It's a matter of simple revenue and cost economics of cloud services. Most cloud app providers' revenues come from selling monthly or annual per-seat subscriptions, which bring in just a fraction of the annual revenue that an on-premise software license with comparable functionality would provide a vendor. The challenge for selling software subscriptions, then, is to reduce operating costs in order to manage with less; otherwise the provider may end up doing much more than an on-premise vendor does, such as maintain multiple versions, run multiple infrastructures, maintain customer-specific code, and perform upgrades, but with fewer dollars. Multitenancy provides the answer, because it spreads the cost of the infrastructure and labor across the customer base; in fact, customers sharing resources right down to the database schema is ideal for scaling.
The economies of scale get even better as the provider adds customers, and customers benefit from this scaling up. As the cloud app provider's costs decrease, it has more room to innovate and grow, thus delivering more value. Even if customers' costs don't drop, over time they should expect to see more value, such as increased functionality.
So, what's the debate about? Those who say multitenancy isn't necessary to making the cloud model work are typically companies that have long made money from on-premise software and don't want to cannibalize their existing revenues. They might offer a subscription for their single-tenant application, but this could simply be the software license, maintenance, and hosting fees divided into monthly payments which almost certainly would be much higher than a comparable multi-tenant application.
What's even more interesting is the "unsure" camp in this debate. These are typically the traditional on-premise vendors that decided to give the cloud a try. They often try to save money by using all or some of their existing on-premise infrastructure and practice for their cloud apps, by avoiding the investment in a new technology infrastructure that supports multitenancy. However, the high cost of replicating and maintaining instances for each single tenant (or customer) eventually catches up with them. They are forced to try approaches where they can share some of the infrastructure, but their fundamental affinity to the old on-premise model usually proves to be a stubborn barrier to changing their software, infrastructure and culture to fully support a shared model. And if they keep on this single-tenant path as they scale up customers, their margins get lower as each new customer sucks up more resources.
For a customer, there can be trade-offs in sharing an application. Think of it like living in a condo versus a house; one management company serves all the tenants, and you may not be any more special than any other tenant. Everyone gets upgraded with a new version at the same time, for example. But for many types of apps, the cost/value formula of multitenancy is the best answer.
Alok Misra is a cofounder of Navatar Group, which provides Cloud apps for the financial services industry and helps software companies launch and support SaaS.
Attend InformationWeek's Virtualization Virtual Event to learn advanced strategies on how a top-to-bottom transformation will increase system availability and ensure better business continuity. It happens Wednesday, March 3. Find out more (registration required).