Tag Archives: cloud computing

Netbooks, Moore’s Law And Which New App Will Come for Rescue?


The rise of Netbooks, of which 21 million units* will be sold in 2009, may put in jeopardy the progression of Moore’s Law, as netbooks cannibalize sales of laptops. As a result of Moore’s Law, the price of a mid-range PC has not changed much since long, but every year the power of the machine you buy doubles the one available a year before. This exponential growth in performance can be easily tracked in the amount of storage that you get each year for a $25 SD card (8GB in 2008 vs 4GB in 2007) or for a $100 hard drive (1TB in 2008 vs 500GB in 2007). For microprocessors the rule is shifting from doubling clock speed, to doubling the number of cores every 12-18 months. This geometrical progression results in a laptop in 2018 with 256 cores and 32 TB solid-state drive.

Nebooks are driving Moore’s Law in a different direction. A Netbook has the power of a PC of 3-4 years ago, just at a fraction of the cost. And the reason netbooks replace laptops is that they can run the same basic applications that most people use: web browsing, email, writing docs or Skype. Although conceived as an ideal second ultraportable computer for road warriors, netbooks could become the prime PC for basic use in the office and at home for many in these times of recession.

Software, and in particular Windows, have driven the evolution of the PC hardware with “hungrier” software versions every few years. But Vista failed to bring any breakthrough features compared to Windows XP, and netbooks are benefiting from that.

What do software developers have in their roadmaps that will need 256 cores in ten years? Video editing, media encoding or photo processing software need powerful CPUs, but unless the future bring us an OS with disruptive user interfaces with 3D virtual reality, HD, plus voice and gesture detection, it is difficult to imagine why we will need 256 cores in a laptop or desktop PC.

What is sure is that Multi-Core CPUs fit nicely with server virtualization, and that will make data centers which host the cloud more cost-efficient. The demand for multi-core processors for servers will sure drive the CPU market, but will it be able to do it at the same pace as desktops were doing before? If we are to believe in the success of the cloud, not only the value is shifting from the OS to the Cloud, also the demand for stronger processors is.

We could well end up in 2018 with 256-core servers in the data centers and ultra small low-cost dual-core devices in our pockets connected to the cloud.

* The Economist: Less is Moore

Intel could report soon the first quarterly loss in 21 years. Even if they say Atom margins are similar to those of the high-end processors, there is always the doubt if cannibalization has anything to do with the loss. On the other side AMD says they will not address the market for netbook processors, instead putting all eggs on the high-end CPUs that will power cloud servers. Will we see the 256-core CPUs somewhere else than in servers?

Modular Data Centers: Containers for Clouds

Mike Manos, Microsoft Data Center Chief, has unveiled in his blog the design of future “Generation 4” Microsoft data centers. With the mission to provide massive easy-to-scale computing infrastructure to power the Cloud, Microsoft envisions pre-assembled containers equipped with a few thousands of servers each and its associated cabling and cooling system.

Microsoft is already using the flexibility of “containerized servers” in their Generation 3 Data Center under construction in Chicago. Generation 4 pretends to take the concept of building block even further with a central spine infrastructure for mechanical, electrical and security components, to which the pre-assembled containers are connected in a plug-and-play mode. The containers are designed for high efficiency, minimizing both footprint and the use of water or air for cooling.

The target is to achieve an average power usage effectiveness (PUE)* of 1.125 by 2012. Taking into account that current average data centers have a PUE of almost 2, that would be a huge achievement in green IT, even though Google alreadly claims to be at a 1.21 PUE in some of their facilities.

Microsoft modular container-based design is ideal for a scalable cloud infrastructure that can adapt computing capacity to demand. On top of that, not only operation costs are reduced due to the optimized footprint and cooling, but also CAPEX is improved with server containers assembled in a single manufacturing plant for far less cost than deploying the servers on-site.

Watch Microsoft video illustrating the Generation 4 Data Centers. An interesting approach to counter attack Google’s thought leadership earned with their patent filing for floating, wave-powered data centers.

Microsoft is taking Cloud computing seriously, as Allan Leinwand of GigaOm writes today:

“… in conjunction with announcements of the Azure Services Platform and Office Live, there is no doubt that the giant in Redmond is aggressively focused on delivering enterprise cloud computing. ”

*Definition of PUE extracted from Google corporate site:

PUE is defined as the ratio of the total power consumed by a data center to the power consumed by the IT equipment that populate the facility:

Power Usage Effectiveness

For example, a PUE of 2.0 indicates that for every watt of IT power, an additional watt is consumed to cool and distribute power to the IT equipment.

Top 5 Disruptions in the Past Twelve Months

These are the top hottest five topics that we have covered over the past twelve months with the biggest potential for a disruption in both the Communications Industry and in our habits.

