Company: Azul Systems, Sunnyvale, CA
Company Description: Azul Systems, the industry’s only company exclusively focused on Java & the Java Virtual Machine (JVM), builds fully supported, standards-compliant Java runtime solutions that help enable the real time business. Azul is a venture-backed company founded in 2002 with offices around the globe, including the US, EMEA and APAC with extensive partner and reseller networks in EMEA & Eastern Europe.
Nomination Category: New Product & Service Categories
Nomination Sub Category: Financial Services

Nomination Title: Azul Systems ReadyNow!

Tell the story about this nominated product or service (up to 650 words). Describe its function, features, benefits, and performance to date. TEXT REDACTED FOR PUBLICATION

ReadyNow! is a revolutionary new technology that solves the “warm-up” problem prevalent in many Java applications. It is included in Azul’s Zing runtime for Java and ensures peak application performance is available without long “warm-up” periods during business-critical times.

The technology provides three key functions - the first is the ability for operations teams to save and reuse accumulated Java optimizations across runs; and the second is a robust set of APIs and compiler directives that give developers control over the timing and impact of JVM de-optimization. The third is a specific ability for developers to flag “hot” methods and make sure they are fully optimized when needed.

ReadyNow! is particularly well-suited for use in the financial markets that require peak system performance at critical moments like market open as well as unpredictable surges in volume or trading behavior.

The ‘warm-up’ problem for Java applications has been a serious issue that has long plagued the financial sector and other markets requiring consistent, real-time performance from their systems. With Zing’s ReadyNow! technology, Azul solves Java’s ‘warm-up’ problem for market makers, traders and financial firms who rely on being able to consistently act on opportunities throughout the trading day.

The Java “warm-up” problem refers to an aspect of the behavior of a Java Virtual Machine (JVM), the software responsible for running Java applications. The Java runtime was designed to start up quickly, then improve performance over time via its just-in-time (JIT) compilation features. The challenge is, code re-compilation and re-optimization happen after the JVM decides the code is “hot” (hence “HotSpot”) – so many Java-based systems exercise key methods in advance to “warm up” their code.

In financial markets, this “warm-up” period can result in lost trading opportunities, lost liquidity, lost revenue, and ultimately loss of customers. These trading losses and missed opportunities can accumulate and have a significant impact on a firm’s market presence and standing in this highly competitive sector.

To compensate for these problems, firms utilize various techniques including executing “warm-up” test data that attempts to mimic real market conditions, issuing and canceling small live trades, and various other programming practices. These workarounds for the “warm-up” problem carry real operational risks, however, sometimes executing accidental trades that can cost millions or in extreme cases cause market “flash crashes.” And yet, even after assuming these risks, the systems may still not realize peak performance at market open due to the unpredictable and dynamic characteristics of a real-world trading environment.

With ReadyNow!,traders, market makers, and risk analysts can feel confident that their Java-based systems, are “warmed up” when they are needed most, without suffering from unexpected stalls and without assuming operational risk. Developers can use ReadyNow!’s robust APIs to direct the compiler to pre-compile code they know will be commonly used, or code that must be fast even though it’s called infrequently. With ReadyNow!’s code profiles, teams can get best performance run after run, day upon day, because profiles accumulate and make controlling compiler directives a simple setup exercise.

There are many key benefits of ReadyNow! Technology, including:

• dramatically simplifies operational processes and reduces risks associated with “warming-up” Java-based financial systems
• ensures consistent, peak performance during critical times of trading, market making, and risk analysis
• reduces if not eliminates the need for synthetic test data or “warm-up” trades on live systems
• further accelerates the adoption of Java for low-latency, real-time financial systems

The key features of ReadyNow! technology include:

• delivers more consistent low latency performance
• allows operations teams and DevOps to use compiled code profiles across runs
• provides developers with APIs that control compilation policy, reduce de-optimization, and direct aggressive initialization tactics that minimize the need to warm-up the JVM
• optimized for 64-bit Linux on x86
• works with Zing’s Azul-optimized JIT compilers

In bullet-list form, briefly summarize up to ten (10) of the chief features and benefits of the nominated product or service.

Zing with ReadyNow! technology is in use in Java-based capital markets trading applications worldwide, and fits well for eCommerce, telecom, and online advertising platforms. Its chief benefits and features include:

Benefits
• Dramatically simplifies operational processes and reduces risks associated with “warming-up” Java-based financial systems
• Ensures consistent, peak performance during critical times of trading, market making, and risk analysis
• Reduces if not eliminates the need for synthetic test data or “warm-up” trades on live systems
• Further accelerates the adoption of Java for low-latency, real-time financial systems

Features
• Delivers more consistent low latency performance
• Allows Java applications to start up fast and stay fast
• Allows operations teams and DevOps to use compiled code profiles across runs
• Provides developers with APIs that control compilation policy, reduce de-optimization, and direct aggressive initialization tactics that minimize the need to warm-up the JVM
• Optimized for 64-bit Linux