RESEARCH TRIANGLE PARK — Just as Lockheed has its “Skunkworks” where famous aircraft such as the B-2 Stealth bomber are developed in secrecy, so too does MCNC have its own group for rather quiet projects.
“If I tell you everything we’re working on, I’d have to shoot you,” says Dave Rizzo, chief executive officer at MCNC. Only nanoseconds later does he break into a big smile.
Well, not quite that secret, points out Dan Stevenson, vice president of Advanced Networking Research (ANR).
Sensitive, yes. But disclosure leading to firing squad?
“There’s no spooky stuff involved,” Stevenson says after chuckling when told the Rizzo story.
Stevenson and company do have a nickname for their latest assignment: “Project Jumpstart”. Sounds better than skunk, right? And the idea fits.
In a sense, the MCNC group is trying to jumpstart existing communication networks to a new level of performance and speed.
Stevenson peels off a layer or two of the obscurity that does surround some of the cutting-edge communication work he is involved with as he discusses a new $1.35 million grant MCNC just received to explore an emerging technology called “optical burst switching”. (Outside the bounds of this conversation was MCNC’s other think-tank group that spends a great deal of time with the feds working on communications encryption technology. Maybe we can talk with them at a future date — provided we can decipher what can and can’t be discussed.)
The money trail
The money is just the latest in a bundle of cash sent MCNC’s way by of the Advanced Research and Development Activity, or ARDA for short. This group is an offshoot of the nation’s Intelligence Community — the friendly folks from the CIA, NSA and the NRO (National Reconnaissance Office, the people who run spy satellites.)
But to better understand what’s going on with optical burst technology and MCNC, it’s a good idea to better understand who is paying the bills.
Here’s how ARDA describes itself on its web site:
“The Advanced Research and Development Activity (ARDA) is an Intelligence Community (IC) center for conducting advanced research and development related to information technology (IT) (information stored, transmitted, or manipulated by electronic means). ARDA sponsors high risk, high payoff research designed to produce new technology to address some of the most important and challenging IT problems faced by the intelligence community. The research is currently organized into four technology thrusts, Information Exploitation, Quantum Information Science, Global Infosystems Access and Novel Intelligence from Massive Data.”
If that blurb doesn’t tell you how important the MCNC project is, then tune in a spook movie such as “Patriot Games” — the first to introduce a worldwide audience to the potential of real-time image from space as Special Forces take out a terrorist training came in the middle of a desert, at night no less — or Will Smith and Gene Hackman in “Enemy of the State”.
And if you want a true life example, just go back a few days in the war on terror when CIA officers, perhaps as far away as the United States, directed a drone aircraft known as a Predator within range of al Qaeda terrorists in far-off Yemen and wiped them out with a Hellfire antitank missile.
Such operations require virtually instantaneous communications. One nanosecond of a delay in image or relaying of command and — poof – Osama bin Laden lives to fight another day. And there have been stories about ‘just missed’ chances.
“We could do those kind of things, yes,” Stevenson says when asked if optical bursting could help improve communications for that kind of operation. “Of course,” he hastens to add, “we’re just speculating.”
What is optical bursting?
Optical refers to light, as in speed of. Bursting refers to how the information is handled at best speed. And what Stevenson and his team is working on is a technique of sending data or other information over networks without having to go through network buffers or converting data from electronic to light (optical) back to electronic.
“Packet switching (as in Internet protocol) requires relatively complex protocols relative to circuit switching and requires a lot of memory in the switching nodes,” Stevenson says. At the routers, information is “buffered” and prioritized before being sent on the network.
One of the problems with Voice Over Internet Protocol as a replacement for standard voice communications has been latency — or delay. Those annoying pauses. Latency also creates havoc for such things as on-line gaming where nearly instant communication is crucial. Packets also sometimes get “lost” and the network must retrieve, reassemble and resend all the packets.
Imagine how much higher the stakes are in the intelligence agencies.
We’re not talking enough time to flip the remote control here, but telecoms, federal agencies, companies such as Cisco and Nortel, have spent untold billions of dollars trying to squeeze latency out and reduce packet loss.
Not all networks are IP based or run fully over optics, and most still have a lot of transmitter-to-receiver direct links.
“In a circuit network (such as standard phone or analog networks), you have to set up in such a way to create a continuous connection of information that goes from transmitter to transmitter and very little buffering of that information takes place,” Stevenson says. “There’s no ‘leakage’.”
The standard networks have their own problems, however. Even if based on optics, the electronic information has to be converted to light then transmitted over fiber and converted back to electronic.
Stevenson says he believes MCNC has developed a solution with the new protocols it is developing.
“Optical burst switching is a technique that sits in the middle ground between the two,” he explains. “We send information in units we call bursts. They can be very short like packets or very long like a circuit. We don’t do any buffering.
“The reason this is important is that it comes down to performance and economics. Whenever you can keep the functionality in an optical format as opposed to converting back to electronics — there’s a lot of money to be saved.”
Steven also points out that since packet switching in optics “requires complex logic (programming software for routers and switches) and buffering, this is beyond the reach of optical technology as a matter of practicality.
“Optical mirrors are very crude, and the cost of memory (in the equipment) can be prohibitive.
“Optical bursting is a way to bridge the way logarithms are done today and the Holy Grail, which is all packets in optical networks.”
So what does that mean? More tomorrow.
Rick Smith is managing editor of Local Tech Wire.