Welcome!

Mobile IoT Authors: Liz McMillan, Terry Ray, Elizabeth White, Pat Romanski, Kevin Benedict

Related Topics: Java IoT, Industrial IoT, Microservices Expo, Open Source Cloud, Machine Learning , Recurring Revenue

Java IoT: Blog Post

A Painless Introduction to Java’s ThreadLocal Storage

Best practices for using another powerful class: ThreadLocal from java.lang

In my last blog, we explored WeakHashMap which is based by WeakReference in Java. Now let's look at some best practices for using another powerful class: ThreadLocal from java.lang, which is also implemented using WeakReference.

What is ThreadLocal? A simple example
As its name suggests, a single instance of ThreadLocal can store different values for each thread independently. Therefore, the value stored in a ThreadLocal instance is specific (local) to the current running Thread, any other code logic running on the same thread will see the same value, but not the values set on the same instance by other threads. (There are exceptions, like InhertiableThreadLocal, which inherits parent thread's values by default.)

Let's consider this example:

We have a TransactionManager class that provide static methods to:

  • Start a transaction with a generated ID
  • Store that ID as a static field and provide a transaction ID getter method to other code logic that needs to know the current transaction ID.

In a single threaded environment, TransactionManager can simply store the ID as a static field and return as is. However, this will certainly not work in a multiple-threaded environment. Imagine multiple threads are using TransactionManager - transaction IDs generated by each thread can overwrite each other as there is only one static instance of transaction ID. One may synchronize and block other transactions to avoid overwrites, but this would totally defeat the purpose of having multiple threads.

In order to solve this problem, ThreadLocal provides a very neat solution:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 public class TransactionManager { private static final ThreadLocal<String> context = new ThreadLocal<String>(); public static void startTransaction() { //logic to start a transaction //... context.set(generatedId); } public static String getTransactionId() { return context.get(); } public static void endTransaction() { //logic to end a transaction //... context.remove(); } }

Different thread that starts transactions via TransactionManager will get its own transaction ID stored in the context. Any logic within the same thread can call getTransactionId() later on to retrieve the value belongs/local to that Thread. So problem's solved!

The Internals of ThreadLocal and How it Works

Let's drill down a little bit into the ThreadLocal's internals. ThreadLocal is implemented by having a Map (a ThreadLocalMap) as field (with WeakReference entry) within each Thread instance. (There are actually 2 maps; the second one is used for InheritabeleThreadLocal, but let's not complicate the picture). The keys of those maps are the corresponding ThreadLocals themselves. Therefore, when a set/get is called on a ThreadLocal, it looks at the current thread, find the map, and look up the value with "this" ThreadLocal instance.

Still confused? I certainly am. Let's look at a real example.

  • Code running in Thread 1 calls set() on ThreadLocal instance "A" with value "123″
  • Code running in Thread 2 calls set() on ThreadLocal instance "A" with value "234″
  • Code running in Thread 1 calls set() on ThreadLocal instance "B" with value "345″

And this is the end result:

Thread 1 (the instance)'s field ThreadLocalMap (m1) has 2 entries:

KeyValue
ThreadLocal A "123″
ThreadLocal B "345″

Thread 2 (the instance)'s field ThreadLocalMap (m2) has 1 entry:

KeyValue
ThreadLocal A "234″

Now if some code logic in Thread 1 calls get() on ThreadLocal instance "A", the ThreadLocal logic will lookup the current Thread, which is instance Thread 1, then access the field ThreadLocalMap of that Thread instance, which is m1, it can then lookup the value by using m1.get(this), with "this" as ThreadLocal and the result is "123″

Now what to watch out for!
Did I hear weak reference for ThreadLocal entries? Does that mean I don't have to clean up? Well, it's not quite that simple.

First of all, the value object put into the ThreadLocal would not purge itself (garbage collected) if there are no more Strong references to it. Instead, the Weak reference is done on the thread instance, which means Java garbage collection would clean up the ThreadLocal map if the thread itself is not strongly referenced elsewhere.

So now the question is: when would the Thread object get garbage collected?

The answer is: it depends, but always assume the thread is long running. 2 common examples:

  • Servlets. The threads that handle servlet requests usually stay alive in the container for the lifetime of the server instance. Code logic that uses ThreadLocal might be referenced indirectly by servlets.
  • Thread pooling java.util.concurrent.Executors. Java encourages recycling threads!

A typical usage of Executor introduced in Java 1.5, if ThreadLocal maps are not cleaned up properly after a transaction is done, next TransactionProcessingTask might inherit values from another previous unrelated task!

1 2 ExecutorService service = Executors.newFixedThreadPool(10); service.submit(new TransactionProcessingTask());

Be careful with initialization of ThreadLocal, below is an implementation of a counter by Thread. Can you tell what is wrong in the below initialzation?

