An Ignoble Experiment

2. Intel preview: Will the next Xeon include SDN, NFV 'inside?'

It's a net neutrality issue of sorts, except that instead of being about the rights of bits, it's about the rights of electrons. The promise of the smart grid is typically described with bold, sometimes grandiose, Walt Disney-like language, and usually includes a sentence or two about how the smart meter in your home or business can work in tandem with your power provider to help better manage the electric grid as a whole, maintain appropriate power levels and conserve energy usage.

That's the same promise that Internet service providers would apply to a smarter net, if the notion of two-way bandwidth control weren't already colored by the "Open Internet" debate. Network management has to happen somehow; and for the electricity delivery network, the changing demands of households and offices necessitate a two-way relationship between the provider and consumer of service.

But while the installation of smart meters in some metropolitan areas has eclipsed 80 percent, by some estimates, there is no single communications standard between power companies and smart meters nationwide. In fact, since communications schemes may have been determined on the fly, there could be hundreds. Meanwhile, smart meters themselves lack the smarts to connect to smart devices, such as hybrid and electric vehicles (EVs), giving them the data they need to throttle back their power draw when the grid approaches overload.

Last week, Ford Motor Company announced an initiative to build a central server with which the nation's power providers could communicate not just with Ford electric cars, but 12 major worldwide brands, including GM. In an interview with FierceEnterpriseCommunications, one of Ford's principal consultants on this project, Dr. Nancy Ryan, who directs policy and strategy for the firm Energy & Environmental Economics (E3), explains the logic behind Ford's move to connect the power company with electric cars without necessarily involving smart meters.

As a former member of the California Public Utilities Commission, Dr. Ryan was a champion of smart meter technology, defending it against complaints that it enabled utilities companies to charge customers more during times of peak demand. Smart meters, she says, already pass utility companies' cost/benefit tests simply for requiring meter readers to come on-premises. It's still theoretically possible, she adds, for smart meters to communicate price signals to charging devices, informing them of the most cost-effective times of the day or the month to draw power.

"Given the amount of smarts that's in cars today, I think it's logical to have some of that control happening inside of the car," remarks Ryan. "I don't think it's any kind of existential threat to smart meters. Honestly, sensing, metering, and control technology keeps on getting cheaper. It's really more the institutional arrangements and partnerships in the market that make it possible to capture the potential of all of those devices."

Partnerships like the one Ford is forging with utility companies will expedite the time when those companies can have much more granular control over the impact that EV charging will have on the power grid. Ford emphasizes that enabling this level of control will only take place on an opt-in basis by the customer. But Ford's value proposition to the customer is based around the notion of helping to prevent overloads. Is there truly a danger from electric vehicles overloading the grid?

"We're a ways away from that really being a problem," admits Dr. Ryan. "Which means we have headroom to solve the problem.

"If you look at things from the wholesale level, EVs have to get up above a couple of million in California, for example, to begin to have the kind of load growth that can make an impact on the bulk power system." Ryan actually believes that declining power load as a result of more efficiently distributed generation will balance out the growth in power loads from increasing numbers of EVs over the next 15 to 20 years.

So if overload won't be a problem--what's Ford's objective? Could it be, as I suggested, a way to enable power companies to offer--perhaps on a prepaid basis--premium tier pricing for EV owners, analogous to the "Netflix premium" pricing models that ISPs are considering?

While Dr. Ryan wouldn't rule out the possibility, she believes that would be a bad idea. There are some pre-paid electricity subscription programs available, including in Texas, she said. "I think it would be progressive politics that informs a lot of customer-facing policy for the residential sector, that would preclude going to pre-paid. There's a kind of visceral... Any kind of customer-facing business model that includes having the possibility of having their service disconnected, would be an incredibly tough sell in California."

For more:
- read the L.A. Times' 2010 account of Dr. Ryan's defense of smart meters
- see Betanews' 2010 coverage of the Comcast/BitTorrent throttling issue

Related Articles:

Ford's next car-charging network could bypass smart meters
Big data's impact on utilities: from smart grid to soft grid [FierceBigData]
Big data analytics sparks reinvention of smart grid [FierceBigData]

Read more about: electric vehicles, Dr. Nancy Ryan
back to top

By Scott M. Fulton, III

Comment |

3. Ponemon: Infosec pros focusing more on protecting info than technology

This week at its Jones Farm, Oregon campus, Intel Corp. will be sponsoring a press-only workshop to discuss the work it's doing with its current "Haswell" generation of 22 nm server processors, especially with respect to a new series expected soon that will include the "v3" moniker. Take it from somebody who used to edit the news for Tom's Hardware: FierceEnterpriseCommunications is not Tom's Hardware. We're not usually about the CPU that runs the server box, but rather the things that plug into it that enable you to use it as a communications tool.

