public static int awaitAdvanceForPhase(Phaser phaser, int phase) {
		return awaitAdvanceForPhase(phaser, phase, () -> Thread.yield());
	}

	public static int awaitAdvanceForPhase(Phaser phaser, int phase, Runnable runnable) {
		while (true) {
			int currentPhase = phaser.awaitAdvance(phase);
			if (currentPhase < 0 || currentPhase > phase) {
				return currentPhase;
			} else {
				if (runnable != null) {
					runnable.run();
				}
			}
		}
	}

The Core Questions

• What are the tasks?
• What is the data?
• Is the data mutable?
• If so, how is it shared?

Here are some other questions that will help lead you down the right road.

• What work does each task need to do?
• What, if anything, does each task depend upon? That is: what does each task have to wait for before it may proceed?

Code To Implement

Required Code

class:	CakePipeline.java
methods:	mixBakeAndIceCakes
package:	pipeline.cake.studio
source folder:	student/src/main/java

method: public static IcedCake[] mixBakeAndIceCakes(Mixer mixer, Oven oven, Icer icer, int cakeCount) (parallel implementation required)

Design a simple pipeline using Phasers. In order to have a finished cake, you need to first mix the ingredients, then bake the cake, and finally ice it with frosting. Although you only have one tool for each step (one whisk, one oven, and one icing spatula), there's no problem with icing one cake while another is baking in the oven. That's where the parallelism comes in. Answering the above questions should give you all the information you need to set up this method. Remember that this needs to be able work for any number of cakes passed through!

Note that icer.ice() returns an object of IcedCake. Collect all the 'IcedCake's you make, and return them in an array.

Warning:Do not forget to ensure that all of your tasks complete before you return.

Optional Challenge

class:	MultiOvenCakePipeline.java
methods:	mixBakeAndIceCakes
package:	pipeline.cake.challenge
source folder:	student/src/main/java

method: public static IcedCake[] mixBakeAndIceCakes(Mixer mixer, List<Oven> ovens, Icer icer, int cakeCount) (parallel implementation required)

As we've discussed in class, not all steps of the pipeline always take the same amount of time to complete. In this example, baking a cake in the oven usually takes much longer than simply mixing the ingredients or icing the cake. The Oven is the bottleneck on the performance of this pipeline. One way to further improve the process then is to simply more ovens. This time, build a similar pipeline to what you made earlier, but use all of the Ovens passed to you in order to speed things up. Here's a couple things to take note of:

• How have the tasks changed? How does this additional oven affect the number of tasks your program will use?

Testing Your Solution

Visualization

class:	CakePipelineVizApp.java	VIZ
package:	pipeline.cake.viz
source folder:	student/src//java

The viz below is for the challenge assignment.

Correctness

class:	CakePipelineTestSuite.java
package:	pipeline.cake.studio
source folder:	testing/src/test/java

When you are passing the tests and your visualization looks good, demo it to an instructor.