Java is one of the most popular and widely used software development platforms on the planet. One of the main reasons for its popularity is the ability of the Java Virtual Machine (JVM) to scale to the largest of applications. Global enterprises such as Netflix, Amazon and eBay all use Java for large parts of their server-side architecture delivering services to millions of customers worldwide under demanding service level agreements (SLAs).
This whitepaper outlines how to drive down Java support costs while ensuring secure, stable operations.
This white paper describes Falcon, a just-in-time (JIT) compiler for the Zing JVM. Falcon replaces the existing C2 compiler, providing improved performance plus much greater flexibility and maintainability. The Falcon JIT compiler is based on technology from the open source LLVM project. This white paper also highlights some of the advantages Falcon offers for Java-based applications when deployed on modern server hardware. The focus is technical.
Written for architects and Java experts, this paper, originally presented to the International Symposium on Memory Management (ISMM), describes how C4 differentiates itself from other generational garbage collectors by supporting simultaneous – generational concurrency: the different generations are collected using concurrent (non stop-the-world) mechanisms that can be simultaneously and independently active. The C4 Continuously Concurrent Compacting Collector, is an updated generational form of the Azul Pauseless GC Algorithm. This paper was originally published in the proceedings of the ISMM 2011, copyright ACM, 2011. Available here under license.
This white paper presents the challenges the Internet of Things (IoT) brings for software development, and illustrates why Open Source Java is an ideal solution that addresses all of these challenges. It also addresses why Java is not as widely used today in IoT as might be expected given its advantages.
It is clear that the direction of software development is primarily moving to the Cloud and using Microservices as a way to do this in a flexible, scalable manner.
This white paper illustrates why Java is the obvious choice for microservices, as it has a number of distinct advantages over other languages. However, by using a traditional JVM, several deployment issues must be carefully considered when choosing an application architecture.
While new In-memory Computing (IMC) techniques hold the promise of better scalability and performance, ensuring consistent, low latency performance isn’t a guarantee for all Java applications. Product choices, runtime components and deployment topologies are essential to maximizing the values of these new IMC solutions, including In-memory Data Grids (IMDGs). This benchmark study compares the response time performance of two different JVMs, namely Azul Zing and Java HotSpot, while running Red Hat JBoss Data Grid at different node sizes. All benchmark results were derived using the open source RadarGun framework and the jHiccup measurement tool.
This benchmark study compares the response time performance of two different Java Virtual Machines (JVMs), namely Azul Zing® and Oracle HotSpot®, while running Apache Cassandra at different throughput levels.
Apache Cassandra is built in Java, which makes it subject to the limitations and performance hiccups associated with traditional Java Virtual Machines (JVMs). Common issues include response time outliers, frequent Java garbage collections, out of memory errors and system hangs. Companies deploying Cassandra spend extensive time tuning and tweaking the JVM to avoid these issues, then need to re-tune each time load (or the application accessing Cassandra) changes.
This white paper describes how Zing allows you to avoid JVM tuning and get your Apache Cassandra deployment launched faster, and improve your Cassandra performance.
In this white paper published by 451 Research, Analyst John Abbott discusses Docker and Java, “For enterprise customers with heavy-duty Java applications wanting the flexibility of virtualization without the performance penalty and unpredictability, Docker could be something of a savior.”
When evaluating innovative new technologies like Zing, delivering a compelling return on investment (ROI) is essential. Written for business people with direct or indirect P&L responsibility, this white paper discusses how Azul’s Zing technology enables firms to leverage the power of Java across the enterprise – even previously sacred high performance/low-latency areas – by dramatically improving performance, resource leverage and above all, ROI to deliver enhanced revenue opportunities and lower costs.
Written for application and line of business owners, this paper reviews the current limitations of enterprise Java applications, challenges of modern deployment topologies such as virtualization and Clouds and the advantages of a highly innovative and elastic JVM that can provide guaranteed predictability, even under load. Azul Systems has adapted award-winning technology from its line of Java accelerator appliances to create the Zing® JVM, a unique, software-based solution. Zing complies with the Java SE standard and is a JVM designed for enterprise applications and workloads that require large memory, high transaction rates, consistent response times and high sustained throughput.
Written for IT professionals, architects and developers, this white paper reviews and classifies the various garbage collectors and collection techniques available in JVMs today. It provides an overview of common garbage collection techniques, algorithms and defines terms and metrics common to all collectors including generational, parallel, stop-the-world, incremental, concurrent, and mostly-concurrent. It further classifies each major JVM collector’s mechanisms & characteristics and discusses the tradeoffs involved in balancing requirements for responsiveness, throughput, space, and available memory across varying scale levels. The paper concludes with some pitfalls, common misconceptions, and “myths” around garbage collection behavior, as well as examples of how certain choices can result in impressive application behavior.
Written for IT and developers, this white paper provides technical detail on the operation of Zing’s garbage collector to explain how it achieves pauseless operation. Based on the Azul C4 (Continuously Concurrent Compacting Collector) pauseless garbage collection algorithm the Zing collector improves application performance and removes barriers to Java scalability by eliminating pause times even at very large heap sizes. Using a combination of software techniques and hardware emulation features, the Azul garbage collector uses ‘concurrent compaction’ to allow applications to continue processing while remapping memory.
Azul is a small company, and challenging Oracle could turn out to be a daunting task. But there’s little doubt that many customers will welcome an alternative source for enterprise Java support beyond Oracle itself which ends support for older versions to fit its own schedule rather than that of its users. Will Oracle fight back? We doubt it. In reality, there’s little interest in individual products down at Redwood Shores, where the focus is firmly on integrated stacks of both software and hardware.
Web portals are ubiquitous in the enterprise and across the Internet today and have become the de facto approach for offering services, ranging from web facing e-commerce, to customer self-service, to corporate portals that are central to daily employee productivity. Portals typically provide a central, session-based, authenticated interaction with multiple internal subsystems, covering a wide range of user operations. Portals will typically carry session state associated with user information for the duration of a user’s session, as well as transactional state during the execution of internal operations. Since portals typically provide a central interaction point between a user and a business-critical set of functionality, response time and availability metrics are key to proper portal operations. In this paper we will review performance metrics for a transaction-centric, Web portal application under specific service level agreements (SLAs).