1 2 3 4 5 6 7 8 9 10 11 12 public class Counter { private static ThreadLocal<Integer> counter = new ThreadLocal<Integer>(); static { counter.set(0); } public int getCountInThread() { return counter.get(); } //.... }

The counter would not get initialized correctly! Though the counter is declared as static, it CANNOT be initialized by having a static initializer, as the initializer only runs once when the first thread references the Counter class. When the second thread comes in, it does not run counter.set(0) on that thread, therefore counter.get() returns null instead of 0! One solution is to sublcass ThreadLocal and override the initialValue() method to assign non-null initial value.

With these in mind, you can probably picture the consequences of not cleaning up after ourselves! An operation that runs on a recycled thread might inherit the values from previous operation on the same thread! Besides, it can also cause memory leaks as the instance stored in ThreadLocal will never get garbage collected if the thread is alive.

As a rule of thumb, always clean up/reset your threadlocal after you have finished your "unit of operation"! Even though the current code might be simple enough to bypass the cleanups, it might be adapted and integrated into servlets/thread pooling later on! After all, cleaning up responsibly is always appreciated both in the realms of programming and real life.

More Stories By Patson Luk

A Java developer who has spent the better part of the last decade working on financial services applications with companies from HSBC to Mobilearth and Parasoft, Patson is experienced in various aspects of computer systems, from large scale enterprise banking system to lightweight mobile payment solutions. He now leads Java instrumentation and tool development for the TraceView product at AppNeta. Patson's focus is on using java bytecode manipulation technologies to gain greater visibility into the full spectrum of Java based technologies. This includes higher level application frameworks from Spring and Struts to Webflow, AppServers from TomCat to JBoss, and Databases from MySQL to Oracle. He's also writes frequently on the AppNeta blog - www.appneta.com/blog

IoT & Smart Cities Stories
The deluge of IoT sensor data collected from connected devices and the powerful AI required to make that data actionable are giving rise to a hybrid ecosystem in which cloud, on-prem and edge processes become interweaved. Attendees will learn how emerging composable infrastructure solutions deliver the adaptive architecture needed to manage this new data reality. Machine learning algorithms can better anticipate data storms and automate resources to support surges, including fully scalable GPU-c...
Machine learning has taken residence at our cities' cores and now we can finally have "smart cities." Cities are a collection of buildings made to provide the structure and safety necessary for people to function, create and survive. Buildings are a pool of ever-changing performance data from large automated systems such as heating and cooling to the people that live and work within them. Through machine learning, buildings can optimize performance, reduce costs, and improve occupant comfort by ...
The explosion of new web/cloud/IoT-based applications and the data they generate are transforming our world right before our eyes. In this rush to adopt these new technologies, organizations are often ignoring fundamental questions concerning who owns the data and failing to ask for permission to conduct invasive surveillance of their customers. Organizations that are not transparent about how their systems gather data telemetry without offering shared data ownership risk product rejection, regu...
René Bostic is the Technical VP of the IBM Cloud Unit in North America. Enjoying her career with IBM during the modern millennial technological era, she is an expert in cloud computing, DevOps and emerging cloud technologies such as Blockchain. Her strengths and core competencies include a proven record of accomplishments in consensus building at all levels to assess, plan, and implement enterprise and cloud computing solutions. René is a member of the Society of Women Engineers (SWE) and a m...
Poor data quality and analytics drive down business value. In fact, Gartner estimated that the average financial impact of poor data quality on organizations is $9.7 million per year. But bad data is much more than a cost center. By eroding trust in information, analytics and the business decisions based on these, it is a serious impediment to digital transformation.
Digital Transformation: Preparing Cloud & IoT Security for the Age of Artificial Intelligence. As automation and artificial intelligence (AI) power solution development and delivery, many businesses need to build backend cloud capabilities. Well-poised organizations, marketing smart devices with AI and BlockChain capabilities prepare to refine compliance and regulatory capabilities in 2018. Volumes of health, financial, technical and privacy data, along with tightening compliance requirements by...
Predicting the future has never been more challenging - not because of the lack of data but because of the flood of ungoverned and risk laden information. Microsoft states that 2.5 exabytes of data are created every day. Expectations and reliance on data are being pushed to the limits, as demands around hybrid options continue to grow.
Digital Transformation and Disruption, Amazon Style - What You Can Learn. Chris Kocher is a co-founder of Grey Heron, a management and strategic marketing consulting firm. He has 25+ years in both strategic and hands-on operating experience helping executives and investors build revenues and shareholder value. He has consulted with over 130 companies on innovating with new business models, product strategies and monetization. Chris has held management positions at HP and Symantec in addition to ...
Enterprises have taken advantage of IoT to achieve important revenue and cost advantages. What is less apparent is how incumbent enterprises operating at scale have, following success with IoT, built analytic, operations management and software development capabilities - ranging from autonomous vehicles to manageable robotics installations. They have embraced these capabilities as if they were Silicon Valley startups.
As IoT continues to increase momentum, so does the associated risk. Secure Device Lifecycle Management (DLM) is ranked as one of the most important technology areas of IoT. Driving this trend is the realization that secure support for IoT devices provides companies the ability to deliver high-quality, reliable, secure offerings faster, create new revenue streams, and reduce support costs, all while building a competitive advantage in their markets. In this session, we will use customer use cases...