By: Bill Tolson, Vice President, Archive360
Software as a Service (SaaS) is the most well-known cloud computing service—but that doesn’t mean it’s the best model available today. Platform as a Service (PaaS) grew in popularity in 2018 due to gaining maturity, emerging as a preferred strategic platform for many organizations that conduct digital business, according to industry analyst Gartner. Ongoing innovation has both broadened and deepened PaaS capabilities, leading Gartner to predict a number of key trends for 2019:
- PaaS solutions are on their way toward ultimately using a spectrum of cloud services
- PaaS is poised to enable custom machine learning and AI capabilities
- There will be far-reaching impacts of serverless PaaS architectures
Gartner also notes: “By 2021, intermediation, including integration, event brokering and API management services, will be one of the top three PaaS responsibilities and essential in large-scale digital business implementations.” With this in mind, let’s explore the differentiation between these two types of cloud services and examine the pros and cons of each to help you determine which is right for your enterprise’s data storage and information management needs:
Reconsidering the Common Option
While SaaS has been the go-to choice when conducting cloud-based business—providing companies with access to a third-party vendor’s cloud-based software and storage—there are reasons why you may want to rethink going blindly to the most common option. SaaS has a number of significant drawbacks to its implementation and ongoing usage. The disadvantage that tops many IT administrators’ lists? Higher costs. There are several other notable problems as well relating to flexibility, security, and various other complications:
- Lack of flexibility. In most cases, SaaS offers only a single application—in other words, no choice. Paired with this one-size fits all approach is the fact that it generally requires locking in to a third-party cloud vendor. Many of these providers try to ensure that this lock-in sticks by storing their customer’s data in a proprietary format, making enterprises dependent on continuing the relationship. Such lock-in limits options for an enterprise, since you’re then stuck with whatever applications are provided, which can lead down the road to not having the specific functionality you need. This situation whereby the third-party vendor controls if you can migrate out of their system, at what speed, and at what cost is known as “data ransoming” or “data prison” for obvious reasons.
- These realities cause a trickle-down effect leading to a host of other issues. When data is stored in a proprietary format, the enterprise is hamstrung in that it can’t employ the latest analytics tools, putting the company at another disadvantage. SaaS also leaves you with no ability to customize performance-based business requirements.
- Security concerns. Since SaaS is a shared platform that serves multiple tenants, the model naturally raises questions about data security. As the customer, you’re forced to adapt to the third-party vendor’s choice of security safeguards, which may be insufficient and not the best choice to meet the specific needs of your enterprise.
- Frustrating complications. Data export can be complicated and frustrating for various other reasons including:
- The third-party vendor’s ability to throttle export speed (resulting in requiring what’s in essence a ransom fee to manage this speed)
- Potential unavailability of data residence
- Limits to the portability of your data (for example, if the cloud provider goes out of business or there is another reason why you want to change service providers, you may be locked in by contract)
- Since the cloud-vendor provides a specialty application that’s designed for that vendor’s own cloud environment, integrating other applications can produce headaches or be impossible
- In legacy migrations, data chain of custody can become problematic, potentially causing legal or regulatory issues
What draws organizations to SaaS service providers in the first place is the initial ease of entering a contract: all you have to do is migrate your data into the cloud of the third-party vendor you choose, and you are provided with software and applications through a subscription model. It sounds great on a surface level to know that you’re using applications that reside on a remote cloud network through which you can easily access from almost any internet-connected device via the web or an API, storing and analyzing data and collaborating on projects. There’s no need to manage, install, or upgrade software since the third-party provider handles that. The costs may be higher but they are predictable, and the cloud provider conducts all system administration.
But be wary of taking a shortcut on the front end without recognizing the consequences of your choice: it’s a major downside to then be locked in to whatever applications are available through the provider without access to others you might need, and to have to rely on the performance parameters that are set, as well as security systems that may be sub-optimal.
“PaaSing” Along New Benefits
PaaS offers another option that provides companies with a cloud environment in which they can develop, manage, and deliver applications. Generally offered as a subscription model based on number of users, the huge advantage of PaaS compared to SaaS is that it’s your data, in your cloud, behind your firewall and under your direct control and management.
If you want to implement agile methodologies, then it’s smart to consider PaaS. With PaaS, you as the customer enter a direct agreement with the cloud platform provider, not with a third-party cloud vendor. Adopting a PaaS solution significantly lowers capital expenditures over on-premise systems and provides a lower total cost of ownership. You control the infrastructure costs and gain fully elastic scalability for expansion as needed.
Cloud solutions based on the PaaS model allow your enterprise to retain direct ownership and management of your own data—for example, by working with your Azure tenancy. While SaaS offerings require uploading your data to a vendor’s cloud where it is often converted from its native format, making it tricky to move out of the SaaS system, you retain ownership of your data when using a PaaS model. This can help with everything from regulatory compliance and eDiscovery requests to information management and elastic search, making the process of data storage and management faster and more consistent than SaaS solutions, as well as legally defensible and customizable to your particular performance requirements.
The PaaS cloud model has many other advantages as well, including:
- Avoids the limitations of a single application designed around a third party’s environment; you can leverage a range of prebuilt tools to develop, customize, and test your own applications.
- Provides an easy way to test and prototype new applications, saving money when developing new applications and speeding up application release.
- Benefitting from PaaS providers publishing many up-to-date services that you can consume inside applications.
- Receiving the PaaS provider’s management of security, operating systems, server software, and backups.
Admittedly, PaaS requires you to commit to more responsibility for your data and applications than a SaaS setup, the latter of which relinquishes this responsibility to a third-party vendor. With PaaS you’re gaining more control over your own data to ensure enhanced application performance, increased security along with a predictable pricing model that works for your business.
About the Author
Bill Tolson currently serves as vice president for Archive360 and is focused on the archiving, migration, governance, regulatory compliance and cloud-based storage of data. Bill has extensive experience in eDiscovery and archiving/information governance from both a marketing and customer perspective.