A senior Java engineer recounts how a team evolved from a MyBatis interceptor for environment-based data isolation to a clean annotation-plus-AOP solution after a junior developer's ThreadLocal-based quick fix caused nested-call bugs and scattered boilerplate across the codebase.
This article explains five common Java memory leak scenarios—long-lived collections, unclosed resources, ThreadLocal misuse, inner class references, and unregistered listeners—with code examples and recommends MAT and VisualVM for leak detection.
This comprehensive guide covers 10 hidden Java memory leak scenarios, a 6-step MAT-based investigation SOP, code-level fixes for ThreadLocal and static collections, emergency mitigation tactics, and long-term prevention standards to eliminate recurring OOM crashes.
This article demonstrates how to implement dynamic data source switching in Spring Boot using ThreadLocal and AbstractRoutingDataSource, covering configuration, annotation-based switching, and runtime data source registration with Druid.
ThreadLocal provides per-thread variable copies via a ThreadLocalMap with weak-reference keys and strong-reference values, enabling request-context storage and non-thread-safe object reuse, but risks memory leaks in long-lived thread pools unless explicitly removed, and requires careful handling for cross-thread propagation.
The article analyzes why Java virtual threads, despite promising million-concurrency with minimal code changes, see little production adoption due to pinning with synchronized, ThreadLocal memory pressure, incomplete framework support, and CPU-bound workload mismatch, while showing benchmarks where they excel in I/O-heavy services but migration costs deter legacy systems.
This article analyzes why Java virtual threads, despite promising million-concurrency with minimal overhead, see little production adoption due to pinning with synchronized blocks, ThreadLocal memory explosions, incomplete framework support, and unsuitability for CPU-bound workloads, while showing benchmarks where they excel in I/O-heavy scenarios.
This article walks through a production-ready Spring Boot multi-tenancy implementation using database-per-tenant isolation, JWT-based tenant identification via Spring Security, AbstractRoutingDataSource for dynamic routing, and ThreadLocal context management, while exposing critical pitfalls in transaction timing, async task propagation, and dynamic tenant registration.
The article examines why the early hype around Java virtual threads has faded, detailing technical pitfalls such as pinning and ThreadLocal memory bloat, incomplete ecosystem support, and misuse in CPU‑bound workloads, while also presenting benchmark data that shows they still excel in I/O‑heavy scenarios.
The article explains why platform threads are costly, how JDK 21 virtual threads eliminate that cost, and provides practical guidance on creating, using, and debugging virtual threads, including pinning pitfalls, rules against pooling, structured concurrency, ScopedValue, and a detailed model‑selection comparison.
The article examines a real‑world incident where ThreadLocal leakage caused user data mix‑ups, then walks through the evolution of context propagation—from implicit ThreadLocal to TTL patches, Reactor Context, ScopedValue, and Go's explicit context—highlighting trade‑offs and best practices for high‑throughput systems.
This article walks through ten essential Java concurrency interview topics—including the Java Memory Model, volatile semantics, synchronized lock upgrades, CAS pitfalls, AQS internals, ReentrantLock versus synchronized, ThreadPoolExecutor parameters, ThreadLocal memory‑leak issues, CountDownLatch vs CyclicBarrier, and deadlock conditions—providing code snippets, tables and detailed explanations to help candidates answer interview questions confidently.
This article reviews common Java concurrency pitfalls—including deadlocks, ThreadLocal memory leaks, Integer‑cache loops, unbounded thread‑pool queues, and swallowed exceptions—illustrates each with code samples, analyzes real production incidents, and provides a concise checklist and tool‑selection guide for safe concurrent programming.
A production outage at 3 am revealed a 10‑second Full GC pause on an 8 GB heap; the article walks through evidence collection, heap‑size rebalancing, switching to G1, fine‑tuning G1 parameters, and code changes that together reduced Full GC pauses to 200 ms and improved latency dramatically.
This article shares production-hardened patterns for multi-tenant SaaS using Spring Boot, covering isolation mode selection, ThreadLocal-based context propagation, MyBatis-Plus SQL interception for automatic tenant_id injection, Feign and async thread propagation, HikariCP tuning, composite indexing, cursor pagination, and real-world pitfalls like interceptor ordering, transaction-bound data source switching, and ThreadLocal leakage.
