Day: May 27, 2005

Utility Computing

InfoWorld has a special report: “HP, IBM, Sun, and others promise a glowing future in which enterprises tap into computing resources where and when they’re needed.”

There are three basic definitions.

Utility as an on-demand computing resource: Also called adaptive computing, depending on which analyst or vendor you talk to, on-demand computing allows companies to outsource significant portions of their datacenters, and even ratchet resource requirements up and down quickly and easily depending on need. For those of us with gray whiskers in our beards, its easiest to think of it as very smart, flexible hosting.

Utility as the organic datacenter: This is the pinnacle of utility computing and refers to a new architecture that employs a variety of technologies to enable datacenters to respond immediately to business needs, market changes, or customer requirements.

Utility as grid computing, virtualization, or smart clusters: This is just one example of a specific technology designed to enable the above definitions. Other technologies that will play here include utility storage, private high-speed WAN connections, local CPU interconnect technologies (such as InfiniBand), blade servers, and more.

These three descriptions are different enough to seem unrelated, but in fact theyre dependent on each other for survival. Should utility computing ever live up to its name — a resource you plug in to, as you would the electric power grid — then that resource needs to be distributed, self-managing, and virtualized. Whether that grand vision will ever be realized is an open question, but at leaast some of the enabling technologies are already here or on the horizon.

China’s Technology Strategy

San Jose Mercury News compares China and the US:

They graduate four times as many engineers as we do.

They lavish generous tax breaks on tech firms.

They support local manufacturers.

They don’t respect intellectual property.

They, of course, refers to China. And the gripes from Silicon Valley business leaders capture in stark and accurate terms the key underpinnings of the growing tech rivalry between the United States and China.

None of these things happened by accident. They happened because China has something that the United States lacks and badly needs: a national technology policy.

he country long ago made a strategic decision that technology was paramount to its development and put in place a systematic policy to create a world-class technology sector. It sometimes runs roughshod over trade agreements or international law, which is wrong. But on the whole, the policy is simply smart.

And it’s just plain dumb for the United States to think it can compete in the tech race against China and other nations without a technology policy of its own.

In China, the importance of tech is articulated at the highest levels of government. “Science and technology are the decisive factors in the competition of comprehensive national strength,” Premier Wen Jiabao said just last month.

When will we wake up in India?

Google Gazing

[via Dr Abhishek Puri] SquirrelNet discusses speculation about GBrowser, GooglePC, Google NetPC and Google ISP. Here is an excerpt discussing Google NetPC:

Some say Google could instead offer a NetPC (a cross between a dummy terminal and a bare bones PC where all data would be stored and served centrally at Google). Per this example of a NetPC setup, you can see that a NetPC looks like a regular PC except for there is no bulky “big box” that makes lots of noise and produces a lot of heat. Just plug the “little box” into the internet and you’re ready to go. Almost all files would be stored at Google, although a small local hard drive or flash drive would still exist locally for caching reasons.

One big advantage of NetPC’s is that you’d never need to install software since Google would provide everything that most people need (email, browsing, word processing, spreadsheets and perhaps a few other things). What about games? If Google launches the Google NetPC, expect game companies like Electronic Arts to build for-fee games directly into the Google NetPC system (housed centrally at Google).

It’s worth pointing out that the Google NetPC would be different than any NetPC on the market, since it would have millions of users. As such, Google could expand the size of the NetPC marketplace, ultimately lowering prices to perhaps $50 to $150 (from current levels of $300 to $500), including keyboard, monitor and mouse. Of course, Google could offer the Google NetPC for free in exchange for a fixed term Google ISP contract.

If these prices sound optimistic, observe how Apple’s iPod has driven down prices of miniature high-capacity hard drives. Google could do for the PC and NetPC what Apple’s iPod did for personal music players.

The Google NetPC also increases the chances of a Google-owned Google ISP. Google could choose to increase the fiber optic “fat pipe” coming out of its Mountain View headquarters. By going this route, Google could buy or build their own ISP, making a Google ISP a reality. Thus, the Google NetPC running GBrowser, the Google Web Accelerator and various other Google tools could exclude the following from the revenue equation: Microsoft, computer manufacturers and various ISPs.

Cameraphones as Personal Storytelling Media

Howard Rheingold writes that “the cameraphone exists at this moment in that ephemeral, potent and confusing phase of its adoption cycle where people are still deciding what kind of social medium it is.”

If recent observations from Keio University researcher Daisuke Okabe can be used to forecast future trends, we will find that the social role of the cameraphone is distinctly different from both the camera and the phone. And although these devices transmit images through the Internet, they are also turning out, rather unexpectedly, to be face-to-face media. It looks like this newly ubiquitous device could be more about flows of moments than stocks of images, more about sharing presence than transporting messages, and ultimately, more about personal narrative than factual communication.

