Alcatel SA (NYSE: ALA; Paris: CGEP:PA) has released a new long-haul DWDM product, manifesting a slow but sure migration to all-optical core networks by carriers (see Alcatel Releases Core DWDM Gear).

The key word here is "slow." Alcatel and its competitors say that despite market reports that core spending's as dead as a dodo, carriers do want more core capacity, particularly in areas such as the Asia/Pacific region and Eastern Europe, where there was less boom-time overbuild than elsewhere.

What's more, carriers haven't given up their ambitions for all-optical cores, the vendors say. It's just that they're not moving quickly to install them. Instead, carriers want to take as long as they can and make as few changes as possible during the optical migration.

The new Alcatel 1626 Light Manager is a case in point. The product offers more capacity than Alcatel's older DWDM core products. To wit: The 1626, when it ships in the first quarter of 2004, will support up to 96 C-band channels at rates to 10 Gbit/s over distances up to 2,100 kilometers. In a future release (commercial availability unspecified), the 1626 will support 192 channels at distances to 4,500 km.

Compare the 1626 with Alcatel's existing 1640 DWDM system, which supports up to 80 channels at rates to 10 Gbit/s at distances to 1,400 km. For ultimate contrast, compare the 1626 with the Alcatel 1686, which supports just 32 channels at 600 to 700 km.

The 1626 is also Alcatel's first long-haul DWDM system to have a reconfigurable optical add/drop multiplexer (ROADM), seen as a key element of future-proofing carrier cores (more on that below).

Alcatel isn't emphasizing the ability of the 1626 to replace its older gear, however. And news of the ROADM is being downplayed. Instead, Alcatel is pitching the 1626 as an incremental stepping-stone to better optical cores. Alcatel's selling the 1626 transponder separately from the ROADM as an "extender shelf," for instance, to increase the performance of the 1640 and 1686.

According to Sterling Perrin, senior research analyst at IDC, Alcatel probably won't hesitate to pitch the next-gen 1626 to carriers requiring greenfield networks. Its ROADM, he says, can be viewed as a kind of incremental two-way optical switch, giving customers a boost toward all-optical networking.

But to those who already have Alcatel gear, it's worth it for Alcatel to give their customers a choice to move ahead in a piecemeal fashion, Perrin says. The message carriers have sent is loud and clear: Suppliers who move too fast risk getting replaced. And core DWDM revenues are still significant enough to make Alcatel and others want to stick around. "Protecting their installed base is important," he says.

Table 1:

The bottom line? Alcatel's looking to catch up with chief competitors in channel count and distance, while at the same time providing a smooth migration path for its installed base. Ciena Corp. (Nasdaq: CIEN) and Lucent Technologies Inc. (NYSE: LU) can send more channels farther. But in real-world implementations, there are many other variables carriers need to consider, including the availability and cost of incremental improvements, the type of amplification, the use of digital wrappers and interfaces, and compatibility with other systems.

— Mary Jander, Senior Editor, Light Reading