At 3 am a production server saw Full GC spike from once a day to ten times a minute because a ThreadLocal holding a large object was never removed, leading to a memory leak that almost took the machine down, and the article explains the root cause, internal mechanics, and remediation strategies.
The article explains that a single ThreadLocal object can be shared across dozens of threads because the actual data lives in each thread's own ThreadLocalMap, outlines how set/get work, highlights thread‑pool pitfalls such as stale data and memory leaks, and provides best‑practice guidelines for static declaration, proper removal, and using TransmittableThreadLocal for cross‑thread propagation.
The article explains ThreadLocal's lifecycle, why userId disappears when spawning asynchronous threads, compares ThreadLocal, InheritableThreadLocal, and Alibaba's TransmittableThreadLocal, and shows how the latter prevents data contamination in thread‑pool reuse.
This article explains how ThreadLocal provides per‑thread variable copies in Java web applications, details its internal storage in Thread objects, showcases ten practical scenarios—from request tracing to async tasks—and highlights common pitfalls such as memory leaks and thread‑pool data contamination, offering concrete best‑practice recommendations.
This article examines five typical pitfalls when using Spring Boot filters—request body consumption, ThreadLocal leakage, duplicate execution, ordering conflicts, and async handling—provides concrete code examples, explains why each issue occurs, and offers reliable solutions to keep your applications stable and secure.
An online service experienced frequent Full GC due to a ThreadLocal memory leak in a thread‑pool scenario; the article walks through GC log analysis, heap dump inspection, root‑cause discovery in ThreadFactory logic, and presents a robust fix using Spring’s TaskDecorator to ensure proper cleanup.
This article explains why ThreadLocal values are not automatically transferred in asynchronous or thread‑pool contexts, compares JDK's InheritableThreadLocal with Alibaba's TransmittableThreadLocal (TTL), and shows how TTL captures, replays, and restores context safely with code examples and diagrams.
This tutorial shows how to implement graceful dynamic data source switching in SpringBoot by combining ThreadLocal with AbstractRoutingDataSource, configuring master/slave databases, adding a @DS annotation with AOP support, and enabling runtime addition of new data sources.
This article walks through building an elegant dynamic data‑source switch in SpringBoot by combining ThreadLocal with AbstractRoutingDataSource, covering the context holder, custom routing class, YAML configuration, testing, annotation‑driven switching, and runtime addition of new data sources.
The article recounts a developer’s painful experience with PageHelper, detailing how duplicate registrations, truncated result sets, and password‑update errors stem from the plugin’s ThreadLocal pagination handling, and explains the underlying code paths and safe usage practices.
The article examines unexpected bugs caused by PageHelper’s hidden LIMIT clause injection in MyBatis projects, explains how ThreadLocal pagination parameters persist across requests, and provides practical guidelines and code examples to avoid and clean up these issues.
This guide shows how to eliminate manual logging of user and order identifiers in a Java e‑commerce system by declaring log placeholders, storing values in ThreadLocal, and using a custom @UserLog annotation with Spring AOP to automatically populate MDC variables for Log4j2.
This article provides a comprehensive guide to Java concurrency and related concepts, covering synchronized lock upgrades, object vs class locks, lazy and double‑checked singleton implementations, ThreadLocal mechanics, reflection usage, annotation scopes, JVM class loading, and Redis cluster threading behavior.
The article shares iFlytek's recent campus salary packages, then dives into a detailed Java interview guide covering Redis data types, key expiration handling, thread safety, ThreadLocal usage, MySQL covering and composite indexes, slow‑query analysis, and SQL‑injection prevention techniques.
This article explains the inner workings of Java's ThreadLocal, why it fails to propagate context in asynchronous scenarios, compares manual passing and decorator‑based approaches, introduces the TransmittableThreadLocal (TTL) library with its CRER workflow, and provides a structured interview answer that showcases deep architectural insight.
The article presents a step‑by‑step, code‑driven design of a Spring Cloud SaaS multi‑tenant solution that balances resource sharing and strict data isolation by using a shared‑database, shared‑schema approach with tenant_id filtering, ThreadLocal context, MyBatis‑Plus interceptors, Redis key prefixing, Sa‑Token session segregation, and Spring Boot auto‑configuration.