1) Mobile Internet, the Internet way.
The iPhone and Google’s Android have revolutionized Mobile Internet in the handset, making web browsing such a cool and better experience that operators have embraced unlimited data plans for the first time, in bundle with these “next-gen” handsets (well, in fact operators did it first with Blackberry for unlimited enterprise email). Downloadable Mobile Apps were already available for Symbian, J2ME, Windows Mobile and Blackberry, but never have they been hotter than they are now on App Store and App Market. The Winners: Apple and Google (and even Blackberry to a certain extent). The Losers: Nokia, Sony Ericsson and Windows Mobile. The short-term winners, long-term who-knows?: The Service Providers that are cashing-in the flat fee Unlimited Data plans, but sacrificing their chances to ever control again the Mobile Internet as they did before.

2) Cloud Computing.
Amazon Web Services, Google App Engine, Google Apps, IBM “Cloud consulting”, or Google releasing Chrome to run complex apps on the browser at full speed, have made headlines in 2008. Startups such as Animoto run on Amazon’s cloud to enjoy flexible on-demand computer scalability (no wonder Amazon is also an investor in Animoto). The hype is such that these days everything is getting cloudy (or re-defined as cloudy) : GPS navigators face competition from the clouds with Google Maps (maps run on the cloud, with real-time info update and interactive recommendations/ads). Even speech recognition now runs in the cloud with Google Mobile App. The Winners: Google getting us to spend more time online, pioneer Amazon that gets economies of scale, and the IT vendors, such as HP or IBM, that will equip clouds. The Losers: Microsoft, for the moment. SaaS running on a browser make the OS less relevant. The catastrophe of Vista has only help the Cloud cause with enterprises eager to consider alternatives. Microsoft is building huge data-centers and will be a big cloud player too. For the moment, the cloud will put pressure on them.

3) Internet TV.
Hulu has brought national broadcast TV shows to the Internet. YouTube has confirmed the revolution of Internet Video, now well into the mainstream. YouTube not only is reaching agreements with content owners to be an ad-based platform for content distribution, but it has been a huge platform for politicians for the US 2008 presidential campaign, and now for President-elect’s change.gov. Not to mention the new habits of kids to search YouTube for information instead of searching Google. The Winners: All of us that now have virtually unlimited video content, including old-time TV jewels, just one click away. The Losers (yet to be confirmed): Broadcasters that will cannibalize higher CPMs revenues of traditional TV, with much lower CPMs prices in the online world. The effect is still small but innovative models will be needed before the cannibalization goes too far.

4) Wireless Broadband. With HSDPA or EVDO widely deployed, the 3G networks are now beginning to deliver on past promises. WiMAX has also seen in the past twelve months many commercial launches, and will play a key role in emerging markets where fixed broadband is not a financially viable option. The Winners: Mobile Operators that will get an additional source of revenues selling Mobile broadband, and opening the door to Machine2Machine applications taking advantage of ubiquitous wireless connectivity. The potential losers (not yet): Fixed Operators will need to keep fixed broadband ahead in terms of performance, to avoid substitution by wireless. Mobile Operators will lose too, as their hopes to make Mobile Internet different from fixed vanish.

5) Netbooks.
Asus EEE PC has been followed by most laptop vendors (supported by Intel low-power low-cost processors such as ATOM)  to bring a new class of 1 kg laptops with less than 9″ screens, ideal to enjoy full Internet on the go. Combined with Wireless Broadband, these low-cost laptops are not only a great companion for road-warriors but also are likely to power Internet access in many emerging countries as an affordable PC solution. The Winners: Millions of new Internet users worldwide, as well as Internet addicts. Flash memory manufacturers. The Losers: High-end lightweight laptop makers, that will face competition from low-end machines with enough power for everyday use. DVD/CD writers and disc makers, with discs replaced by memory cards and the disc writer by a card reader.

Converging into Clouds

Online players, such as Google and Amazon cope all bets to reign over the Clouds, benefiting from their global reach and huge experience on web scalability. Others foresee IBM -talking about utility computing for decades- or HP, after acquiring EDS, as strong contenders to put IT systems on the clouds.

But what about Telcos? Wouldn’t they want to play too? At then end, operation of network services is what they have been doing for ages. Telcos own the connectivity, which is a key element in cloud computing. They can promise their business customers a guaranteed bandwidth and latency, that other on-line Internet players, simply can not commit. And they are close to their corporate and SME customers, to whom they already sell PABXs, internet access, leased lines and mobile phones. So telcos know their market in their local countries better than Google. Do not underestimate the power of the Operators. They have the strength to being able to gather an ecosystem of partners they can bring to their enterprise customers.

Internet giants, IT players and telcos, they all want a place in the cloudy sky.

SingTel to package SaaS offerings
IBM launches four new cloud computing centers

Google Chrome: Get Equipped for Clouds

Google has surprised the world with its new browser. The product introduction through a comic was a brilliant exercise of showing the superior features of their browser, most of them quite technical, in an simple, plain, easy to understand manner.

Google wants the browser to make the OS irrelevant. The intent was already clear when they launched Gears to power up Google Docs and Google Readers by giving them access to local storage so that applications could run on the browser even when offline.  Chrome is the next step to make sure that the browser is able to run complex and rich applications, as fast as they would run on the OS. 