But there is a fundamental shift in data center architecture under way, and it has to do with networking and, well, enterprise communications. When virtualization started taking off as a genuine technology rather than just a cool trick, Intel was one of the CPU manufacturers that helped standardize and popularize the technology by embedding virtualization support functionality in its hardware. Software-defined networking is the next major technology to make positive, fundamental changes in data center architecture--the idea that the design of a network can be virtualized, and therefore pliable and elastic rather than fixed. Network functions virtualization derives from SDN: the notion that the components and devices that comprise the typical network, can all be rendered as virtual machines and run more efficiently by CPUs.

Intel is expected to be embedding SDN, and quite possibly NFV, support functionality into its CPUs in the coming months. It's part of the company's new strategy to embed function-specific features into its chip designs, often at the request of partners, and sometimes (so Intel tells me) on the exclusive request of certain partners, who may not even make these features publicly known. We're already seeing this with a special model of Intel's Xeon E7 v2 processor, built at the request of Oracle. Intel will continue to build a general-purpose core architecture, of course, but for certain vertical market segments and for specific customers, it will supplement that architecture with special features.

So I'll be asking Intel some very specific, special-feature questions this week:

Could a "bare-metal" server based on a near-future Xeon v3 processor enable, say a server based on Brocade's Vyatta platform to, say, accelerate its routing functions using embedded microcode, or an OpenFlow-based switch to improve its handling of memory, similar to the way Intel accelerates virtualization with its VT technology now?
Will a Xeon v3-based server provide enhanced support for virtual machine security that might give one vendor--say, VMware or Citrix--at least a temporary edge over the others, if it adopts these features first?
Will cloud data centers begin utilizing a type of Intel processor that is somewhat different--and, in later years, very different--from the type used in enterprises for on-premise operations, such as data warehouses and business process management?
Could communications service providers such as Alcatel-Lucent begin offering hybrid IP telephony services within the next few years, using less-expensive on-premise servers with "Intel inside" and not as much in the cloud?
How soon will Intel be able to offer communications service providers systems that match the flexibility, and perhaps even the low power consumption, of the first ARM64 servers-on-a-chip, introduced just last May?

Stay in touch with FierceEnterpriseCommunications this week as I begin our in-depth look at Intel's communications hardware strategy shift, and how it impacts you and me in the communications field.

For more:
- see APM and Canonical's joint press release from last May on ARM64 servers-on-a-chip

Read more about: NFV, Xeon E5 v3
back to top

By Scott M. Fulton, III

Comment |

4. Dialogic: What contact centers cannot do with WebRTC by itself

Quite often, when you see a story here that's tagged with the word "Ponemon," it's not good news. Once Google's indexing bot realizes I'm not misspelling "Pokémon," it might even be worse news. Survey data--especially the kind that's commissioned by a software provider--usually highlights the kind of bad news that makes readers want to buy the software provider's products.

But what I've liked about the Ponemon Institute from the very beginning, besides its thorough attention to detail, is how honest the company can be. A Ponemon report released late last month, commissioned this time by Adobe, is a fine example. This particular report questioned some 2,300 infosec professionals in 18 different sectors worldwide.

And yea, the key finding was supposed to be that some 58 percent of those polled either agree or strongly agree with the notion that the BYOD trend has made it difficult for them to minimize the security risk to their companies' vital information. Gee, not really a surprise there.

My key takeaway, though, is this: Some 67 percent of infosec pros polled worldwide either agree or strongly agree with the statement that their company's current security policy focuses on protecting their vital information rather than protecting the "IT stack"--the technology with which that information is presented to users, and whose relative stability would usually make their lives less stressful.

This, to me, is a positive sign. With about the same number of people responding that their businesses are focusing on people rather than technology as central to the risk management process, this tells me that infosec is focusing on the social and interpersonal dimension of protecting information.

Infosec professionals are taught, from the day they set foot in the training room, that information is a series of electrons hovering nervously in the fortress of some memory system somewhere, just waiting to be knocked loose by some evil perpetrator. But new principles of governance in information security are making businesses focus on information as a resource exchanged between people by way of technology. In this new medium, people are no longer the weakest links in the chain of handoffs between storage media. Technology provides the convenient, yet volatile, mechanism for efficiently transferring information between people, when those people are the strongest links.

Like I've been saying, all computing is a communications process.

So the IT department doesn't have much say any more about the devices people bring with them into work. Concentrating upon devices as the cornerstones of information security wasn't an effective strategy anyway. As the Ponemon report puts it, "Restrictive, preventative security controls are not realistic. Prevention-centric security strategies are seen as less effective. As revealed in this research there is a shift towards people-centric security strategies that emphasize individual accountability."

For more:
- see the Adobe / Ponemon report, "Global Insights on Document Security" (reg. req.)

By Scott M. Fulton, III

Comment |