June 9, 2021
A Case for Cloud – Costs to Consider
It can be difficult to determine if a move to the Cloud makes financial sense for your company, but there are some techniques and tools that can be applied to build a business case. Jeff Hutchens is the Chief Information Officer at a leading IT Outsourcing firm EAG 1Source, and he and his team specialize in deploying IT Strategies for the Oil and Gas industry. In this article, Jeff highlights the cost benefits of migrating to Cloud, different cost models, and tips for how to avoid runaway costs.
Your Data Center is Costing too Much
On-going maintenance and hosting costs add up over time and do not benefit your bottom line. Jeff sheds light on additional implications to consider:
- Current hardware spend and refresh cycle: Hardware that exists in your data center must be replaced on regular basis which can be a large capital outlay. Refresh cycles vary but are typically done every 3-5 years, which includes the cost to replace servers, network equipment, firewalls, in addition to maintenance costs of the hardware.
- Current virtualization spend: The need for specialized virtualization software can be eliminated if your company migrates to cloud as this functionality is natively available within both Azure and AWS.
- Environmental: Data centers are expensive to run and maintain. If you host the systems in your office, you are incurring the cost of the space required, electricity, and air conditioning. Additionally, larger facilities require specialized power which drives cost up exponentially.
- People costs: Depending on your company’s current Data Center support model, people costs can be reduced by leveraging Software as a Service (SaaS) or Managed Service Provider (MSPs). In some cases, this cost can even be eliminated since the base hardware maintenance and replacement cost will be removed when a Cloud strategy is adopted.
- Current Software Maintenance: Moving to a SaaS model can potentially reduce your overall software maintenance as the cost is factored into the vendor’s overall pricing.
In some cases, depending on your security policies, your company may need to purchase virtual equipment, which is still a fraction of the cost required to run a Data Center.
Pre-Paid or Pay as you go? It Depends.
There are two models that are generally available with most Cloud Vendor providers: Pre-Paid and Pay as you go.
- Pre- Paid: Reserved/dedicated capacity where some or all the resources are pre-paid (typically for one or three years) which provides a significant cost savings. This model will provide the most value if your compute needs are relatively, stable and you can pay the annual or multi-year fees in advance.
- Pay as you go: This is the simplest model as you only pay for the resources when you use them. It allows companies to scale up and down as needed and is an ideal approach if you cannot fully commit to a Cloud migration, or if you have more volatile infrastructure needs.
- Hybrid Software: This can be used with either the pre-paid or pay as you go models. Cloud providers can provide the license for the operating system/database (e.g. Windows/Linux and MS SQL) or companies can purchase and provide their own. It is less expensive for Companies to purchase their own software but does require separate administration of the licenses.
These models can also be combined if you have a core set of services under the Pre-Paid model and the remainder under the Pay as you go model. Agents can be loaded on your current servers that will estimate what the equivalent service would be in the Cloud Vendor’s environment, and then use their calculators to estimate costs. Additionally, you may take the following routes when attempting to estimate costs:
- EAG 1Source has a pre-built cost model to help companies understand what the true costs would look like when considering a move to Cloud services.
- Microsoft and Amazon provide tools to help estimate size and costs for the migration to their Cloud environments.
- Your Cloud Services Provider or Managed Service Provider can help with these estimates.
Avoiding Runaway Costs
So you made the decision to move to the cloud. Cloud Providers can provide almost unlimited resources in terms of compute, memory and storage and they make it incredibly easy to provision. This can quickly turn into a large, unexpected invoice that you then must explain to your company’s leadership. Jeff Hutchens mentioned, “Moving to the cloud isn’t just about what the technology can offer you. With all the capabilities, companies can easily overlook what they really need.” He encourages companies to consider the following ideas to evade unexpected costs:
- Process: Adding processes require sign-off when scaling up resources. Once in the Cloud, it is incredibly easy to throw additional memory or processors at a problem. Ensure you have an established process that estimates these costs and requires sign-off prior to incurring costs. The threshold for sign-off will vary with each company, but this can help ensure there are no surprises at the end of the billing cycle.
- Performance: Some solutions may not be best suited in the Cloud, which include graphically intense applications or systems with very large data sets. In the Energy Sector, certain Geology solutions have not always performed well. This can typically be addressed with specialized services, but the costs may outweigh the benefits. Jeff recommends handling these scenarios on a case-by-case basis to determine what is best for your Company.
- Migration/Cost Control: Outside of the cost models mentioned above, there are additional ways to help cut costs:
- Continued Consolidation/Virtualization: Use the migration to the Cloud to further consolidate and virtualize systems which will keep Cloud costs low.
- Headless Services: There are several services that don’t require a base Operating System. For example, you can provision a SQL server that does not require Windows. Not all applications can utilize these technologies, but when applicable it can help reduce the number of servers and overall costs.
- Scripting/Automation/Scaling: System scaling can be used to dynamically scale resources up and down based on when they are most needed. Some applications may not need any resources after business hours or on the weekends. These services even be shut down completely and then automatically started during business hours.
- Containers: More applications are being developed that can take advantage of microservices and containers. This is another option for scaling application resources dynamically for when they are most needed.
EAG 1Source enables companies to power their I.T. platform with full-scale technology services, reengineered strategies, predictable costs, and unmatched customer service with every interaction. Through our 17+ years of experience and team of skilled experts, we’ve analyzed every corner of I.T. to unlock and deliver greater value to our clients. Our services range from cyber security management to server management, cloud migration, architecture rationalization, and more.