This article explores the limitations of ThreadLocal—memory leaks, data contamination, inheritance issues, and performance overhead—and introduces Java's ScopedValue as a safer, more efficient alternative, covering its core design, basic and advanced usage, performance benchmarks, migration strategies, and real‑world web application examples.
This article explores the limitations of ThreadLocal, introduces Java's new ScopedValue feature, provides detailed usage examples, performance benchmarks, migration strategies, and compares both approaches to help developers decide when and how to adopt ScopedValue in modern backend applications.
This guide covers essential Java interview topics—including synchronized lock upgrades, object vs class locking, lazy and double‑checked singleton patterns, ThreadLocal mechanics, reflection usage, annotation scopes, Redis cluster behavior, and a linear‑time algorithm challenge—providing clear explanations, code examples, and practical insights for developers preparing for technical interviews.
This article explains how the Transmittable ThreadLocal (TTL) Java agent works, why improper usage can cause context contamination, memory leaks, and CPU spikes, and provides real production cases, code examples, and practical recommendations to avoid these pitfalls.
This article explains how to eliminate manual logging of user and order identifiers in Java e‑commerce services by using Log4j2 placeholders, ThreadLocal, MDC, a custom @UserLog annotation, and Spring AOP to automatically enrich log messages with these contextual fields.
ScopedValues provide a more efficient, memory‑safe, and virtual‑thread‑friendly alternative to ThreadLocal for managing per‑thread context in Java, eliminating manual cleanup, reducing memory leaks, and simplifying context propagation across threads, as demonstrated by code examples and performance benchmarks.
This article explains how improper use of ThreadLocal in Java can lead to memory leaks and OutOfMemoryError, outlines common scenarios that trigger OOM, and provides best‑practice solutions such as proper removal and alternative ThreadLocal types.
This article explains ThreadLocal's role in Java concurrency, the memory‑leak risk when its remove() method is omitted, introduces the new ScopedValue feature in JDK 20 as a safer alternative, and demonstrates practical Spring and StructuredTaskScope examples with complete code snippets.
This guide covers evaluating engineer hourly wages, interview self‑introduction tips, local vs distributed cache choices, ThreadLocal usage, memory‑leak detection with MAT, deadlock troubleshooting, Spring Bean lifecycle, AOP implementation, and Redis‑based distributed locks, plus resources for Spring Boot projects.
An extensive performance benchmark compares four GUID generation techniques—UUID, Snowflake algorithm, ThreadLocal exclusive counters, and atomic counters—across varying thread counts, revealing that ThreadLocal excels in high‑concurrency single‑machine scenarios while Snowflake remains stable for distributed use, with detailed optimization recommendations.
This article explains how Java threads interact using volatile for visibility, synchronized blocks for mutual exclusion, wait/notify (and Condition) for coordination, piped streams for one‑way byte communication, as well as Thread.join and ThreadLocal utilities, providing concrete code examples, best‑practice guidelines, and performance considerations.
ThreadLocal has long served as Java’s go‑to tool for thread‑local data, but its memory‑leak risks and performance penalties become pronounced with virtual threads, while the newer ScopedValue offers automatic cleanup, immutable safety, and superior efficiency, making it a compelling replacement in high‑concurrency scenarios.
This tutorial explains how to implement dynamic data source switching in Spring Boot by combining ThreadLocal with AbstractRoutingDataSource, covering custom context holders, configuration, annotation‑driven switching, dynamic addition of data sources, and complete code examples with test results.
FastThreadLocal, Netty’s custom thread‑local implementation, replaces JDK’s ThreadLocal by using an indexed array to avoid hash collisions, offering faster access; this article explains its background, core classes, source‑code mechanics, performance trade‑offs, and cleanup strategies within Netty’s architecture.
This article explains how ThreadLocal creates per‑thread variable copies, demonstrates its usage with sample Java code, illustrates the internal storage structure, and reveals why improper handling can lead to memory leaks when thread‑local values persist in thread pools.
This article examines the design of Netty's FastThreadLocal, compares its performance and memory‑management advantages over Java's ThreadLocal, and outlines suitable high‑concurrency use cases, implementation details, and practical considerations for developers.
This article explains why Netty introduces FastThreadLocal instead of relying on JDK ThreadLocal, detailing its background, core principles, source code structure—including InternalThreadLocalMap, FastThreadLocalThread, and FastThreadLocal implementations—performance implications, resource reclamation mechanisms, and practical usage within Netty’s memory allocation.
