Scaling the New Bar for Latency in Financial Networks

By Brian Quigley

August 9, 2010

The bar for what qualifies as a fast connection or “low latency” networking has always been higher in finance than in other areas of corporate networking. It’s never been quite this high, however.

The increased use of innovative algorithmic-trading strategies has levied unprecedented pressure on financial firms to seek out and remove any possible delays that could threaten the successful execution of automated buy and sell orders. Significant time and resources are spent to achieve even modest, incremental improvements in latency at every point in the ecosystem of processes and systems that undergird algorithmic trading.

Managers of financial networks have found that the process of transporting data from one building to another is a common source of those delays. The fiber optic links between stock exchanges, alternative trading systems, colocation providers and information feeds all introduce additional latency into the system. Once, those amounts were deemed negligible — even in financial networking. Today, they are simply unacceptable.

Contemporary Trading’s Latency Intolerance

High-frequency trading (HFT) and other forms of algorithmic trading have emerged since the late 1990s. In all of them, a computer model has a predefined set of rules that automates the process of buying and selling. The computer receives various inputs from throughout the world (market data on price and volume, labor statistics, employment information, for example). The data is parsed and monitored in real time, and automated buy/sell decisions are executed according to how the model interprets the incoming intelligence.

Since the first trade to the market gets the best price, the delivery of a buy or sell order must be as fast as possible. Just a little more than a year ago, firms were concentrating on removing milliseconds from their network; today, a mere 250 nanoseconds make a difference.

The solution is not a matter of contacting the local phone company and deploying its most current high-bandwidth connection to the buildings that house an exchange. Even adopting the leading-edge, highest-speed model of router or multicore processor computer blades might not successfully address the financial world’s hypersensitivity to latency.

Traditionally, a firm has sought to improve its algorithimic trading processes by focusing first on the computer model itself. A lot of mathematicians have invested a lot of hours into tweaking the models that predict how the world is going to behave. Next, firms have sought to boost efficiency by optimizing the servers that process information against models, as well as the switches connecting those servers. More and more frequently, firms often locate or colocate clusters of computers in the financial exchanges themselves to root out proximity delays. Today, a firm’s various building-to-building transport connections are attracting attention from IT staffs.

It’s an entire ecosystem that must be optimized to wring as much latency as possible from algorithmic trading, and many financial firms now seek to take greater control over all of it.

From Buy to Build

Historically, when a financial firm required a connection between two buildings, it procured service from a phone company by ordering a circuit at a monthly fee, and the system would run through the phone company’s network, which typically would have some routing functionality for linking the two locations. Two key things have changed in the last few years.

One, algorithmic trading has crystallized scrutiny on the latency created every time traffic must pass through a router or take an indirect path to a central office. (It’s easy enough to quantify the sum of these delays; any member of the IT staff can look up the impact on the router.) Two, a combination of advances in optical technology that are specifically designed to eliminate latency, plus the lower costs and expanded availability of dark fiber, have enhanced a financial firm’s business case for building a dedicated infrastructure.

The performance gains are in the orders of magnitude — from 120 microseconds of latency for a carrier-provided service between buildings to 40 microseconds in a privately operated, optimized infrastructure. Service reliability can be enhanced with 24-by-7 monitoring predicated on the knowledge that 30 minutes of downtime to a financial firm can mean millions of dollars in lost revenue. The firm becomes more flexible to react and scale for new opportunities, as circuits can be turned up as needed. And it gains a path to continued innovation when partnering with vendors who understand the firm’s low-latency requirements and are under the gun to roll out enhancements to continually shave delays.

These developments have financial network managers asking, “Who can I partner with to sell me dark fiber, and what is the path that the fiber will take?” The ramifications of the answers are far-reaching. Financial firms must ensure that they are not sinking their money in a dark fiber route that snakes jaggedly under roads and up and down manholes, given the fact that eight inches of fiber can translate into a nanosecond of delay.

The next question is, “What is the latency of that path?” That answer demands an understanding of how optical transport works.

Getting into the Glass

There are four common optical networking functions where managers of financial networks often can realize game-changing efficiencies in latency budgets. When traffic beams across glass fiber, it encounters equipment that performs amplification, color conversion, dispersion compensation and regeneration. Each of these Wavelength Division Multiplexing (WDM) functions can be absolutely necessary to successfully carry out transport depending on the specific network environment, but each certainly also will inject some delay into the process. How much delay is something that financial managers must quantify and control if they are to squeeze every drop of latency from their end-to-end infrastructures.

