Azul Systems Partners with Solarflare for Latency and Jitter Solution

Azul Systems a provider of robust technology solutions for firms working in Java based environments has entered into a partnership with Solarflare, a leader in application-intelligent 10 Gigabit Ethernet (10GbE) networking software and hardware as the companies work together to reduce end-to-end Java application latency and eliminate jitter.

Azul Systems has been providing firms with its Azul Zing™ Java Virtual Machine (JVM), a solution to reduce jitter and an advance garbage collector for high frequency fast response trading systems. The collaboration means Azul and Solarflare will improve throughput, lower latency and eliminate jitter for Java applications. In fast moving markets not all JVMs can keep up with object allocation rates under higher transaction volumes. This can lead to delayed trade processing and transactions clogging up consequently impacting the Execution in trading strategies. Azul’s Zing works ideally with Solarflare's OpenOnload product thus reducing latency by 18% and eliminating max outliers by seven times.

"Azul's and Solarflare's best-of-breed products have been tested to ensure they work together seamlessly," said Bruce Tolley, vice president, solutions marketing at Solarflare in a statement. "We've taken all the necessary steps to deliver exactly what our customers require to do business in today's fast-paced business environment. And although our focus at this time is on capital markets, customers from a wide variety of industries can profit from our collaboration with Azul," added Mr Tolley.

With high frequency trading accounting for nearly 70% of trading volume in the USA and an estimated 35% in the UK, trading participants are constantly exploring low latency trading solutions in the arms race to find alpha.

"Our valued customers in the capital markets see incredible advantages in a collaboration between Azul and Solarflare," said Scott Sellers, CEO and founder of Azul Systems in the official press briefing. "This partnership was formed in direct response to that vision, and is a strong example of the value technology vendors can create by working together for the benefit of our common customers."

Trading terminals and applications are built-in Java as banks, brokers and hedge funds look for Scalability in the development process. Java built applications provide firms with key advantages relating to the development and maintenance of the software, however this comes with a price, jitter. Java developers have tried some techniques to rectify the issues of latency and jitter, which include; building their own object pools, avoiding allocation and writing code in the java syntax.

Firms active in the FX markets have been using Azul’s Zing to overcome some of the latency related issues, LMAX, the first FX MTF partnered with Azul earlier this year.