Finisar at Interop Las Vegas

Next week, Finisar will be heading to Las Vegas for Interop 2009. From the looks of the conference agenda, this year’s show is shaping up to be a good one.

We’ll be there, showcasing our active optical cable line and talking about how these interconnects can help reduce energy consumption and take the heat out of datacenters. We’ll also be checking out some panels and have our ears tuned for the latest trends in networking.

If you’re going to be in the neighborhood, feel free to stop by the Finisar booth (#2612) and say hi. In fact, the first 25 visitors who mention this blog will get a snazzy Quadwire car.

Look forward to seeing you there.

Active Optical Cables and the truce between Optics and Copper

This week’s guest blogger is Jan Meise. Jan is responsible for Strategic Marketing at Finisar and his work in strategic marketing allows him to bring a very interesting perspective to some of the new trends in active optical cables. I’m hoping Jan will be a regular guest blogger on the site.
*************************

Almost out of nowhere and with little time for everyone to digest, a new market segment emerged in mid-2007 in the optical component industry called Active Optical Cables (AOCs).

Put simply, an AOC has an electrical interface at both ends. The electrical to optical conversion is embedded into the cable wire harness and uses fiber optics cables as the transport medium.

And so new is the Active Optical Cable segment that analysts are still struggling to build a consensus forecast. While Lightcounting is estimating a 2010 AOC market of some $14m, IGI is two logarithmic steps more aggressive and sees potential for $1.1bn revenue in the same year.

Beyond revenue expectations, Active Optical Cables have the great potential to end the war between Optics and Copper. As Karen Liu of Ovum outlined in her article, there will be ‘universal ports’ for copper and optics allowing both to ‘help the total market’ as synergistic “frenemies”.

In this new post war era, systems are thereby using common host connectors enabling Data Center customers to deploy copper and/or active optical cables based on needed cable lengths and diameters, bend radii and of course, cost.

By moving the signal conditioning chips from the host board into the cable ends, electrical interfaces on the host board can be simplified, ultimately keeping the overall system cost and power need to a minimum.

Being able to choose among passive copper, active copper and active optical cables, end-users pay and consume power only according to their specific need.

While first concepts of Active Optical Cables were originally developed in the 1990s, the commercialization of those cables started in 2007 with products for InfiniBand, 10GbE, DVI and HDMI applications.

Finisar introduced its first two families of Active Optical Cables, Laserwire™ and Quadwire™ in 2007 and 2008, respectively. While currently mostly focused on 10GbE and InfiniBand, we have seen adjacent markets for those families all the way from industrial to consumer applications.

In subsequent posts I will take a closer look at those various markets but, in the meantime, you can check out the product briefs for Laserwire and Quadwire.

I invite your comments on Active Optical Cable technology, or even better, give me your best shot for the AOC 2010 revenue forecast! You can also follow me on Twitter.

I will be demonstrating Finisar’s cable products at Interop next week. If you would like to chat in person come by the Finisar booth #2612 anytime.

CML and EML see eye to eye

We believe that CML™ (Chirp Managed Lasers); can be a very complementary product (and in some cases even a substitute product) to EML (Electro-Absorption Modulated Lasers).

EMLs have been used as a primary transmitter type in many 10Gb/s module applications. Modules such as the current Finisar 40km/80km TDM and DWDM XFP today use EMLs.

CMLs have slightly different characteristics than EMLs. Specifically, CMLs have back-to-back eye diagrams with lower mask margin than EMLs. On the flip-side, CMLs offer higher output optical power, and much better performance after transmission through fiber.

However, the key point is that CMLs are fully interoperable with EMLs.

In fact, our Chief Scientist Daniel Mahgerefteh published a paper in Lightwave in November 2008 dedicated to this topic. The article, aptly named “CML and EML see eye to eye,” discusses how the two transmitter types are fully interoperable.

So, although CMLs may not win the beauty contest when looked at under a scope, CMLs will enable better performance over fiber. As Daniel so eloquently put it, “Using mask as a measure of performance is similar to trying to determine the speed of two race cars by comparing how clean their engines appear rather than by which of them crosses the finish line first.”

In addition, CML-based XFPs have very nice features not offered by EML:
• CMLs can work on 50GHz ITU grids (currently EMLs are only available on 100 GHz grid);
• CMLs offer 4x100GHz/8x50GHz narrow tunability;
• CMLs support L-band as well as C-band (EMLs only currently support C-band); and
• Inherent immunity to SBS (Stimulated Brillouin Scattering)

What are your thoughts on the differences between CMLs and EMLs? Feel free to let us know in the comments below.