Ethernet QFSP

Avago has developed a new four-channel pluggable similar optic QSFP transceiver module for 40 Gigabit Ethernet applications.

The section is completely biddable to the IEEE 802.3ba 40GBASE SR4 design and provides a solution for multilane data communication and interconnects applications that incorporate four independent 10 Gbps data lanes in each direction to provide 40 Gbps cumulative bandwidth. In addition to 40 Gbps Ethernet interconnects, this transceiver can be used in datacom/telecom switch and router connections, as well as data aggregation and backplane applications.

With data rates of 10 Gbps for up to 100 metres using OM3 fibre, the transceiver is designed to operate over multimode fibre systems using a nominal wavelength of 850 nm. The electrical interface uses a 38-contact edge type connector while the optical interface uses either an 8 or 12 fibre MTP (MPO) connector. The element also provides greater electro-optical concert to enable low jitter required for high-speed computing, server clustering, Infiniband and Ethernet switching and core routers. Additionally, this transceiver is hot pluggable for ease of installation and servicing, and backward compatible with 5 and 2, 5 Gbps per channel applications.

Infiniband to Outpace Ethernet's Expected Force

Every good plan in networking ultimately seems to be borged into the Ethernet protocol. Even so, there's still a place in the market for its key rival in the data center, InfiniBand, which has constantly offered more bandwidth, lower latency, and often lower power use and cost-per-port than Ethernet.

But can InfiniBand keep outrunning the tank that is Ethernet? The members of the InfiniBand Trade Association, the association that manages the InfiniBand specification, think so.

InfiniBand, which is the result of the fusion in 1999 of the Future I/O spec espoused by Compaq, IBM, and Hewlett-Packard and the Next Generation I/O contending spec from Intel, Microsoft, and Sun Microsystems, represents one of those rare moments when key players came together to create a new technology then kept moving it forward. Sure, InfiniBand was relegated to a role in high-performance computing clusters, lashing nodes together, rather than flattering a universal fabric for server, storage, and peripheral connectivity. Roadmaps don't always pan out.

But since the first 10 Gb/sec InfiniBand products hit the market in 2001, it's InfiniBand, more than Ethernet, which has kept pace with the discharge core counts in servers and gigantic storage arrays to feed them, which stipulate massive amounts of I/O bandwidth in the switches that link them. Which is why InfiniBand has persisted despite the offensive of Ethernet, which jumped to Gigabit and then 10 Gigabit speeds while InfiniBand evolved to 40 Gb/sec?