We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.

The Connected Web

Phil Wainewright

Why Bother with Multi-tenancy?

Vote 0 Votes

Many people are starting to question whether the purest form of provider-hosted multi-tenancy is really necessary. Some people have even thought that I was questioning it myself, when I reported Intalio's launch a couple of weeks ago of a multi-tenant application platform that customers can deploy on-premise. I was certainly glad to fuel the debate, but I come firmly down on the side of on-demand multi-tenancy — I believe that on-premise equivalents are a poor relation, for reasons that I'll explain shortly. But first, let's review what other people have been saying.

Last week I kicked off a debate in the ebizQ Forum around the question,
Should SaaS Vendors Offer On-Premise Options?. The responses covered a surprisingly broad spectrum of opinions, but the one I thought was the most insightful was from Tarak Modi, who suggested the question ought to be, 'Can the SaaS Vendor Offer On-Premise Options?' His question highlights how easy it is to talk about this on-demand versus on-premise question as if it's just a matter of degree, some kind of scale in which a 'hybrid' solution is a half-way mark. His point, though, was to draw attention to the fact that many SaaS applications are architected in a way that doesn't allow an on-premise deployment.

There's been plenty of talk about hybrid models in the past week. My blogging colleague Andrew Yee drew attention to Jive Software's decision to "leapfrog ... the SaaS model in favor of federated private and public clouds," as described in a InformationWeek blog posting. This reports Chris Morace, SVP of products, explaining that Jive found "multitenant solutions weren't getting true enterprise penetration," and he goes on to opine that "the enterprise will adopt around virtualized clouds, and multitenant will have a sweet spot for SMBs."

Last week also saw the announcement of Tibco Silver, a development platform designed to help enterprises build, deploy and manage cloud applications — but which, although offered as a cloud service, is set to be sold as conventional licensed software. Tibco, too, believes enterprises want to control and own their own infrastructure, and wants to give them the tools to do so, even if from time to time they may want to requisition the cloud as an extension of their in-house infrastructure.

The trouble with all these compromised models and captive clouds is that people are worrying so much about what they might lose when they adopt true SaaS and cloud solutions that they don't stop to think what they might be giving up by staying on-premise.

In a comment to my Intalio write-up, Appirio's Narinder Singh noted the loss of the instant feedback loop that multi-tenant providers benefit from when operating their applications on behalf of customers: "... ability to see how all customers are using all of your system at any time so you can improve the right part of the products, etc."

Then last week, Zoli Erdos on CloudAve wrote about The Hidden Business Model in SaaS, in which he discussed the potential for multi-tenant providers to collect aggregated data from their customers' use of their applications, "not simply representing additional revenue sources for the SaaS vendors, but also enabling them to deliver enhanced services to their customers, services that were simply not possible in the previous, behind-the-firewall fragmented data model."

Because multi-tenancy is newer than the conventional ways of doing things, we haven't yet fully appreciated all the benefits that accrue from having this global view of customer activity, both in terms of fine-tuning and enhancing the product and also providing feedback and benchmarking back to customers on their use of its functionality. These and other facets of multi-tenancy do make it worth bothering with, and we don't yet know the half of the additional value they bring.


Phil, when an instance is managed on premise, an Intalio expert will be onsite and will make sure the instance is updated constantly as well as giving feedback to avoid this problem as much as possible.

Here is a link to the complete offering:

Phil -
I find this mixture of Multitenancy and Cloud computing very strange.

Multitenancy is an application architecture. Cloud Computing is a hosting model. Why would we ever try to trade off one for the other. Doesn't this say more about the ISV's thoughts on what they are really delivering to add value?

I'm working on a product development that is planned for multitenancy. I know it will also have to go on premise. It can do that. HOWEVER, changing the hosting model will also change the services provided. We will not guarantee updates etc to an on-premise client. That is part of the cost of separating the services from the software.


On-premise vendors have to make the leap towards a unified architecture for their on-premise and on-demand software. We were early to recognise that we had much to gain by enabling our web-based HRMS (Powerapps) with multi-tenancy in order to have a Saas offering. This allows us to provide all the required features to both the target segments without maintaining two products. Our Saas based HRMS (www.emportant.in) offers ALL the features as the traditional on-premise system. Net result - productivity gains for us, no-compromise for our customers.

Sandeep Todi

Phil Wainewright blogs about how businesses are using the Web to get better plugged into today's fast-moving, digital economy.

Phil Wainewright

Phil Wainewright specializes in on-demand services View more

Recently Commented On

Recent Webinars


    Monthly Archives