Concurrency

Question 1

What are the 6 stages of a threads lifecycle?

Answer 1

Ready to run, 
running, 
waiting, 
sleeping 
blocked on I/O or synch
Dead

Question 2

What is event dispatch?

Answer 2

The gui thread within java

Question 3

When does a single thread die?

Answer 3

When its run method returns or .stop is called on it

Question 4

What class do the .notify, and .wait methods belong to?

Answer 4

Belongs to the object class

Question 5

What object do the sleep and, start and yield belong to?

Answer 5

Thread object.

Question 6

When is a class thread safe?

Answer 6

If it behaves correctly when accessed by multiple threads. Regardless of the scheduling or interleaving of those threads by the runtime. With no extra synchronisation on the part of the calling code.

Question 7

Why should you avoid sharing states between threads wherever possible?

Answer 7

Sharing states leads to race conditions and deadlock?

Question 8

What are the ways to create a new thread?

Answer 8

Calling .run() on any of the following:
A new anonymous thread class, 
A new thread with a lambda inside. 
New thread subtype with the run class overridden, 
New runnable implementation passed as a parameter to a new Thread.

Question 9

Implementing runnables vs extending threads? Whats the difference.

Answer 9

Inheritance should be used when you want to override existing functionality, what existing behaviour are we trying to override in thread? Instead we seek to extend the existing behaviour by giving it something to run. Furthermore, implementing runnables creates loose coupling in our program and allows our runnable to inherit something else.

Question 10

What is a race condition?

Answer 10

A race condition occurs when two or more threads can access shared data and they try to change it at the same time. Thread scheduling means that their order is random. The random timing and simultaneous access can result in unexpected results.

Question 11

What are check then act and read then write.

Answer 11

Neither of these operations are atomic, check then act can make the condition that was checked become outdated midway throughout execution.
Whilst in read then write two threads can read a value at the same time causing the wrong write value.

Question 12

Why can even primitives behave unexpectedly?

Answer 12

The JVM caches values for periods of time. Meaning that the change in value may not be a publicly available resource.

Question 13

What is write back caching?

Answer 13

When stuff is placed in the cache and then eventually put into main memory in batches.

Question 14

What is write through caching?

Answer 14

When the processor is told to put straight from the cache into main memory.

Question 15

Which is faster out of write through and write back caching? Which is safer?

Answer 15

Write back is faster. Write through is safer.

Question 16

What is a thread?

Answer 16

A flow of control within a program.

Question 17

What are the advantages of threads?

Answer 17

• Reactive/realtime system.
• Responsive to user input.
• Concurrent server requests.
• Take advantage of multicores/processors.
• Supports object decoupling

Question 18

What methods for thread control are deprecated?

Answer 18

.stop, .suspend and .resume. Because they suddenly seize control away from the thread mid execution.

Question 19

What are daemon threads?

Answer 19

They are essentially like background support task threads. Once all non daemon threads die, then all daemons die aswell and the program finishes.

Question 20

Can primitives be locked?

Answer 20

No

Question 21

What can be synchronized?

Answer 21

Methods or blocks.

Question 22

What does synchronization do?

Answer 22

When threads are executing and they encounter a synchronized method or block, in order to execute the following code, they most own the lock to a specificied object.

Question 23

What does synchronization do?

Answer 23

When threads are executing and they encounter a synchronized method or block, in order to execute the following code, they must own the lock to a specified object.

Question 24

What is an intrinsic lock/monitor lock/monitor?

Answer 24

Every object has an intrinsic lock associated with it. Threads which need exclusive access to an objects fields/methods must own the intrinsic lock for that respective object.

Question 25

What happens when a thread holds the monitor for an object?

Answer 25

All other threads which wish to access the object must wait until the monitor lock has been released to them.

Question 26

What happens when static methods are synchronized?

Answer 26

A separate monitor exists for the entire class which must be acquired to utilise any synchronized static fields.

