I am talking about Lightfleet Inc. a rather new company, which plans to sell servers far more efficient than existing high-end systems. Current architectures imply a serious slowdown when complex calculations are involved, as each microprocessor waits for data being processed by other chips. The actual congestion may appear when many CPUs try to simultaneously send messages to each other. In order to solve this problem, Lightfleet prepares server configurations which will use 32 dual-processor chips from Intel Corp and innovative laser technologies.
Lightfleet engineers decided to exploit lasers in ways that are not usually used for communications. In today's configurations, optical-networking devices send tightly focused pulses of light down strands of fiber-optic cabling. Lightfleet, on the other hand, does the opposite thing: it uses lenses to spread out laser beams and bounce the light off a mirror to send data around a system. This way, each microprocessor is installed with a laser transmitter and a set of devices that receive beams of light carrying messages from other chips. The light is reflected off the mirror and passes through focusing lenses to the receivers, reaching each CPU chip in considerably shorter times.
Messages from each processor, or any combination of them, are simultaneously sent to all the other microprocessors. Each receiver only picks out the messages intended for it, because of special addressing information sent with the light beams. Because the system sends light through air and not through optical cabling, Lightfleet avoids the need for wiring and associated switching circuitry and software. Thus, CPU messages do not collide and the overall information traffic does not get jammed.
Scientists at Lightfleet admit that the adoption of their new laser technology could be hindered by issues regarding the software rewriting.