SaaS is short for ‘software as a service’ and represents a software delivery model. Compared to the traditional on-premise software model, the application and user data are hosted in the cloud which users then access from their personal computers. As a model of software delivery, SaaS represents a branch of the overarching cloud-computing model.
Cloud-computing works by delivering computing services as a utility over the internet rather than as a product; the two other branches being Infrastructure as a Service (IaaS), in which physical or virtual machines are licensed for running firewalls, storage, networks, etc., and Platform as a Service (PaaS), in which providers license out programming language execution environments, databases, web servers, etc. to software developers.
SaaS has been quickly growing as the preferred way of using enterprise software for companies small and large. But, looking back on its history, cloud-computing is still a fairly new concept. Only in 2008 the first open source cloud platform Eucalyptus was launched, paving the road for private cloud hosting. Business and enterprise applications sprung to live in 2009 following the boom of cloud-computing. In this article we look closely at the strengths and weaknesses the SaaS model offers to the enterprise market.
Software is ubiquitous in today’s business world. Computer and web applications are integrated anywhere and everywhere the budget can afford because they maximize efficiency – from tracking shipments across multiple countries, to managing large inventories, to improving customer service or knowledge base management. Ever since software innovation entered the business world it has been treated as a physical component that companies have to own and run on their internal infrastructure or computer network.
An analogy for this relation could be a product manual or a how-to book for using a product. Before the inception of cloud computing, a company had to supply a copy of the manuscript to every employee to ensure that everyone was on the same page and could collaborate. Today, the instructions, or rather the software and user data, can be accessed on the cloud via an internet connection.
The SaaS model has become prominent over the past few years leaving the traditional software model competitively at a loss and lacking in many technical respects for two main reasons. First, the technology has arrived to support cloud based and ‘on demand’ computing – underlying technologies that support web services and service-oriented architecture have matured and new developmental approaches have become popular. Likewise, broadband service has become widespread around the world and employees today are more wired to the web than ever before.
Second, the benefits that the SaaS model offers are continuously outpacing those of the traditional model. For years, companies have run software on their own internal networks. Any updates, upgrades, or fixes to the software had to be done on premise and required the use of company hardware and in-house IT specialists. SaaS has removed the need for organizations to handle the installation, set-up and often daily upkeep and maintenance. By repackaged traditional computer services, applications and the data inherent to them, and managing everything from a centralized cloud server, SaaS has revolutionized the traditional on-premise software model.
The growth of the SaaS model shouldn’t come as much of a surprise. Simply put, the new model provides the same services for a lower price by outsourcing, or what in the tech industry has been called Business Process Outsourcing (BPO). India has been renowned for picking up the slack, US$10.0 billion in 2011 to be exact. And in fact, BPO has become an extremely important tech strategy with a 7.3% IT outsourcing rate since 2003. And, if a company can save money by outsourcing its business functions, customer service duties, IT responsibilities, then it can surely outsource the operation of its software. SaaS is a perfect example of BPO. It delivers the functionality of a full software system without the overhead of operating the application. It’s a win-win situation, except maybe for those in the IT Department.
With the hyper growth of the unreliable and unnecessary applications of the dot-com bubble fading into the dark ages of the tech era, today’s businesses are greeted with a myriad of competing software providers that offer real business solutions. The recent – or maybe not so recent in tech years – growth of the SaaS model has demonstrated that first and foremost, SaaS costs much less to deploy, upkeep, and update for manufacturers which translates into saving for consumers.
SaaS alleviates the costs of traditional licensing fees and decreases the cost of IT support and infrastructure maintenance that all software comes bundled with. The traditional software often depends on an internal IT infrastructure on which the services can be hosted. This adds another cost layer. Facilities and their operation – all the switches, servers, routers and storage devices for the data centers produce a cascading cost which the SaaS model conveniently circumvents.
In addition to fewer upfront costs, SaaS is also easier to customize once it is deployed – no need to reinstall software on every machine in the office, everything can be done over the cloud. For instance, a recent study by the Software and Information Industry Association (SIIA) showed that the total cost of ownership of a SaaS CRM software package was only 10 – 20% of the cost of a traditional software package for one of largest CRM providers on the market. The numbers demonstrate that software providers and businesses alike can be better off with a transition to a SaaS model of operations.
SaaS applications are predominantly based on a multi-tenant architecture where a single version of the software is used for all customers. Some other SaaS providers rely on a virtualization model and create different virtual copies of their application to cost-effectively manage a large number of customers. Contrasted with the traditional hardware software, which utilizes multi-instance architecture where separate hardware instances are set up for different client organizations, multitenancy is structurally more complex but offers three reasons: cost savings, data aggregation, and release management.
