Azul Systems the award-winning leader in Java runtime solutions, and Pontus Networks, a specialist in improving the performance of software in financial institutions, have launched an Azul Zing®-based edition of PontusVision Thread Manager (PVTM), Pontus’s flagship product which reduces capital and operational IT infrastructure expenditures by improving the way software threads run on specific CPU cores. In a recent benchmark study, the combination of Zing and PVTM helped an investment bank improve the latency of its FX pricing system without changing a single line of code, by reducing the pricing system’s peak latency from receipt of a raw market data tick to sending a FIX market data snapshot to clients by 20X.
Both buy- and sell-side organizations require predictable behavior in their systems. If a sell-side market maker has a predictably slow system, the prices can be adjusted by applying the appropriate level of spreads to prevent the bank from being targeted because they are providing stale prices. If a pricing system has lower average latency, but higher peak latencies than a predictably slow system at a competitor, the sell-side organization can easily wind up randomly sending inaccurate prices to the market, and be cherry-picked by a nimble buy-side organization.
Azul Zing is often an appropriate choice for Java-based applications and systems that require consistent low latency and/or high scalability. Zing eliminates the operational interruptions and response time outliers often experienced by Java-based trading, risk or compliance applications across a wide variety of asset classes. Zing’s ReadyNow!® technology also solves Java’s warm-up problem by allowing applications to start up fast (and fully optimized) at the opening bell and stay fast even as market conditions change throughout the day.
Pontus’ PVTM was designed to help financial institutions reduce the need for weeks of performance analysis by highly-skilled Java engineers. PVTM is an easy to deploy solution that accelerates software by optimizing the execution layout of threads within production servers. PVTM improves the performance of applications -- improving latency in FX pricing systems and delivering better throughput for batch- and Database-centric workloads. The benefits of deploying PVTM include up to 270% performance gains, plus the ability to improve performance in minutes instead of weeks -- and without hardware or code changes.
Leo Martins, Founder at Pontus Networks, said: “In a nutshell, Thread Manager reduces capital and operational expenditure by solving a performance problem in a minute as opposed to a month. By offering clients a Zing edition of our solution, we not only guarantee lower latencies, adherence to SLAs and a twenty fold improvement in performance, but crucially we can help eradicate the industry-wide issue of arbitrage. Moving forward, we are looking to work with Azul in the big data space and solve industry-specific issues there as well.”
Scott Sellers, CEO of Azul Systems, said: “Pontus’s decision to work with Azul on developing a joint solution highlights the importance of using the right JVM for Java-based FX applications that require predictable low latency and rapid scalability. In today’s ultra-competitive financial markets where resources are stretched, having a proven joint solution which can be deployed rapidly with zero changes to existing code or hardware offers a great advantage to both buy-side and sell-side institutions.”