What you look for when you host big data operations on the cloud is decidedly different from what you look for with the hosting of normal apps and data.
There are clouds that specialize in running mission-critical applications like ERP (enterprise resource planning), and software as a service (SaaS) provider clouds that specialize in providing services to specific industry sectors such as finance or logistics.
Likewise, when it comes to managing large sets of data that are a combination of structured, unstructured and semi-structured data, there are clouds that specialize in big data.
The clouds that manage big data have the ability to quickly scale up large amounts of storage and processing. They can invoke big data technology that is designed to ingest many different forms of data and then perform algorithmic operations against the data.
Some clouds excel in offering infrastructure as a service (IaaS), which provide enterprises with all of the infrastructure and tooling they need to perform their own big data operations. In other cases, cloud big data service providers offer consulting and expertise and outsource staff as well.
Thus, it can be difficult for companies to determine what selection approaches and criteria to consider when they look for a big data cloud provider.
SEE: Microsoft Power Platform: What you need to know about it (free PDF) (TechRepublic)
1. Understand what your big data processing needs are
Understanding where your organization fits on this spectrum is the first thing to do before you look for a big data cloud services provider.
There are large enterprises that are far along in deploying their big data strategies. They have on-staff expertise and have already successfully deployed big data modeling and analytics.
Their interest in big data cloud services might be purely to scale out infrastructure as they need it, so they don’t have to make expensive capital investments in their data centers. In a situation like this, an enterprise will likely seek out a cloud big data services provider that has the ability to scale out storage and processing infrastructure on demand.
At the other end of the spectrum are companies that lack internal expertise and resources for big data but still need to process it. These companies are more likely to look for a big data cloud provider that provides both infrastructure and add-on services such as consulting, data management and even the ability to develop and maintain big data applications for the company.
Between these two extremes are companies that want a combination of both self-managed and provider-managed big data services.
2. Align yourself with a big data services provider that is compatible with your existing IT infrastructure
If you’re already running big data applications on a particular platform, it’s important to find a big data cloud services provider that also supports it.
In addition, if you have a set of security and governance standards you internally apply to big data, it’s important to find a big data cloud services provider that can provide the same level of governance and security, along with utilities and tools in the cloud that can affect those governance and security standards.
3. Review the cloud provider’s tools and resources
Does the cloud provider have a robust set of tools and utilities your staff can easily learn and use to manage your big data operations in the cloud? And if you’re a company that lacks big data expertise, does the big data cloud provider have consulting and outsource staff available that can manage your big data for you, based upon your specifications?
4. Don’t forget about the network
The cost of reliable data communications between your big data cloud provider and your own data center isn’t cheap. It’s helpful if your big data services provider has a facility within your geographic region. A more localized cloud facility can lessen costs and also the possibility of service interruptions.
5. Will the big data cloud service provider work with you on cost?
A recent Pepperdata survey revealed that over one-third of companies were exceeding their cloud budgets for standard services by up to 40%.
In the big data cloud world, determining cost for processing and storage is even more difficult because there is so little cost history. You might not know how much big data your application will ingest or process or how much you will need to pay to store the data.
A final word
The good news for companies seeking big data cloud services is the number of mature cloud providers that can deliver a wide array of big data services. Among them are Amazon AWS, Microsoft Azure, Google, IBM and others.
Once you understand your company’s present and future big data requirements, you can seek the vendor that is most aligned with your IT direction.