My colleagues here at CloudAve have talked about the recent Gmail outage at length – I’ll not speak to that topic it seems to have been well covered.
Krish wrote an excellent post (his posts are always excellent by the way) claiming that Cloud Computing users should always consider two basic requirements to ensure the security and availability of their data;
- Redundancy
- Diversification
The redundancy part I agree with entirely and I commit to this strategy in my own world. As an example – my desktop files (yes, unlike others I love with out 24/7 connectivity so I still have files on my machine) are backed up and synced to not one but two providers (syncplicity and sugarsync – both great products) and appear on three separate hard drives (a RAID pair on my desktop and also on my laptop HD), while all out family images are replicated on two harddrives and two separate web services (flickr pro and picasa).
As for the diversification part I’m going to risk a flaming and disagree with Krish. He said that;
When small businesses move their data and applications to the web, it is very important to spread it among different providers. If all the applications from mail to calendar to word processing to web analytics are with a single vendor, any lockout like the one Zoli mentioned in his post will result in cutting of all the apps used by the small businesses. It is important to select different vendors for different apps so that any lockout will not bring down the entire business to its knees. For example, a small business can pick Google Apps for email and calendaring needs, Zoho or Thinkfree for productivity apps, Pbwiki for Wiki needs and Freshbooks for accounting needs. Such a diversification of vendors ensures that the entire operation of a business is not brought down by such lockouts or outages. I do agree that it is a big hassle to manage so many different vendors but it is nothing compared to the loss a business will incur if the whole operation is brought down by lockouts or outages.
I’m involved in several small businesses as well as doing advisory work for a number of others. The common element that all SMBs have is very limited time resource. They’re not in a position to have significant IT support and, in the case of your typical “Joe the plumber” every extra mouse click is another barrier to them getting out and doing the work that actually brings in the dollars (and as an aside – for the example I used of Joe the plumber – that’s a lot of dollars – how does an unregistered plumber earn over USD250k/annum?).
I actually believe that having all ones applications provided by a single vendor, or at least aggregated by a single vendor, is a good strategy for SMBs. It reduces the complexity, creates one “point of call” for service and simplifies their experience in terms of user interface and functionality.
No the real issue here is one of SLAs. There is no problem having all your SaaS needs met by one provider – so long as that provider has good reliability backed by an SLA – and this is the issue with Google – the fact that it goes down is bad enough, but the fact that there is no service or support when it goes down is criminal.
So… we can’t both be right here. My vote is for the minimum number of application providers for SMBs. Krish’s vote is for diversification – what do you, the community, think is the best approach?

Don’t fall over Ben but I agree. The one stop shop is the winner in the small business market. Keep it simple, satisfy the key needs and integrate.
Jim – I’m staggered – I never thought I’d see the day!
Ben, I agree with you and Zoli about the hassles of multiple vendors. Of course, it convenient to deal with single vendor. But, both you and Zoli are taking a short term view, which could very well be important for small businesses, but I am taking a long term view. It is about having standardized APIs, open standards for protocols, etc.. Over a period of time, I am going to explain my philosophy on this and I will keep revisiting this again and again. I have seen few others talking about it too. Hope I could convince you guys then
Regarding the lack of IT department in small businesses, it will be there soon. This is not the long term view I mentioned above. I know of couple of companies in India trying to do this, ie. trying to be the IT dept. of small businesses with presence in the cloud. It is just matter of time before India becomes the backend of small businesses. This could very well be the short term solution for small businesses. For few dollars a day, they could have an IT department sitting in India helping them with a diverse set of vendors, taking care of the backend operations and, if needed, even integration.
Probably, I am not putting it well here but I promise to write about these ideas in my posts in the coming month or two. Hopefully, I could convince you then
.
Krish, don’t convince us too fast, that way we’ll have a facsinating conversation thread here
[..] And Taking the Dissenting View… [..]