- Collaboration Shatters Latency and Consistency Barriers for Java Applications
- Azul Zing plus Hazelcast Enterprise provide 40% overall performance improvement
- With Zing, max latency is reduced by 28X over Oracle’s HotSpot JVM
- Hazelcast’s ideal use cases include caching, ecommerce, payments processing, and financial services
Azul Systems (Azul), the award-winning leader in Java runtime solutions, today announced that it has partnered with Hazelcast, the leading open source in-memory data grid, to ensure explicit Service Level Agreement (SLA) assurance for Java-based in-memory computing applications. By combining Zing plus Hazelcast Enterprise, Java-based businesses are now able to deliver consistent in-memory application performance without the risk of performance-robbing Java runtime artifacts.
A recent benchmark showed that the Zing/Hazelcast combination had a maximum latency of 8 milliseconds at the 99.99th (four nines) percentile. The same workload using the Oracle HotSpot JVM had a maximum latency of 225 milliseconds, an improvement of over 28X when using Zing. Across the entire latency profile range, average transactions were 40% faster with Zing. In all cases the configuration was identical, using 2-node clusters and both 1 and 2 GB JVM heaps. Full benchmark results are available on the Hazelcast website.
SLAs can be formal or informal, depending upon the business requirements. Lower peak latencies and predictable performance are key to driving revenues and reducing exposure in business-critical applications. Here’s how the Zing/ Hazelcast combo is delivering value and SLA compliance today in three important use cases:
Ecommerce: As everyone who has completed a web-based shopping transaction knows, slow websites are bad for business, especially during the holiday shopping season when demand is at its peak. 200 milliseconds per online transaction is ideal – in most cases 2 seconds is tolerable, and within 5-10 seconds of a screen freeze the customer vanishes. Online buyers won’t wait – they will abandon the session or the shipping cart and move on to a provider with a better user experience and a faster site. Today a leading ecommerce powerhouse uses Hazelcast with Zing to meet their SLA targets and ensure that customer inquiries and transactions happen quickly – every time, even under high, unpredictable loads of the holiday season.
Payments processing: Credit card companies face a balancing act between their need to ensure that a transaction is valid and the very brief time available to make a go/no go decision. Their merchant clients and end consumers won’t tolerate long wait times – and if credit card validation takes too long and payment is approved, the payment processor bears the fraud risk. By adding the Zing runtime to their Hazelcast-based risk platform, payment processors can meet their SLA targets while reducing their fraud exposure.
Financial services: Low latency is a requirement within capital markets, where the proliferation of algorithmic trading requires firms to react to market events faster than the competition to increase profitability of trades. For example, when executing arbitrage strategies the opportunity to “arb” the market may only present itself for a few milliseconds before parity is achieved. To demonstrate the value that clients put on latency, a large global investment bank has stated that every millisecond lost results in $100m per annum in lost opportunity. ¹ Hazelcast combined with Zing slashes latency and shaves precious milliseconds of latency, capturing revenue and profits otherwise unattainable.
“Achieving predictable and capped latency has been a challenge in Java systems and a major historical weakness of the use of Java," said Greg Luck, CEO of Hazelcast. “Achieving predictable latency with Java has been traditionally managed with smaller heaps and garbage collection (GC) tuning. But with Azul Zing we can achieve a latency cap of <10ms and at the same time lift performance, eliminating costly tuning and achieving performance consistency, a win-win for latency sensitive applications.”
“Businesses are constantly faced with balancing the need to develop and deliver innovative new capabilities and features for their revenue generating platforms with development costs and time-to-market requirements," said Scott Sellers, Azul Systems president and CEO. "Java remains the world’s most popular and productive language for the enterprise, however the constant tuning and tweaking needed to try and achieve consistent performance is expensive and often times futile. Azul Zing, combined with in-memory datagrids like Hazelcast Enterprise, deliver rock-solid consistent performance that scales to meet the demanding requirements of today’s high growth businesses.”
¹ "Wall Street's Quest To Process Data At The Speed Of Light". Information Week.
Azul Systems, the industry's only company exclusively focused on Java and the Java Virtual Machine (JVM), builds fully supported, certified standards-compliant Java runtime solutions that bring the power of Java to the enterprise, the embedded community and the IoT. Zing is a JVM designed for enterprise Java applications and workloads that require any combination of low latency, high transaction rates, large working memory, and/or consistent response times. Zulu is Azul's certified, freely available open source build of OpenJDK. Cost-effective Zulu Enterprise subscriptions extend OpenJDK support to Java-based businesses who prefer open source solutions. Zulu Embedded is Azul's 100% open source build of OpenJDK for Intel, ARM, and PowerPC -based designs that require custom configurations and long support life cycles.
Azul Systems, the Azul Systems logo, Zulu, Zing and ReadyNow! are registered trademarks.
Java and OpenJDK are trademarks of Oracle Corporation and/or its affiliated companies in the United States and other countries. All other trademarks are the property of their respective holders.
Hazelcast is the leading provider of operational in-memory computing with tens of thousands of installed clusters and over 16 million server starts per month. The Hazelcast operational in-memory computing platform helps leading companies, like Capital One, Chicago Board Options Exchange, Deutsche Bank, Ellie Mae, and Mizuho Securities USA, manage their data and distribute processing using in-memory storage and parallel execution for breakthrough application speed and scale.
Hazelcast’s developer-friendly approach makes it easy to modernize existing applications while providing a platform for building new innovative solutions. Hazelcast is headquartered in Silicon Valley’s Palo Alto, with offices in Ankara, Istanbul, London, and New York City.