The demand for cost-effective high-performance cryptography solutions for data-centers continues to grow under the weight of almost constant threats of hacking, both strategic and malicious. At the present time, asymmetric cryptographic solutions and bare fiber links form two of the most critical weak points for data-center security. Both of these security incursions could be defended against by deploying adequate quantum key exchange (QKD) solutions.
But as CIR sees things, several factors are holding back QKD deployment in the data center and suppliers of QKD systems must address these issues if they are ever going to tap into the data center market; an addressable market that is larger in volume terms than these suppliers have ever had available to them before. The most obvious problem is that QKD firms have not yet come up with QKD solutions that would be affordable to most data center managers.read more
400 GigE is the first Ethernet standard not to incorporate copper cable. There is some talk about implementing 400 GigE for ultra-short reach links on a board or even board-to-board. But for the typical data center applications – rack-to-rack and beyond – it’s fiber all the way. An open question is how will the fiber be implemented? The options here being field termination and active optical cables.read more
While polymer-based data communications has found viable niches in home networking and in the automotive industry, there has been no success in bring polymers in the data center. Polymer datacom start ups have risen and fallen. One recalls, especially, Telephotonics, which had very ambitious plans to create a full range of polymer-based optical components for the telecommunications industry. Large specialty chemical firms – including Dow, DuPont and Dow Corning -- have promised polymer optical components but said promises have not produced much.
CIR believes that this is about to change and polymer photonics is finally emerging as a viable solution for data centers. Whereas the old polymer photonics was mostly focused on plastic optical fiber (POF), the new approach is all about polymer waveguides. And while the old “polymer photonics” was technology in search of an application, today’s polymer photonics lowers the cost of handling “big data” in routers, switches and data centers which makes for an actual business case.read more
For now 400 GigE is barely in its deployment phase, and its use will be confined to hyperscale data centers at first. But we have all seen how the fastest data rates make their way downward. It is not that many years since 10 Gbps was considered “bleeding edge.” Today some laptops have 40 Gbps (Thunderbolt) interfaces. So it may not be many years before the industry must contend with the cabling infrastructure opportunities that 400 Gbps will bring in its wake.
Cabling opportunities change at a slower pace than data rates. When data center managers install a new cabling infrastructure, they want assurance that it will survive several generations of networking protocols. Data centers are reluctant to abandon old cabling technologies. Even though the death of copper has been proclaimed in the data center since the 1990s it hasn’t come close to happening yet What has happened is that copper cabling has improved its packaging – CAT 7 is something a lot more than the old “telephone wire” – and its price has increased. But copper has endured so far.
CIR believes that there are two emerging opportunities that have emerged as the result of increasingly noticeable vulnerability of fiber optics. These are (1) Improved intrusion detection and (2) quantum encryption.read more
The growth of data centers is a constant theme in the trade press. Typically, this growth is presented as an encouraging trend, creating demand for a variety of systems and components. But for data center managers hyperscale data centers are mostly cost centers; necessary evils that require cost reduction and cost analysis, and do not produce profits for their owners.
All this has consequences for optical component manufacturers. These implications have both positive and negative implications for the component sector. On the one hand, the cost sensitivity of data centers points towards the need for plain vanilla standardized components that are hard to distinguish in the marketplace. On the other hand – and somewhat paradoxically – CIR believes that it also indicates a new and potentially profitable direction for components companies in the form of what we shall call “smart optical components.”read more
After the relative failures of the Thunderbolt 1 and Thunderbolt 2 cabling systems, Thunderbolt 3 appears set for success as the “one cable to rule them all.” Thunderbolt 3 is billed as a user-friendly cabling system, robust enough to support everything from high-definition video to high-res audio to personal storage networking. It operates at 40 Gbps, twice the data rate of earlier Thunderbolts. Thunderbolt 3 also supports the most popular consumer electronics interfaces including the USB Type-C connector, USB 3.1, DisplayPort 1.2 and HDMI 2.0. As an aside, some insiders are now talking about an 80 Gbps Thunderbolt 4 at some time in the not-too-distant future.
The proof that Thunderbolt 3 is catching on is that many important laptop OEMs are now adopting Thunderbolt 3. In the past the Thunderbolt reality has been an Intel-Apple one. Thunderbolt was born at Intel at the end of the last decade, but despite Intel’s early high hopes, the only big computer/consumer electronics firm to have implemented the earlier versions of Thunderbolt was Apple. Thunderbolt 1 and Thunderbolt 2 did not see much (or indeed any) love at the Windows/PC companies. But this has changed with the advent of Thunderbolt 3. Acer, Asus, Dell, HP, and Lenovo began to show real interested in Thunderbolt in 2015 and all have high-profile Thunderbolt 3 capable laptop models on the market today.
Meanwhile, Apple has gradually strategically positioned Thunderbolt to be its key interface. In some Apple products, the Thunderbolt port is the only port in the box.read more
Active optical cable (AOC) suppliers are spreading their wings into video and consumer electronics markets, but most will remain focused on the data center – especially the HPC environment – for years to come. In CIR’s opinion, AOC suppliers have good reasons to be optimistic in their pursuit of dollars from the data center. Higher data rates continue to lead managers to install more fiber. Multicore CPUs and many-core GPUs (graphics processing units) make the boundaries between HPC and more conventional forms of computing become fuzzier.
There can be little doubt that most of the strategic focus of the AOC manufacturers will continue to be on the data-center market for the next few years. Here they will find a relatively mature market and a fairly clear path for further penetration of the market. Thus we expect to see a growing range of AOCs equipped with the appropriate IB, Ethernet, and Fibre Channel connectors. However, the market for AOC sales outside of the data center is looking increasingly attractive for a number of reasons.
In 1990s fantasies of the all-optical network, electronics played a minimal role. Today it seems that the higher bandwidth demands go the more high-performance silicon is actually needed.read more