“Cameraphones enable an expanded field for chronicling and displaying self and viewpoint to others in a new kind of everyday visual storytelling,” wrote Okabe, in a paper delivered at a conference in Korea at the end of 2004. Okabe’s findings make a case that cameraphones represent a new opportunity to tell the story of our lives to ourselves as well as to others, and to share a sense of continuous, multisensory, social presence with people who are geographically distant. Tokyo youth have added a visual element to the flow of phone calls and text-messages among small groups of intimates that Okabe and colleagues have come to call “distributed co-presence.”

Qualcomm’s One Cellphone Chip

News.com writes:

Qualcomm unveiled several chips designed to underpin inexpensive cell phones with MP3 players and cameras, pitched at emerging handset markets in Latin America, India and China.

The three new processors–the 6010, 6020 and 6030–are the first in Qualcomm’s line of single-chip products. Each contains everything–the modem, radio transceiver, power management, multimedia engine and security features–it takes to make a basic cell phone. Single-chip cell phones, in theory, reduce handset makers’ costs, which translates to lower-priced phones for markets where cell phone penetration is very low.

The lineup is a sign of how single-chip design is becoming more attractive to handset makers. Texas Instruments, a major provider of mobile processors, has already announced it has developed a single chip for cell phones, while No. 1 handset maker Nokia has one in development.
Disharmony on your cell phone

Qualcomm’s new chips should enable manufacturers to create a less-expensive version of phones using its CDMA 1X standard. (Two years ago, the technology was the power behind state-of-the-art phones, but those handsets have since been supplanted by faster and more processing-heavy models.) The newly released 6020 chips, for instance, support MP3 players, and the 6030 can handle both a camera and a music player.

TECH TALK: The Coming Age of ASPs: Technology Building Blocks (Part 3)

4. Single Sign-On

Identity Management is the broader theme. Access to applications and processes can be controlled via an identity manager, and users can be given a single login and password which works across all the applications. Companies like Oblix offer services in this space.

5. Ajax

Ajax was coined by Jesse James Garrett of Adaptive Path. It uses a combination of asynchronous Javascript and XML to create rich client applications. What is interesting about the Ajax idea as demonstrated in Googles use for its Maps and Suggest feature is that it can help creative interactive client-side applications and potentially also reduce the bandwidth needed for server-side interactions. This could be very useful for SMEs in Emerging Markets (SMEEMs) because the broadband infrastructure is not yet fully in place.

6. Thin Clients and LAN-Grid

One of the key factors in ensuring the success of both the automation of SMEEMs and the ASP model is that there will be a computer on every desktop in the enterprises. We are still far away from this reality. In countries like India, perhaps there is just over 1 computer for every 10 employees who need to process information across the 40 million who work in 4 million SMEs. Putting the computing infrastructure in place is therefore a critical part of ensuring the success of the ASP model.

It should be possible to build a thin client for no more than $110 (Rs 5,000) about $70 for the computer and $40 for a refurbished monitor. (A new monitor would add about $40 to the cost.) The thin client itself would have the guts of a cellphone with the capability to run Linux, a browser, and support remote desktops so as to shift processing and storage to the server. It would also have the capability to connect key drives via the USB ports. By using the chips used in mobile phones, we can dramatically reduce the cost of building a thin client. [For comparison purposes, an x86-based thin client would probably cost about 70% more at $120.]

Besides making computing affordable, thin clients also increase manageability since these computers would not fall prey to viruses, require no upgrades for 5-7 years (or even longer), and have no local hard disks on which users could load their own programs. Thus, the thin clients can be remotely managed reducing the total cost of ownership.

Complementing the thin clients would be a LAN-Grid a server-centric computing platform which would deliver the necessary applications. It would centralise all processing and storage. The estimated cost for a LAN-Grid infrastructure would be about $50 per user, given a minimum of 20 users in an enterprise. The LAN-Grid can be managed and updated remotely.

Taken together, the thin clients and LAN-Grid can help create a 1:1 computing infrastructure (one employee, one desktop) for as little as $160 in capital expenditure. Amortised over a 4 year period, the cost per user would work out to less than $5 per month, even assuming some server maintenance and upgrade costs. By ensuring that every user in the enterprise has a desktop connected to the Internet, ASPs can get a jumpstart for their services.

Next Week: The Coming Age of ASPs (continued)

Continue reading TECH TALK: The Coming Age of ASPs: Technology Building Blocks (Part 3)