Every day, new application workloads are being developed for online services, and many companies flounder because they’re unable to process growing data volumes at a sufficiently fast pace. Consequently, they may not always make the best and most informed decisions and may be missing out on the opportunity to generate new revenue streams, improve customer satisfaction or spot emerging trends. It’s especially true for the financial sector, where rapid data processing depends on access to high-speed analytics and the ability to crunch big data sets.
The financial-services industry has always been ahead of the technology curve, as it generally has the funds to purchase the latest solutions. Two decades ago, Internet development and large technology investments drove unprecedented advances for financial services. According to PWC, banks spent an estimated $215 billion on IT worldwide in 2014, including hardware, software and overall services, and that number has only grown since then.
Unsurprisingly, many financial companies are exploring cutting-edge Non-Volatile Memory Express (NVMe) storage systems to stay ahead of the performance curve and keep profit margins growing. Offering lower latency than previous flash-storage options, NVMe storage systems enable financial-services companies to process their data at high speed, delivering timely (and in some cases real-time) findings from their business-analytics software and directly improving their bottom line.
Real-Time Data Analytics
In his book The Flash Boys, Michael Lewis describes how in financial traders have spent large sums of money to increase network speed by just a couple of milliseconds. It proves how vital speed and reliability is to businesses, where the cost of downtime can translate to lost revenue. In the financial-services industry, downtime can have a dramatic impact on companies when fast transactions and information can make or break a business.
Real-time data analytics provides a near constant stream of data, allowing for immediate action on events, such as risk management, sales and trades. In finance, real-time analytics are critical to high-speed trading, fraud detection and transactions. Financial institutions that don’t adopt real-time analytics are putting themselves at a disadvantage by missing out on the ability to make rapid decisions.
In many cases, the analytics environment for financial firms can’t scale with the data set, especially if the firm has turned to servers with internal SSDs to meet their need for low latency. Adding more capacity to the existing architecture is extremely expensive and disruptive, which is why many companies are looking to high-performance NVMe over Fabrics solutions to meet their performance requirements without compromising on reliability, availability or scalability.
Processing Data at High Speed With NVMe Storage
Latency poses a major problem in the world of high-speed analytics and big-data management. Financial-sector data has a limited life span and quickly ceases to be relevant or actionable. With this fact in mind, management of big data sets and fast processing of business-analytics data is critical, as the time window for extracting the intelligence and then making accurate decisions is tight. If a financial-services company is experiencing high latency with legacy flash storage technology, it can lose millions of dollars in financial trading.
NVMe is a new interface protocol for flash storage and is designed to take latency out the data path, as it enables an exponential increase in the parallel access to flash media. It’s flexible and can be deployed in shared-storage systems over any supported high-speed network, such as Ethernet, InfiniBand or even legacy Fibre Channel, while maintaining the low latency that real-time workloads require.
NVMe storage solutions have shown strong performance on industry-standard benchmarks, with the fastest examples performing as fast as or even faster than servers with internal NVMe SSDs for financial workloads. This strong performance has fueled early adoption by financial firms, which require very low latency and strong performance. NVMe is on the path to becoming the de facto standard for solid-state media. According to G2M Research, the NVMe market will exceed $57 billion by 2020, with a compound annual growth rate (CAGR) of 95 percent.
IDC estimates that by 2020, up to 450 billion online business transactions will occur each day. With more automated transactions and trades happening in financial services, high-speed data processing is crucial for the most accurate decision making. Investing in high-performance NVMe storage behind high-speed networks and servers will greatly help in delivering smarter and quicker decisions.
Financial firms need to gain business insight by transforming available data into meaningful, useful and actionable information. Through NVMe storage, they’re monetizing data-driven insights and solving challenges in real time, which ultimately leads to faster growth for businesses and a stronger bottom line.
About the Author
Julie Herd is the director of technical marketing at E8 Storage. Julie has been a product-management leader in the storage industry for many years, spanning both hardware and software product management and marketing. Before joining E8 Storage, she led product-management teams at companies such as NetApp, Nexenta and BlueArc (acquired by Hitachi Data Systems). Julie started her storage career at Legato Systems (acquired by EMC) and earned an M.B.A. from Santa Clara University.