That is why Chrome includes a fast Javascript Virtual Machine, called V8, designed to run complex Javascript code at the speed of native OS applications. It uses WebKit for rendering, the same engine as Apple’s Safari, and it has been heavily tested with millions of web pages cached in Google’s search platform, to make sure that developers will not need to adapt web applications to yet another browser. As Google will use its browser to drive web innovations, it is open source so that new inventions will be easily available to other browsers.

Quick to render pages and run Javascript, simple clean user interface, more secure against crashes (than IE7, which is easy), includes anti malware and anti-phishing protection, and new nice features, as the Incognito navigation. If you thought that browsers was just a commodity with few room for differentiation, meet Chrome. The perfect fit to run applications on the Cloud safely and fast.

See clip above on Google presentation of Chrome yesterday.

In the Clouds

Cloud Computing is a fancy term these days that applies to many different things. Concepts such as Software as a Service (SaaS), On-demand Computing, Utility Computing, Managed Services or even older ones as Hosting or Outsourcing, seem to all fall under the Cloud buzzword.

The basic concept is that the Cloud provides IT infrastructure (processing, storage, database or applications) on-demand, on a pay-per-use basis. The most renown examples of Cloud Computing are Google App Engine and Amazon Web Services. Microsoft recently launched Live Mesh, enabling users to upload files, folders, and sync PCs and devices info through the Live Mesh cloud. Web applications such as Salesforce.com or even Facebook, are often referred as clouds, providing SaaS, and Facebook giving APIs to develop applications. Even the recent HP bid to acquire EDS, has been tagged as a move to reinforce HP position in the Cloud market.

A vivid example of what Cloud computing can represent for an online startup can be found in NYT’s article Cloud Computing: So You Don’t Have to Stand Still.

[…] Animoto, an 18-month-old start-up in New York that lets customers upload images and music and automatically creates customized Web-based video presentations from them; […] earlier this spring about 5,000 people a day were trying it.

Then, in mid-April, Facebook users went into a small frenzy over the application, and Animoto had nearly 750,000 people sign up in three days. At the peak, almost 25,000 people tried Animoto in a single hour.

To satisfy that leap in demand with servers, the company would have needed to multiply its server capacity nearly 100-fold, says Stevie Clifton, 30, a co-founder […]. But (they) had neither the money to build significant server capacity nor the skills — and interest — to manage it.

Instead, they had already worked with RightScale, a cloud services firm in Santa Barbara, Calif., to design their application for Amazon’s cloud. That paid off during the three-day surge in growth, when Animoto did not buy or configure a single new server. It added capacity on Amazon, at the cost of about 10 cents a server per hour, as well as some marginal expenses for bandwidth, storage and some related services.

The example illustrate the opportunities that Cloud Computing represents for entrepreneurs for prototyping applications with little investment. Startups can launch without the risk of being killed by success, as they are able to easily adapt IT infrastructure capacity to their growth. See “Cloud” price lists from Amazon and Google below:

Amazon Web Services pricing in USA:

    $0.15 per GB-Month of storage used
    $0.100 per GB – all data transfer in
    Data Transfer
    $0.170 per GB – first 10 TB / month data transfer out
    $0.130 per GB – next 40 TB / month data transfer out
    $0.110 per GB – next 100 TB / month data transfer out
    $0.100 per GB – data transfer out / month over 150 TB
    $0.01 per 1,000 PUT, POST, or LIST requests
    $0.01 per 10,000 GET and all other requests*
    * No charge for delete requests

Google App Engine provides 500MB storage and up to 5 million pages per month on the free package. ReadWriteWeb discloses that App Engine price will be:

    $0.10 – $0.12 per CPU core-hour
    $0.15 – $0.18 per GB-month of storage
    $0.11 – $0.13 per GB outgoing bandwidth
    $0.09 – $0.11 per GB incoming bandwidth

The old promise of Outsourcing to convert CAPEX into OPEX is now at the service of the entrepreneur. Clouds can be a true engine for innovation.

Picture: Ricardo Carreon

Cloud Computing: Welcome to Google’s Cloud

Google has just launched App Engine, the service to give developers access to Google Cloud, or in other words, to the same Google infrastructure that runs G-Mail, Google Apps or Google Earth.

App Engine provides a number of tools to develop Web applications that will run on Google data centers. The basic service will be free, and will provide for capacity to serve 5 millions pages per month, including 500MB storage and 10GB data transfer per day.

As you can see in the video above, Apps Engine is easy to use and supports applications in Python. Other languages will be supported later.

Amazon was already offering access to its Cloud with Amazon Web Services, claiming 330.000 developers, and Google will be a strong competitor. Start-up companies like Bungee Labs were already offering Platform-as-a-Service, and while these companies will be hit, Google service will create an awareness on Cloud computing, that will make them target for acquisition from other big Internet players not yet in the Cloud trend.

As GigaOm suggests, this service could be a base for other development platforms like OpenSocial for social networks and Android for mobile. Could Google fulfill the promise of Java, “build once, run on any platform”?

To learn more about Cloud Computing and its buzz, here is a good article from NYT: What Cloud Computing Really Means.