186,000 Miles Per Second. Not Just a Good Idea, It's The Law

I was somewhat chagrined last week when I saw that my friends at SliconAngle ran a blog post endorsing another blogger's brilliant idea that all we had to do to reduce intercontinental WAN latency was to adjust the speed of light. To give them some credit for high school physics, they weren't referring to changing one of the principle constants of the universe (although they did invoke Star Trek's all-powerful Q). Instead, they were implying that faster fiber optic cable was around the corner.

Howard Marks

November 4, 2010

3 Min Read
NetworkComputing logo in a gray background | NetworkComputing

I was somewhat chagrined last week when I saw that my friends at SliconAngle ran a blog post, Boosting the Web: Fiber Optic Cables and the Speed of Light, endorsing blogger Denton Gentry's brilliant idea, laid out in Toward A Faster Web: Increase the Speed of Light, that all we had to do to reduce intercontinental WAN latency was to adjust the speed of light.  To give them some credit for high school physics, they weren't referring to changing one of the principle constants of the universe (although they did invoke Star Trek's all-powerful Q). Instead, they were implying that faster fiber optic cable was around the corner.

Those of us who actually studied science may remember from our college years that the speed of light in a vacuum is a constant. When light passes through matter, such as silicon, interactions between the light photons and the electromagnetic fields of the matter itself slow down the photons. A transparent substance's refractive index describes how much slower light travels in that substance than in a vacuum. The chemically doped, fused silica core of today's fiber optic cable has a refractive index of 1.48 so light travels in it 67.57 percent as fast as in a vacuum. In his original post, Gentry says if we could just make fiber with a 10 percent lower refractive index, then a ping from San Francisco to Tokyo could take 90ms instead of the 100ms it takes today. He also says "I admit that I have absolutely no understanding of the chemistry involved in fiber optics."

I studied chemistry. I can tell you it ain't gonna happen. And as someone that's been intimately involved in WAN applications for twenty years, the difference between 100ms and 90ms doesn't matter. Latency is important, and the high-performance trading guys on Wall Street worry about every microsecond, but from a user point of view the difference between 90 and 100ms isn't worth worrying about. To run well with 90ms of latency you need to eliminate TCP and any other chatty protocols and keep everything you can local. It's no different for 100ms.

So let's look at the idea of lower index of refraction (IR) fiber. To get the 10 percent reduction our friends are looking for, we'd need to push the index of refraction down to that of pure water (1.33). That's well below the IR of any common solid. IR is related to the interactions of photons with the material they pass through, thus the IR is tangentially related to density. A material with an IR of 1.33 would probably have to be less dense than today's fused silica, which is more than twice as dense as water.

Even if we had such a material (call it unobtianium-tetranaquadide) that has a 1.3 IR, it would be less dense and therefore less strong than fused silica. Such fibers would need all sorts of reinforcement to make cables for the bottom of the ocean. Some thin film coatings approach 1.36 IR, but remember that the cladding of an optical fiber has to be lower than the core as well, so to get a 1.33 core we'd need a 1.29 or so cladding material--all of which need to be at least as transparent as fused silica. I'm not waiting for the speed of light to get faster.  Better to work on the software.

About the Author

Howard Marks

Network Computing Blogger

Howard Marks</strong>&nbsp;is founder and chief scientist at Deepstorage LLC, a storage consultancy and independent test lab based in Santa Fe, N.M. and concentrating on storage and data center networking. In more than 25 years of consulting, Marks has designed and implemented storage systems, networks, management systems and Internet strategies at organizations including American Express, J.P. Morgan, Borden Foods, U.S. Tobacco, BBDO Worldwide, Foxwoods Resort Casino and the State University of New York at Purchase. The testing at DeepStorage Labs is informed by that real world experience.</p><p>He has been a frequent contributor to <em>Network Computing</em>&nbsp;and&nbsp;<em>InformationWeek</em>&nbsp;since 1999 and a speaker at industry conferences including Comnet, PC Expo, Interop and Microsoft's TechEd since 1990. He is the author of&nbsp;<em>Networking Windows</em>&nbsp;and co-author of&nbsp;<em>Windows NT Unleashed</em>&nbsp;(Sams).</p><p>He is co-host, with Ray Lucchesi of the monthly Greybeards on Storage podcast where the voices of experience discuss the latest issues in the storage world with industry leaders.&nbsp; You can find the podcast at: http://www.deepstorage.net/NEW/GBoS

SUBSCRIBE TO OUR NEWSLETTER
Stay informed! Sign up to get expert advice and insight delivered direct to your inbox

You May Also Like


More Insights