by Dow Smith
The “Cloud” looms large over CIO playbooks and budgets with the opportunity to capture leading cloud providers economies of scale, growing services, and capabilities. 2018 saw many OpEx budgets for cloud spend exceeded, and not by just a little. The question became would cloud spend decrease as a result?
It appears not, as migrating infrastructure, platforms, and applications are still accelerating. Katie Costello and Sarah Hippold of Gartner forecast that worldwide cloud spend will grow by 17.3% in 2019. What is even more astonishing is that, of the cloud categories of Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS), and Software-as-a-Service (SaaS), IaaS is forecast to grow 27.6 percent in 2019 to reach $39.5 billion, up from $31 billion in 2018.
Microsoft has reported its fiscal earning for Q1 2019, and it is not surprising to see Azure revenues up by an incredible 76%. Similarly, Amazon Web Services (AWS) cloud business grew 46% in the third quarter alone, and it now accounts for $2.1 billion in operating profit and 56% of total operating earnings for Amazon.
How can we consume more services and spend less?
Clearly, cloud adoption has accelerated and will continue to grow for the foreseeable future. In this rapid migration to the cloud, have we left money on the table? Where did we leave it? What lessons can we immediately adopt to recapture it?
The good news: the leading cloud providers continue to very rapidly evolve their offerings with over 140 services now available from AWS and nearly the same number offered by Azure. Included in that evolution are tools that afford cloud stewardship over costs and consumption thresholds.
In addition, we are maturing our migration planning capabilities. Here are five important areas that will need time investments when considering IaaS and PaaS inputs for cloud architectural designs:
Understand Available Services:
Including accounts, subscriptions and services models
Cloud providers have reserved instances for dedicated use models. While the resource may not be shared, a substantial discount can be realized with Reserved Instances. Conversely, inconsistent and bursty compute needs may be better and more cost-effectively served with Spot Instances which rely on additional available cloud provider capacity at a much lower cost.
Utilize tags within your resource groups. Tags can be applied to departments or different teams to provide different cost reporting views. AWS Cost Explorer and Azure Portal provide views into resource usage and data usage reports. Thresholds and alerts can be set to help IT stay in front of cloud spend and burn rates.
Move Test into Design Phase:
Cloud instances can be tested quickly and inexpensively
There are many interactions between, among, and across instances of security, compute, storage, and network designs. Providers have options available to you based on the geolocation of your data, your data sensitivity, and your application tiering methods. IT designers can easily over-engineer lower tiered applications with good intentions, but with costly results. One real-life example of this was a well-intended application owner who told its cloud builders the design input for data sensitivity was “high availability,” while the business had graded the platform a “best effort” data recovery tier. The result was an overly expensive design across two cloud provider regions for both compute and storage, significantly driving up monthly services fees. This is just one of many reasons why it is a good idea to include the business owners in design review sessions prior to testing and moving into production.
Investigate then Delegate:
Cloud-native capabilities are growing at a monthly pace
There was a time when healthcare organizations, for example, thought that the cloud was a threat to security and compliance. Today, many healthcare enterprises’ compliance and reporting objectives are being met through AWS and Azure capabilities. The following are some of the conforming reports available for your cloud environments from both providers:
- Service Organization Controls (SOC) 1, 2, and 3
- ISO/IEC 27000 Series Reports (various)
- General Data Protection Regulation(GDPR)
- Multi-Tier Cloud Security (MTCS) Singapore
- National Institute of Standards and Technology (NIST) Cybersecurity Framework (CSF)
IT will need to adapt its current procedures for the cloud
IT will need to consider the native capabilities offered and not simply try to apply on-premise procedures for provisioning and managing infrastructure and computing resources. Standing up servers can now happen in minutes, not days and should be automated based on computing vertical and horizontal power as needed, not based on traditional capacity estimates for capital spend. Some key planning questions include:
- Who has the authority to provision additional compute power?
- What is the approval procedure?
- How can we delegate authority and maintain compliance?
Do not set it and forget it (or the monthly bill will surprise you)
Estimate your costs in advance with the native toolsets available. Compute time, availability, uptime requirements and data egress all drive costs. Monitoring for both security anomalies as well as underutilized instances is a leading practice. Daily and weekly reporting is available and should be an internal standard when considering the value of proactively managing costs. If staff skills are an issue, look to your partners or a new 3rd party capable of delivering via its Managed Service offerings.
Cloud services can be commissioned and decommissioned immediately, which is one of the value propositions of the cloud to begin with. By running an agile support contract or staff, costs can be reduced, and resources optimized.
Summary: Buying more from cloud providers can cost less in 2019
It is a good idea to review IT plans, playbooks, and procedures while the IT gears are still turning. Check with your partners to see if they can assist you with cloud health-checks in Cost Management, Automation & Optimization, Security & Compliance or Cloud Business Continuity.
Whether your organization has already moved workloads into the cloud, is planning additional migrations, or needs to create hybrid instances for legacy platforms, 2019 can be a year for efficient cloud services growth. As always, iVision is here to help engineer a solution that is perfect for you. Contact us today to find out how.