Back to Blog

Asena Woodward

Ever Thought About Competing on Latency? Try Cutting It by 37%. (New Report & Infographic)

Mahwah, New Jersey, located just over 30 miles from Wall Street, is home to a 400,000 square foot fortress: the NYSE Euronext data center. Fun fact: The town’s name originates from the Lenape word “wawewi,” which means “Place Where Paths Meet.” It’s an apt name, even if in 2015 we’re talking virtual paths meeting.

The action has migrated from the physical trading floor to the virtualized data center. Within the facility NYSE has set aside 60,000 square feet as a colocation space for hedge funds and investment firms engaging in high-frequency trading (HFT). It offers “the ability to collocate equipment in an environment designed for continuity, high security and fairness.” In the interest of this fairness, and per the dictates of the US SEC, every fiber optic cable connecting the colocation customers’ hosts to the network is exactly the same length. What’s interesting is that this hasn’t leveled the playing field so much as moved the field elsewhere—these companies subsequently do everything they can to reduce latency within the host. (You can read more about the “micro-second market” here.)

Technology, not the trade itself, is the competitive edge.

The Playing Field Has Expanded into the Virtual World

The High Cost of Latency - smallSure, the high-stakes world of finance, especially high-frequency trading, is an extreme example of the impact of technology on the bottom-line, but it is a real one.

We no longer live in a world where the competitive edge is isolated to the product or service—every transaction, process, and interaction in the supply chain contributes to a business’s competitive edge. It has long been true in the physical world—think Amazon and the lengths they’ve gone to give their customers almost instant gratification—and how addictive that can be.

The same holds true in the virtual world and the major culprit is latency. Here too Amazon has a story: every 100ms of latency cost 1% in sales. In terms of Amazon’s 2014 revenue of $88.94 billion, 1% is $889,400,000. On that kind of scale, the project laying the first ever trans-Arctic Ocean fiber optic cables is practically a steal at $1.5 billion, cutting latency between London and Tokyo by 30% (from 240 to 170 milliseconds).

Skip to Infographic? The High Cost of Latency (and what to do about it)



Latency Doesn’t Care How Big or Small is Your Organization

While most companies are operating on smaller budgets, latency is still a critical issue. VMTurbo Founder, Shmuel Kliger, calls latency the “bane of the software-defined universe”:

“Latency, however, is poised to become the bane of the software-defined universe. For as more and more is abstracted into software, the less we can keep track of the physical location of workloads…. The result of which might not only be slow response time, but in a world of gigabit and beyond switches where a second of packet loss is larger than War and Peace on Blu-Ray, might lead to application failure – ever lost a Skype call before? Imagine something much more mission critical.” (emphasis own)

So on the one hand latency sucks profits and on the other it can cause straight up failure in your mission-critical applications. The profit and risk implications make latency a concern for organizations of any size or budget—that includes yours.

Principled Technologies Report: Using VMTurbo Cuts Latency by 37%

Principled Technologies recently evaluated the impact of VMTurbo’s demand-driven control system on virtual and cloud environments. It deployed VMTurbo into a DRS-enabled VMware vSphere ESXi 5.5 cluster designed to represent a typical unit of deploying mission-critical applications within enterprise data centers. Principled Technologies then began executing VMTurbo’s recommendations for driving the environment to a perpetual state of health. The recommendations included decreasing vCPU allocation for SQL Server VMs, allocating additional memory to increase performance and moving certain VMs to more responsive storage. In doing so, Principled Technologies found that VMTurbo decreases latency by 37%.

principled-technologies-vmturbo-repor-smallClick to Read Principled Technologies Report

What’s going on “under the hood”? VMTurbo combats latency by assuring application components get the compute, storage and network resources they demand to perform. It does this whatever the size of your environment or the scale of your applications.


Don’t Let Latency Stand; Minimize It to Increase Your Competitive Edge

Lots of things contribute to competitive advantage. But, given latency’s make-or-break impact, doing something about it can only help. Reports and infographics are all well and good, but do you know how latency is affecting your organization? What would a 37% reduction in latency mean for you? What would it mean for your organization’s bottom-line?

Think on it…

Hopefully now you’re thinking it might be a good time to try VMTurbo out for yourself.