This article presents a comprehensive guide to implementing full‑chain log tracing, unified exception handling, permission interception, request latency monitoring, and standardized log formats in enterprise‑level Java projects using Spring AOP, thread‑local context, and Redis caching.
This article explains how to use SLF4J's MDC with Logback to assign a unique trace ID to each request, propagate it across threads and services, and configure log patterns so that logs become fully traceable for easier debugging in distributed systems.
This article explains the inner workings of Java's ThreadLocal, analyzes how weak and strong references can cause memory leaks in thread pools, demonstrates typical leak scenarios with code examples, and provides diagnostic tools and defensive coding practices to prevent such leaks.
This article explains the internal design of Java's ThreadLocalMap, covering its core Entry structure, hash table storage, linear probing for collision resolution, key operations (set, get, remove), memory‑leak scenarios, automatic cleanup mechanisms, and practical usage patterns such as thread‑context propagation and Android Looper.
ThreadLocal provides per‑thread variable storage to achieve thread isolation, improving concurrency performance, and is essential for safe data handling in Java, but improper use can cause memory leaks, data contamination, and other issues; this article explains its internal mechanism, common scenarios, pitfalls, and best‑practice guidelines with code examples.
This article explains ThreadLocal’s internal mechanism, why improper use can cause OutOfMemoryError in thread pools, demonstrates the issue with a Java example, and provides best‑practice guidance—especially calling remove()—to prevent memory leaks while preparing for interview questions.
This article explains the internal workings of Java's ThreadLocal and Netty's FastThreadLocal, compares their memory management and lookup mechanisms, provides code examples, and offers guidance on when to choose each for high‑concurrency backend applications.
Upgrading to Spring Boot 2.7.1 (Spring Framework 5.3.21) alters the execution order of @Around, @Before, @After, @AfterReturning, and @AfterThrowing advice within the same @Aspect, causing ThreadLocal data loss and prompting a recommendation to consolidate logic into a single @Around advice.
This article walks through the complete process of implementing dynamic data source switching in a SpringBoot application, covering the core concepts of thread‑local context, abstract routing, custom annotations with AOP, configuration, code examples, and important considerations such as transaction management and performance.
This article analyzes unexpected behaviors caused by PageHelper in a Java backend project, such as duplicate user registration, limited query results, and password‑reset errors, and explains how ThreadLocal pagination parameters, startPage(), and cleanup mechanisms lead to these problems while offering practical debugging tips.
This article explains the motivation behind Netty's FastThreadLocal, details its internal design using an indexed array to avoid hash collisions, analyzes the core source code of InternalThreadLocalMap, FastThreadLocalThread, and FastThreadLocal, and discusses performance implications and reclamation strategies.
This article demonstrates how to build a dynamic data source switching mechanism in Spring Boot by leveraging ThreadLocal for thread‑scoped context and AbstractRoutingDataSource for routing, covering manual implementation, annotation‑driven switching, and runtime addition of new data sources.
Facing the need to query multiple databases in a Spring Boot application, this guide walks through building a custom dynamic datasource solution using ThreadLocal and AbstractRoutingDataSource, covering core implementations, annotation‑based switching, dynamic datasource addition, configuration, and comprehensive test results.
Even when you replace HashMap with ConcurrentHashMap and use ThreadLocal for per‑thread data, hidden pitfalls in thread‑pooled environments can cause data leakage and race conditions, but proper cleanup and atomic operations can restore safety, as demonstrated with concrete Java examples.
This article explains how ThreadLocal stores values per thread, details the internal set and get mechanisms, discusses potential memory‑leak problems caused by stale entries, and provides best‑practice guidelines such as using remove() and InheritableThreadLocal for child threads.
ThreadLocal provides a simple yet powerful way to give each thread its own independent variable, eliminating shared-state conflicts; this guide explains its underlying mechanism, creation methods, basic get/set operations, a complete example with console output, and important usage considerations such as potential memory leaks.
This article explains how Java's ThreadLocal provides thread‑specific storage, enabling safe context management across thread pools, web request handling, database connections, logging, and global data propagation, thereby preventing interference and improving concurrency performance.