Question 27

What happens when static methods are synchronized?

Answer 27

A separate monitor exists for the entire class which must be acquired to utilise any static fields.

Question 28

What is reentrant synchronization.

Answer 28

Threads cannot acquire a lock held by other threads. They can reacquire the same lock repeatedly. When doing so, the number of lock releases must equal the lock acquisitions for the lock to be properly released.

Question 29

What causes deadlock

Answer 29

When two threads are both waiting for each others locks in order to continue, neither thread can progress, causing them to wait indefinitely.

Question 30

What causes deadlock

Answer 30

When two threads are both waiting for each others locks in order to continue, neither thread can progress, causing them to wait indefinitely.

Question 31

What is the modern toolbox for concurrency?

Answer 31

Java.util.concurrent

Question 32

Do atomic classes inherit from their similarly named classes?

Answer 32

No. Though AtomicInteger extends Number.

Question 33

What does the Atomic set of classes do?

Answer 33

Supports lock free thread-safe programming on single variables. Each object represents a corresponding primitive and contains helper methods.

Question 34

What does java.util.concurrent.locks do?

Answer 34

Provides an alternative model of synchronization within Java, essentially it expands the things that locks can do. it can replicate any functionality from block structured concurrency.
New ideas include:
-Add different types of locks (such as reader and writer locks).

• Not restrict locks to blocks (allow a lock in one method and unlock in another).
• If a thread cannot acquire a lock (for example, if another thread has thelock), allow the thread to back out or carry on or do something else—a try-Lock() method.
• Allow a thread to attempt to acquire a lock and give up after a certain amountof time

Question 35

What is a countdown latch?

Answer 35

A synchronization pattern which allows multiple threads to agree on a minimum amount of preparation to be done before any thread can pass a synchronization barrier.

Question 36

How is a countdown latch implemented?

Answer 36

You must use
new CountDownLatch(int)
then use cdl.await() which will cause all threads to wait until calls to .countdown means that cdl reaches 0.

Question 37

How does ConcurrentHashMap work?

Answer 37

Placing things in a hashmap only really requires you to lock the bucket being altered.

Question 38

Why is ConcurrentHashmap useful?

Answer 38

Increased performance and thread safety without the any explicit synchronization of blocks.

Question 39

What is java.util.concurrents alternative to an arraylist?

Answer 39

CopyOnWriteArrayList
If you look at the underlying array reference you’ll see it’s marked as volatile. When a write operation occurs (such as in the above extract) this volatile reference is only updated in the final statement via setArray. Up until this point any read operations will return elements from the old copy of the array.

The important point is that the array update is an atomic operation and hence reads will always see the array in a consistent state.
Can cause low performance when many write operations occur and can also cause stale values to be read in an iterator.

Question 40

When is CopyOnWriteArrayList bad?

Answer 40

It is bad when write operations outweigh read operations. This is because it must copy the underlying array every time which affects performance.

Question 41

What are the benefits of the queue?

Answer 41

The queue provides a simple and reliable way to dstribute processing resources to work units.

Question 42

What is java.util.concurrents implementation of a queue?

Answer 42

The blocking queue. When trying to put() to the queue, it will cause the putting thread to wait for
space to become available if the queue is full.
When trying to take() from the queue, it will cause the taking thread to block
if the queue is empty

Question 43

How does a transfer queue differ to a blocking queue?

Answer 43

It just has one extra operation transfer, this directly transfers a work item to a waiting receiver thread

Question 44

What does the callable interface do?

Answer 44

Has one method call, which runs some code and returns a value.

Question 45

What does the future interface do?

Answer 45

It is used to represent an asynchronous task, it uses .get() to try and get a result or otherwise block until it is ready. Also has .cancel if you don’t want to wait anymore.

Question 46

What is FutureTask interface?

Answer 46

It is a combination of the future and runnable interface.

Question 47

What is the ScheduledThreadPoolExecutor?

Answer 47

It is the backbone of thread pool classes