The realities of achieving ultra low latency in the network and the combination of solutions, such as connectivity and proximity, that are needed to address the challenges facing financial services organizations.
By Felipe Alvarez
When it comes to financial services networks, low-latency is king. In a recent survey, IT professionals across multiple industries were asked what is the most important quality in a network. For respondents from the financial services industry low latency was the single most important quality in a network, outpacing reliability and network performance. In short - time is money.
The survey also revealed that more than 90% of respondents consider their network provider to be a critical component of their business. Of these, only 21% felt that their network offered them a distinct competitive advantage. However, the vast majority of those who felt their network was in fact a competitive advantage were from the financial services industry. In no other industry is the need to connect fast and first more critical than in financial services. In this industry, IT decision makers understand the importance of partnering with a proven expert in the industry that understands how to deliver speed and performance.
The State of the Industry
The state of the financial services market is dynamic. It is an industry focused on microseconds as every financial transaction has direct impact on business. In no other industry do we see so clearly the direct correlation of quality and quantity on the bottom line of a business. In fact, according to Information Week, a millisecond advantage in trading applications can be worth over $100 million a year to a major brokerage firm.
Trade volumes are increasing. According to Bloomberg, equities trading averaged 9.75 billion shares per day in April 2010, compared with 8.27 billion shares in March 2010 – an increase of 18 percent in 30 days.
The decentralization of trading markets is a growing trend in the US equities market. The number of markets has risen to more than 50, including at least 30 dark pools. The primary exchanges continue to cede market share to alternative trading systems. Volume transacted away from exchanges has risen to 31.1 percent from 27.6 percent in October 2009, according to data compiled by Barclays and Bloomberg.
In the last decade, automation has taken hold and is growing rapidly. Electronic Computer Networks (ECNs) are facilitating trades outside of exchanges and are gaining market share. Among exchange operators, NYSE Euronext had the largest overall equities market share with 26.6% of trading, followed by Nasdaq OMX with 22.1%. ECNs Direct Edge had 10.8% and Bats Exchange had 9.5% in April 2010, according to Barclays data.
The decentralization, growth in automation and demand for speed leads to certain challenges in network connectivity – including access, proximity, security,
latency and scalability.
Challenges
Lowering latency continues to be the greatest challenge of financial services institutions. Service providers are continuously looking at building and piecing together segments to existing routes between major financial centers and key exchanges and datacenters in order to offer lower latency services - reducing the mileage and straightening the route to exceed current thresholds.
At the same time, network connectivity requirements are expanding and can be to any combination of the primary exchanges, ECNs and/or colocation and datacenters. Proximity to these locations directly impacts latency and performance.
As ECNs emerge and gain market share, connectivity to these exchanges will be expected wherever they emerge. Financial services institutions need to be able to quickly access reliable connectivity to new locations. Superior network flexibility is required to meet this decentralization.
Scalability becomes a challenge, as many current networks were not architected to efficiently scale as automated trading volumes grow. With trade volumes growing anywhere from 30 to 110% year over year, a service provider’s provisioning intervals can help or hinder an exchanges time to market.
The security and reliability of a network is always a challenge and one that increases with the decentralization of the trading exchanges and the increased complexity of the network.
Addressing security, flexibility, and scalability challenges must be done in a manner that does not impede transaction speeds.
The Winning Combination
With all of these challenges, what is the optimal combination of solutions to achieve optimal business performance and gain a competitive advantage?
The answer: A network designed and purpose built to meet all of the challenges seen by the financial services industry – low latency, proximity, and connectivity. Only addressing one or two of the challenges leaves the network open to the introduction of latency or other challenges.
The lowest latency networks will offer any-to-any connectivity across the network, with direct connections between locations that do not carry traffic back to a central routing hub. The network should have the capability of transporting multiple high Gig channels and utilize cutting-edge technology that dramatically reduces per-node latency while at the same time addressing scalability, ease of provisioning and deliver time to market. The network needs to be architected with technology that enables software provisioning of services in hours or days not weeks or months. In addition, the network needs to be reliable and fault tolerant. It is simply not acceptable to sacrifice these qualities to gain low-latency. A financial service’s network should require full performance guarantees.
The next step to reduce route mileage is changing the site proximity to the key financial services locations. Colocation and datacenters are a key solution to shorten the route and, as a result, centers are being designed with new considerations that address the financial services industry. Space management is taking into consideration the need to provide equal or greater computing power with less space. Financial services applications are utilizing high-performance computing and increased transaction processing. The trend of blade servers and high-density rack configuration are being adopted. High-density colocation is a design philosophy that is modeled for variable demand. The design focuses on zones, not space, and accommodates high consumption. Cooling is done in hot aisle containment zones that neutralize the heat more than cooling, supporting a denser configuration of equipment per cabinet. For the financial services industry even inches introduce latency.
Financial institutions need the flexibility and connectivity to key exchanges and datacenters of their choice. When choosing a network it needs to have the optimal site connectivity and to accommodate high-growth locations, connecting colocation facilities, datacenters, and financial exchanges as well as major financial centers.
It is essential that service providers for the financial services industry understand that the winning combination for superior network performance includes:
• ‑Full-performance, guaranteed, ultra low latency networks;
• Connectivity to leading financial centers;
• ‑Connectivity and proximity to leading financial institutions and high profile exchanges; and
• ‑State-of-the-art colocation facilities that offer site proximity, connectivity and support for high performance computing.
This strategic combination has proven to be successful for financial services businesses – delivering a significant competitive advantage and superior business performance.
Felipe Alvarez, President & COO, Sidera Networks, email: Felipe.Alvarez@sidera.net; web: www.sidera.net.
Reach Wall Street's leading technology products and services in the financial industry.
TICKER Editorial Calendar Deadlines, Themes & Suggested Content