This article explains Java's ThreadLocal mechanism, detailing its internal ThreadLocalMap structure, core set/get/remove operations with full code examples, and demonstrates how multiple ThreadLocal variables are stored per thread, providing a clear understanding of thread‑local storage in backend development.
The article examines the performance impact of excessive logging in large‑scale Java systems and proposes request‑level sampling with cross‑thread identifier propagation, offering practical component‑based solutions, implementation considerations, and a concrete code example for backend developers.
Log volume can cripple high-throughput Java systems, especially during traffic spikes, so this article explores practical strategies for request-level log sampling, cross-thread trace propagation, and component-based APIs to balance observability with performance, including code examples and implementation considerations.
The article compares four techniques—manual ThreadLocal copying, TaskDecorator, InheritableThreadLocal, and Alibaba's TransmittableThreadLocal—for passing user context and other data from parent to child threads in Spring Boot async execution, recommending the TaskDecorator and TransmittableThreadLocal approaches for production use.
This article explains four techniques—manual setting, TaskDecorator, InheritableThreadLocal, and TransmittableThreadLocal—for propagating context such as user information from a parent thread to child threads in Spring Boot asynchronous execution, providing code examples and recommendations.
This article summarizes how to correctly declare, monitor, configure, and name Java thread pools, explains common mistakes such as unbounded queues and ThreadLocal contamination, and introduces dynamic tuning techniques and open‑source solutions for robust backend concurrency management.
This article analyzes why Spring MVC fails to retrieve request header fields in multithreaded scenarios, explains the underlying ThreadLocal storage mechanism, critiques a common but flawed solution, and presents reliable approaches using CountDownLatch, manual RequestContextHolder propagation, and request caching to ensure header availability across threads.
This guide shows how to automatically include userId and orderId in Java log statements by defining MDC placeholders, storing context data in ThreadLocal, creating a custom @UserLog annotation, and using Spring AOP to inject these values without manual code changes.
This article explains how SpringBoot interceptors work, demonstrates custom interceptor implementation, shows how misuse of ThreadLocal can cause memory leaks and OOM errors, and provides a fix by moving cleanup logic to afterCompletion, complete with code samples and performance screenshots.
This article explains what memory leaks are in Java, outlines common causes such as unreleased resources, static collections, misuse of ThreadLocal, improper variable scope, inner‑class references and off‑heap memory, and provides practical solutions and best‑practice recommendations to avoid them.
This article provides a comprehensive overview of Python multithreading, covering basic concepts, the _thread and threading modules, thread creation methods, synchronization primitives like Lock and RLock, thread-local storage, thread pools, and the differences between multithreading and multiprocessing for both CPU‑bound and I/O‑bound workloads.
This article explains how to enforce data permission control in Java backend applications by using MyBatis mapper annotations, ThreadLocal rule propagation, and a MyBatis interceptor that parses and rewrites SQL statements with permission constraints, providing minimal intrusion to existing business logic.
Learn how to enforce data permission rules in Java applications by using MyBatis mapper annotations, ThreadLocal context propagation, and custom MyBatis interceptors that modify SQL via Druid's AST parser, ensuring minimal intrusion to existing business logic.
This article explains how using a shared class attribute to store open ports in a Python security‑testing framework can cause data contamination across threads, demonstrates the problem with example code, and presents three solutions—reinitializing the attribute, using contextvars, and employing threading.local—to ensure thread‑local isolation and accurate port scanning.
This article compares Java's standard ThreadLocal with Netty's FastThreadLocal, explains their internal mechanisms, shows practical code examples, runs a JMH microbenchmark, and discusses performance, memory management, and usage scenarios to help developers choose the right thread‑local storage solution.
This article analyzes common anomalies caused by the PageHelper pagination plugin in a Java backend project—such as duplicate user registration, limited query results, and password‑reset errors—traces them to ThreadLocal misuse, explains the internal workflow of startPage, getLocalPage, and clearPage, and offers practical recommendations to avoid these pitfalls.
This article explains why standard ThreadLocal mechanisms fail with thread pools, introduces the design of a custom solution using YesThreadLocal and YesRunnable, and then details how Alibaba's TransmittableThreadLocal (TTL) implements context capture, replay, and restoration for seamless asynchronous execution.
This article explains how Spring's @Transactional and @Async annotations can be applied to the same service methods, the implications for transaction boundaries across threads, and the precautions developers must take to maintain data consistency and proper transaction management.
