Cloud Computing & Africa

For the last week or so, I’ve been pondering a thought provoking article in the July 2008 issue of Wired - ‘The Petabyte Age,’ by Chris Anderson. I call it thought provoking because Anderson asserts that current statistical and theoretical methods will be trumped by the availability of HUGE amounts of data - PetaBytes (1000’s of terabytes), and the ability to process it. The basic idea is that with enough amounts of data, the numbers will eventually speak for themselves. I suppose the other reason this peaked my interest is that about a month ago I wrote about 'Crowdsourcing and the future of crisis reporting'. With enough data, you’d be able to ‘quite possibly’ predict future human and environmental disasters! While large amounts of data exist, the problem to date has been how to process it, and that’s how cloud computing comes into place.

Cloud computing is a relatively new word. Some people have used the term grid computing to refer to the same thing. It basically refers to the outsourcing and consolidation of computing services and resources. I’ll explain some more so that you can understand. The biggest nightmare to most companies IT departments has been dealing with scalability and the processing power or storage that is required to address it. Enter cloud computing companies - which would basically provide you with on-demand capacity as you grow. Some big companies you know have already been providing such services for the last few years; Amazon, Google, IBM and many more, each striving to be the next general-purpose computing platform. Basic services like Google Docs and Amazon S3 are modest examples of cloud computing.

Lately I’ve been following Google’s forays into Africa. Especially interesting has been the introduction of the Google Global Cache (You can read more about it on White African’s blog). Google in their infinite wisdom has come up with a way to deal with the voracity for data on the African continent. Because bandwidth demand surpasses supply, they would store the web on their servers and serve it up to users based on the shortest path to location, thereby improving performance. Some would call this cloud computing at its best. This would save on bandwidth costs for ISP’s, and improve performance for their customers. Wow!

If you don’t already know by now, that ‘Wow!’ in my last paragraph was meant to be sarcastic. While I applaud google’s efforts, google is inherently a corporation whose goal is ultimately profit. Google China is a prime example of how this all comes into play. Search results for ‘Tiananmen Square’, ‘Falun Gong’ or ‘Tibet’ yield far different results within China than outside it. The fallacies of cloud computing start to become clear. If a whole continent outsources its computing needs, isn’t this data then open to manipulation by the 3rd entity (Google) and other political and profit motivated interests?

Don’t get me wrong, I think Africa stands to benefit a great deal from ‘this’ faster access, but I think this means we need to step up our efforts to increase bandwidth instead of depending on a 3rd parties for access to information. We are at the dawn of the age of information, opening up new paradigms of thinking (some jokingly call it the age of mis-information). Just as happened during the industrial age, new products and integrated services are jostling for attention and marketplace attention. Ultimately there will be just a few winners.

del.icio.us:Cloud Computing & Africa digg:Cloud Computing & Africa reddit:Cloud Computing & Africa blogmarks:Cloud Computing & Africa Y!:Cloud Computing & Africa

9 Comments so far

  1. hash @ July 5th, 2008

    Nice picture. I can’t decide if the mtoto looks clueless or scared. :)

    The whole idea behind GGC is good, but it’s only a short-term solution. What we really need is direct undersea cables with greater bandwidth. The truth is that Google is doing it for their own interests, fortunately those align with most everyone else at the moment.

  2. david @ July 5th, 2008

    Hash, I think the kid is clueless… LOL.

    We do need the direct undersea cable, and its sad to see the GGC come up as a result of that delay.

  3. fimbo @ July 5th, 2008

    David,
    I think the real revolution in data processing in the African Enterprise will happen at the Gigabyte level, not the pretentious terabye or even petabye.
    This is especially so if you are serving your pages to dialup and mobile users whose needs end with texts, logos and thumbnails.
    Gigabye’s/terabyte only advantage is serving millions videos, audios and pics.
    It is already feasible to run a modest box with a 500 GB RAID server on a FOSS/LAMP platform from a room in Nairobi, whats left is for a gutsy Kenyan to do get in the “cloud” with it.
    We sometimes forget that Google and Yahoo started in dorm rooms.

  4. david @ July 5th, 2008

    @fimbo - when you put all of Africa’s text and image requirements for the internet, it can add up to terabytes and petabytes. Serving all that information back to the user has posed a problem because of the bandwidth bottleneck.

  5. Ssembonge @ July 5th, 2008

    Google is in the business of internet search. Whatever they do is meant to improve user experience and give them an edge over their competitors.

    I guess it makes a big difference if you get 18,300,000 search results in 0.54 seconds rather than 300,000 in 1.5 seconds.

    The good thing is that people have a choice when it comes to internet search and can use other search engines if they prefer.

    I’m with google on this one.

  6. Peter Njenga @ July 7th, 2008

    Google is doing this not because they love us, but because they see a money minting opportunity in serving us information.
    The undersea fibre optic cable is the only way to go for East Africa.
    I have just finished ranting about our own telecomms companies not giving a damn about the internet, yet we use their service to access thge internet. Just look at the web sites of Telkom, Celtel Kenya (if you’re lucky to get through) and Safaricom (nice Homepage, poor innards).
    See my post at http://www.peternjenga.blogspot.com.
    Nice blog David.

  7. chris @ July 7th, 2008

    the China comparison is a poor one because while Google’s preference would be to provide the exact same service in every country, it is forced to comply with local laws and regulations in order to be allowed to operate a local entity in a particular country. China is not an ally to those who support the openness of the internet, so as long as Kenya and other African governments promote openness, they will have no problem with Google.

  8. david @ July 7th, 2008

    @chris - the China comparison is not a poor one. If anything I think it is quite appropriate. You yourself state that ‘Google’s preference would be to provide the exact same service in every country..’. It is their preference, but they can’t because they’re forced to comply with regulations in certain locales. Yes we know China does not support openness - how do we know what level of openness will be guaranteed in Africa?

    The Google Global Cache is a noble effort, and I’m all for it personally - the key is just to realize how powerful a tool it would be in the wrong hands. It will definitely make things easier for dictatorial regimes in Equatorial Guinea, Libya, Eritrea and Sudan that already spend huge amounts of money on internet censorship and control. Will Google comply with their laws and regulations?

  9. Dean J. Garrett @ July 13th, 2008

    The best recent trends in the cloud computing space, specifically SaaS is all the smaller vendors like http://www.HostedDatabase.com and others, who offer simple-to-use and low-cost apps that appeal to non-U.S. (especially African) customers. The consumers of these apps benefit from the simplicity and low-cost.

Leave a reply

Subscribe without commenting