  • Amplification — Optical networks frequently rely on Erbium Doped Fiber Amplifiers (EDFAs) to boost a traffic signal and offset the weakening that occurs across an optical span. Latency in the microseconds is the not-uncommon impact of the conventionally, widely deployed EDFA architectures. Consequently, financial networks supporting algorithmic trading must instead utilize emergent amplifiers optimized to yield lower latency.
     
  • Color conversion — In WDM optical networking, traffic is “transponded” — or, converted to a color of light — for delivery of a signal across a pair of glass fiber. Similarly, multiple colors of light are aggregated across a single fiber; multiple 10 Gbit/s services, for example, are “muxponded” into a single pair of fibers. Again, a financial network’s transponding and muxponding functions must be optimized for low-latency transport because conventional techniques such as Optical Channel Data Unit (ODU) encapsulation and thin film filters introduce too much delay.
     
  • Dispersion compensation — One of the ways that a traffic signal can degrade over the span of an optical link is “chromatic dispersion.” This is a phenomenon in which a signal effectively smears into a spectrum of hues, and it’s particularly common for data travelling at high speed. Installing kilometers of dispersion compensating fiber (DCF) has provided a remedy in some optical networks, but it’s not a wise approach in infrastructures supporting algorithmic trading. Latencies are too great. Optimized methods, like Fiber Bragg gratings, offer a low-latency alternative for counteracting chromatic dispersion.
     
  • Regeneration — Another method for preventing signal degradation over the course of a glass fiber connection is regeneration. Low-latency approaches to the function can help managers of financial networks claim significant efficiencies, because commonly deployed methods of regeneration produce considerable delay.

Conclusion

Contemporary finance is a race between markets. Firms are concerned with nanoseconds of latency in the processes and the systems that underlie algorithmic trading. Shaving eight inches of fiber equates to a one-nanosecond lead. Every element in a firm’s strategy is being evaluated for improvements, and optical fiber transport is targeted as a prime area for optimization.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

Advancing Modular Supercomputing with DEEP and DEEP-ER Architectures

February 24, 2017

Knowing that the jump to exascale will require novel architectural approaches capable of delivering dramatic efficiency and performance gains, researchers around the world are hard at work on next-generation HPC systems. Read more…

By Sean Thielen

Weekly Twitter Roundup (Feb. 23, 2017)

February 23, 2017

Here at HPCwire, we aim to keep the HPC community apprised of the most relevant and interesting news items that get tweeted throughout the week. Read more…

By Thomas Ayres

HPE Server Shows Low Latency on STAC-N1 Test

February 22, 2017

The performance of trade and match servers can be a critical differentiator for financial trading houses. Read more…

By John Russell

HPC Financial Update (Feb. 2017)

February 22, 2017

In this recurring feature, we’ll provide you with financial highlights from companies in the HPC industry. Check back in regularly for an updated list with the most pertinent fiscal information. Read more…

By Thomas Ayres

HPE Extreme Performance Solutions

O&G Companies Create Value with High Performance Remote Visualization

Today’s oil and gas (O&G) companies are striving to process datasets that have become not only tremendously large, but extremely complex. And the larger that data becomes, the harder it is to move and analyze it – particularly with a workforce that could be distributed between drilling sites, offshore rigs, and remote offices. Read more…

Rethinking HPC Platforms for ‘Second Gen’ Applications

February 22, 2017

Just what constitutes HPC and how best to support it is a keen topic currently. Read more…

By John Russell

HPC Technique Propels Deep Learning at Scale

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community. Read more…

By Tiffany Trader

IDC: Will the Real Exascale Race Please Stand Up?

February 21, 2017

So the exascale race is on. And lots of organizations are in the pack. Government announcements from the US, China, India, Japan, and the EU indicate that they are working hard to make it happen – some sooner, some later. Read more…

By Bob Sorensen, IDC

ExxonMobil, NCSA, Cray Scale Reservoir Simulation to 700,000+ Processors

February 17, 2017

In a scaling breakthrough for oil and gas discovery, ExxonMobil geoscientists report they have harnessed the power of 717,000 processors – the equivalent of 22,000 32-processor computers – to run complex oil and gas reservoir simulation models. Read more…

By Doug Black

Advancing Modular Supercomputing with DEEP and DEEP-ER Architectures

February 24, 2017

Knowing that the jump to exascale will require novel architectural approaches capable of delivering dramatic efficiency and performance gains, researchers around the world are hard at work on next-generation HPC systems. Read more…

By Sean Thielen

HPC Technique Propels Deep Learning at Scale

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community. Read more…

By Tiffany Trader

IDC: Will the Real Exascale Race Please Stand Up?