This article examines the performance impact of excessive logging in high-throughput Java services and presents three practical approaches—ThreadLocal-style wrappers, explicit flag propagation, and a decoupled component with an extensible API—to achieve cross-thread request-level log sampling while preserving traceability.
ThreadLocal in Java provides thread‑local variables by maintaining a per‑thread map, enabling data isolation, simplifying inter‑thread data transfer, and storing context information, while requiring careful handling to avoid memory leaks, thread‑pool contamination, and overuse.
This tutorial demonstrates how to build a custom dynamic data source solution in Spring Boot by combining ThreadLocal and AbstractRoutingDataSource, covering the implementation of context holders, routing logic, configuration, testing, annotation‑driven switching, and runtime addition of new data sources.
This article explains how PageHelper's static ThreadLocal pagination parameters can be polluted by exceptions or thread reuse, leading to incorrect query results, and provides detailed analysis of the plugin's internals together with practical solutions such as invoking clearPage() via filters or aspects.
This article explains how to use SLF4J's MDC to tag logs with a request ID, why the ID is lost in asynchronous threads, and how to apply a decorator pattern (MDCRunnable) so that the request ID is correctly propagated and searchable across all threads and thread pools.
This article explains why MyBatis' default SqlSession implementation is not thread‑safe, how Spring‑MyBatis integration introduces SqlSessionTemplate and its dynamic proxy, and how transaction‑bound ThreadLocal management ensures safe SqlSession usage across multiple threads.
This article provides a step‑by‑step guide to Java concurrency, covering core concepts such as threads, runnables, thread lifecycle, synchronization primitives, wait/notify patterns, volatile variables, ThreadLocal storage, and how to design immutable objects for thread‑safety, all illustrated with concrete code examples and detailed explanations.
Unexpected duplicate registrations, truncated query results, and password‑reset SQL errors in a Java backend were traced to PageHelper’s ThreadLocal pagination state persisting across requests when startPage() is called without immediate query execution, so developers should ensure immediate execution or manually clear the ThreadLocal to prevent stale LIMIT clauses.
The article analyzes unexpected behaviors caused by PageHelper in a Java backend project—such as duplicate registrations, limited query results, and password update errors—by dissecting its source code, ThreadLocal usage, and the importance of proper startPage() and clearPage() handling.
This tutorial explains how to switch between multiple databases in a Spring Boot application by implementing ThreadLocal and AbstractRoutingDataSource, covering code implementation, configuration, testing, annotation-based switching, and dynamic addition of data sources with detailed examples and practical guidance.
This article walks through implementing dynamic data‑source switching in SpringBoot by combining ThreadLocal with AbstractRoutingDataSource, covering manual code, configuration, annotation‑driven AOP, and runtime addition of new data sources, complete with test cases and sample code.
This article demonstrates how to build a custom dynamic data‑source solution for Spring Boot by using ThreadLocal to store the current datasource key, extending AbstractRoutingDataSource for routing, configuring multiple datasources in application.yml, and optionally adding annotation‑driven switching and runtime datasource registration.
This guide explains how to implement dynamic data source switching in SpringBoot by leveraging ThreadLocal and AbstractRoutingDataSource, covering basic concepts, code examples for context holder, routing datasource, configuration, annotation‑driven switching, and runtime addition of new data sources with practical testing steps.
This article explains the fundamentals of ThreadLocal, its memory‑leak pitfalls, introduces the new ScopedValue feature in JDK 20, demonstrates practical Spring‑based use cases, and shows how StructuredTaskScope enables safe structured concurrency with code examples.
This article explains how to build a custom dynamic data‑source solution for Spring Boot by leveraging ThreadLocal and AbstractRoutingDataSource, provides complete Java code for context holders, routing datasource, configuration, annotation‑driven switching, and runtime addition of new data sources.
This article analyzes a real production incident caused by improper thread‑pool usage, explains common misconceptions about core parameters, demonstrates faulty code examples, and provides concrete recommendations for configuring thread pools, handling timeouts, and avoiding deadlocks in Java backend services.
This article explains the ThreadLocal mechanism, its memory‑leak pitfalls, introduces the new ScopedValue feature in JDK 20 as a safer alternative, and demonstrates practical Spring and virtual‑thread examples using StructuredTaskScope for modern Java backend concurrency.