The cost savings that the multi-tenant architecture creates result from consolidating IT resource into a single operation. Every instance of an application incurs a certain amount of processing and memory overhead which can be substantial when there are many customers. Multi-tenancy reduces costs by consolidating everything in a single software instance and amortizing it over many customers. One problem that might sometimes arise is when scaling becomes necessary. Scaling the entire software structure rather individual parts is more difficult, however, most SaaS applications rely on third party cloud platforms – Amazon’a EC2, Heroku, Blue Box, to name the big ones – which allow for secure and easy scaling.
The second benefit of the multi-tenant architecture is the data aggregation and data mining opportunity that a single instance provides. Whereas multi-instant operations require mining of potentially different database schemas, multi-tenant architecture stores all the information on the same database.
Lastly, multitenancy simplifies the release management process. In a traditional release process, packages containing updates and database changes must be distributed to desktop computers and server machines. These packages then have to be installed individually on every machine. With the multi-tenant architecture, the package only needs to be installed on a single server from which the application runs. This greatly simplifies the release management process.
In February of 2012, Veracode Inc., one of the leading security testing firms for cloud-based applications, published piece titled “Outsourcing the problem of software security”. Veracode commissioned an analyst firm Quocirca to examine how UK and US businesses are deploying in-house developed and commercially acquired software. Most importantly, Quocirca looked at measures that were in place for ensuring software security.
Quocirca interviewed 100 medium to large organizations in the financial services, manufacturing, retail, distribution, transportation, and other commercial sectors. The first observation tantamount across all business sectors was that the use of SaaS and mobile applications is not now widespread. The concern brought up by this is the breadth of security issues especially because SaaS applications are predominantly web-based. Customers are realizing the risks and are increasingly demanding security, with 50 percent of US SaaS users and 20 percent in the UK demanding that their software was secure.
For a long time, businesses have been weary of outsourcing their software capabilities because it meant that the security of sensitive information would be left in the hands of a third party. However, modern technology provides better remote security and data redundancy tools then were available before and SaaS vendors are able to protect customer data better than most customers do on their own. In fact, it has been shown that having an on premise network often enables companies with a fake sense of security. While the data is stored locally it is still vulnerable to a wide host of attacks which are usually shielded by nothing more than a firewall.
Many businesses turn to managed security services providers who analyze security logs and prevent inappropriate penetration or data leakage. Often times these are valuable services that a company might not be able to provide internally. Having a managed service for security in the cloud, however, makes security affordable for smaller business. Instead of having on-site appliances, emails can be filtered for spam and viruses in the cloud by a provider.
In a nutshell, while cloud security has been a booming topic among experts of network security, the sophisticated protection that SaaS applications provide today is becoming quite dependable and, on some level, is even surpassing the level of protection that on-premise counterpart applications offer.
Another concern that companies are vocal about is the level of flexibility when it comes to development. Working with a cloud framework might seem more problematic because the data is hosted on a remote server which probably will not have the same environment as an in-house system. However, IaaS and PaaS models offer resources that can mirror in-house servers without hindering the development process. In addition, these servers may be more available and accessible than those of the corporate data center.
Elasticity of the cloud brings better system availability even with fully redundant data center servers. Sony is using IaaS for fast spin-ups of video game demos that often support several hundred players at a time. Today, Sony is in the process of moving to a private cloud in order to avoid resource bottlenecks when they spin up demos of new features. Sony has noted that costs for the private cloud will be about equal to the data center. However, the team will benefit mainly through resource elasticity and accessibility, since developers would otherwise have to go to IT to procure resources.
While SaaS applications are powerful and, if the trend continues, will soon replace most of the antiquated business software platforms, not all companies can or should make an immediate switch. Any IT department knows that the tech market is filled with UI/API tweaks, spec upgrades, newer versions of software, etc. that produce a never ending conveyer belt of mediocre improvements. Switching from one type of technology to another is a timing decision that takes foresight and planning.
The first step that a company should take is to evaluate how important the transition from a traditional on-premise to a cloud hosted SaaS model is. Cutting or conserving costs is an obvious starting point. For established organizations, the migration usually begins with applications like –forums, knowledge bases, email, document management, and other help desk services.
For organizations that are in the early stages of development, SaaS is the best place to start. Applications are plentiful and investments in start-up time and costs are minimal. With SaaS, the entire cloud stack is all bundled into one: infrastructure, management, installation, and other product usage. Moreover, SaaS offers variety – it mirrors anything that can be found in the application world today.
At Safeharbor Knowledge Solutions we understand the power of SaaS. That is why our SmartSupport platform and accompanying tools, such as the SmartTEST Article Optimizer, are entirely SaaS. SmartSupport integrates into any web environment in minutes and can be accessed from anywhere. We take care of all the back-end and IT needs so you can run your business without the worry of downtime or software updates. Scaling is seamless and maintenance is free. SmartSupport takes full advantage of the multi-tenant SaaS architecture. Because all your data is kept in a single database, you have immediate access all of to your community forums and knowledge base content from one place. Coupled with the power of the SmartSearch engine, the search capabilities of SmartSupport are extremely robust. Click here to learn more about SmartSupport!
Author: Dmitry Minyaylov