February 21, 2017

So the exascale race is on. And lots of organizations are in the pack. Government announcements from the US, China, India, Japan, and the EU indicate that they are working hard to make it happen – some sooner, some later. Read more…

By Bob Sorensen, IDC

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

Azure Edges AWS in Linpack Benchmark Study

February 15, 2017

The “when will clouds be ready for HPC” question has ebbed and flowed for years. Read more…

By John Russell

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

For IBM/OpenPOWER: Success in 2017 = (Volume) Sales

January 11, 2017

To a large degree IBM and the OpenPOWER Foundation have done what they said they would – assembling a substantial and growing ecosystem and bringing Power-based products to market, all in about three years. Read more…

By John Russell

US, China Vie for Supercomputing Supremacy

November 14, 2016

The 48th edition of the TOP500 list is fresh off the presses and while there is no new number one system, as previously teased by China, there are a number of notable entrants from the US and around the world and significant trends to report on. Read more…

By Tiffany Trader

Lighting up Aurora: Behind the Scenes at the Creation of the DOE’s Upcoming 200 Petaflops Supercomputer

December 1, 2016

In April 2015, U.S. Department of Energy Undersecretary Franklin Orr announced that Intel would be the prime contractor for Aurora: Read more…

By Jan Rowell

D-Wave SC16 Update: What’s Bo Ewald Saying These Days

November 18, 2016

Tucked in a back section of the SC16 exhibit hall, quantum computing pioneer D-Wave has been talking up its new 2000-qubit processor announced in September. Forget for a moment the criticism sometimes aimed at D-Wave. This small Canadian company has sold several machines including, for example, ones to Lockheed and NASA, and has worked with Google on mapping machine learning problems to quantum computing. In July Los Alamos National Laboratory took possession of a 1000-quibit D-Wave 2X system that LANL ordered a year ago around the time of SC15. Read more…

By John Russell

Enlisting Deep Learning in the War on Cancer

December 7, 2016

Sometime in Q2 2017 the first ‘results’ of the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) will become publicly available according to Rick Stevens. He leads one of three JDACS4C pilot projects pressing deep learning (DL) into service in the War on Cancer. Read more…

By John Russell

IBM Wants to be “Red Hat” of Deep Learning

January 26, 2017

IBM today announced the addition of TensorFlow and Chainer deep learning frameworks to its PowerAI suite of deep learning tools, which already includes popular offerings such as Caffe, Theano, and Torch. Read more…

By John Russell

HPC Startup Advances Auto-Parallelization’s Promise

January 23, 2017

The shift from single core to multicore hardware has made finding parallelism in codes more important than ever, but that hasn’t made the task of parallel programming any easier. Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Leading Solution Providers

CPU Benchmarking: Haswell Versus POWER8

June 2, 2015

With OpenPOWER activity ramping up and IBM’s prominent role in the upcoming DOE machines Summit and Sierra, it’s a good time to look at how the IBM POWER CPU stacks up against the x86 Xeon Haswell CPU from Intel. Read more…

By Tiffany Trader

Nvidia Sees Bright Future for AI Supercomputing

November 23, 2016

Graphics chipmaker Nvidia made a strong showing at SC16 in Salt Lake City last week. Read more…

By Tiffany Trader

BioTeam’s Berman Charts 2017 HPC Trends in Life Sciences

January 4, 2017

Twenty years ago high performance computing was nearly absent from life sciences. Today it’s used throughout life sciences and biomedical research. Genomics and the data deluge from modern lab instruments are the main drivers, but so is the longer-term desire to perform predictive simulation in support of Precision Medicine (PM). There’s even a specialized life sciences supercomputer, ‘Anton’ from D.E. Shaw Research, and the Pittsburgh Supercomputing Center is standing up its second Anton 2 and actively soliciting project proposals. There’s a lot going on. Read more…

By John Russell

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

IDG to Be Bought by Chinese Investors; IDC to Spin Out HPC Group

January 19, 2017

US-based publishing and investment firm International Data Group, Inc. (IDG) will be acquired by a pair of Chinese investors, China Oceanwide Holdings Group Co., Ltd. Read more…

By Tiffany Trader

Dell Knights Landing Machine Sets New STAC Records

November 2, 2016

The Securities Technology Analysis Center, commonly known as STAC, has released a new report characterizing the performance of the Knight Landing-based Dell PowerEdge C6320p server on the STAC-A2 benchmarking suite, widely used by the financial services industry to test and evaluate computing platforms. The Dell machine has set new records for both the baseline Greeks benchmark and the large Greeks benchmark. Read more…

By Tiffany Trader

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

Intel and Trump Announce $7B for Fab 42 Targeting 7nm

February 8, 2017

In what may be an attempt by President Trump to reset his turbulent relationship with the high tech industry, he and Intel CEO Brian Krzanich today announced plans to invest more than $7 billion to complete Fab 42. Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This