The Value Proposition of De-Verticalization

The strategic logic for de-verticalization is illustrated by the trend which began in the 1990s of outsourcing facilities’ maintenance and janitorial services.

A company that specializes in–let’s say–underwriting insurance policies must dedicate its mindshare and resources to that function if it expects to compete at the top of its field. While it may have had talented janitors with the necessary equipment on staff, and while clean facilities are certainly important, facilities maintenance is a cost center that does not provide a strategic return on what matters most to an insurance company. Wouldn’t it make more sense for both insurance and janitorial experts to dedicate themselves separately to being the best at what they do and avail those services to a broader market?

This is even more true for a data center. The era of verticalized technology infrastructure seems largely behind us. Though it’s a source of nostalgia for us geeks who were at home among the whir of the server rack fans, it’s easy enough to see why shareholders might have viewed it differently. Infrastructure was a cost center within IT, while IT as a whole is increasingly seen as a cost center.

The idea of de-verticalization was first pitched as something that would save money and allow us to work more efficiently. The more efficient part was intuitive, but there was immediate skepticism that budgets would actually shed expenses as hoped. At the very least it would be a long haul.

The Road to Performance and Cost Optimization

We find ourselves now somewhere in the middle of that long haul. The efficiencies certainly have come to pass. Having the build script deploy a new service to a Kubernetes cluster on the cloud is certainly nicer than waiting weeks or months for a VM to be approved, provisioned, and set up. But while the cloud saves the company money in the aggregate, it doesn’t show up as cheaper at the unit level. So, it’s at that level where anything that can be shed from the budget will be a win to celebrate.

This is a good position to be in. Opportunities for optimization abound under a fortuitous new circumstance: the things that technologists care about, like performance and power, dovetail precisely with the things that finance cares about, like cost. With the cloud, they are two sides of the same coin at an almost microscopic level. This trend will only accelerate.

To the extent that providers of computational resources (whether public cloud, hypervisors, containers, or any self-hosted combination) have effectively monetized these resources on a granular level and made them available a la carte, performance optimization and cost optimization sit at different ends of a single dimension. Enhancing a system’s performance or efficiency will reduce resource consumption costs. However, cost reduction is limited by the degree to which trade-offs with performance are tolerable and clearly demarcated. Cloud resource optimization tools help organizations strike the ideal balance between the two.

Choosing the Right Cloud Resource Optimization Solution

With that premise in mind, selecting the right cloud resource optimization solution should start by considering how your organization wants to approach the problem. This decision is informed by overall company philosophy and culture, what specific problems or goals are driving the initiative, and an anticipation of where overlapping capabilities may fulfill future business needs.

If the intent is to solve existing performance issues or to ensure continued high availability at future scale while knowing (and having the data to illustrate) you are paying no more than is necessary, focus on solutions that lean heavily into performance-oriented optimization. This is especially the case for companies that are developing software technology as part of their core business.

If the intent is to rein in spiraling costs or even to score some budgeting wins without jeopardizing application performance, expand your consideration to solutions that offer a broader FinOps focus. Tools with a FinOps focus tend to emphasize informing engineers of cost impacts, and may even make some performance tuning suggestions, but they are overall less prescriptive from an implementation standpoint. Certain organizations may find this approach most effective even if they are approaching the problem from a performance point of view.

Now that many organizations have successfully migrated large portions of their application portfolio to the cloud, the remaining work is largely a matter of cleaning up and keeping the topology tidy. Why not trust that job to a tool that is purpose-made for optimizing cloud resources?

Next Steps

To learn more, take a look at GigaOm’s cloud resource optimization Key Criteria and Radar reports. These reports provide a comprehensive overview of the market, outline the criteria you’ll want to consider in a purchase decision, and evaluate how a number of vendors perform against those decision criteria.

• GigaOm Key Criteria for Evaluating Cloud Resource Optimization Solutions
• GigaOm Radar for Cloud Resource Optimization

If you’re not yet a GigaOm subscriber, you can access the research using a free trial.

The post Save Money and Increase Performance on the Cloud appeared first on Gigaom.

Furthermore, as cloud usage continues to outpace the rate at which IT operational analysts can be hired, automated optimization of these resources directly impacts the bottom-line of the cloud bill and the effectiveness of existing IT staff (who are freed up to work on higher-value business objectives). Taking an hour to determine whether a machine would benefit from less or more vCPU may seem hardly worth the time and effort but may reveal an imbalance that can generate significant excess spending or risk if neglected at scale. This is the kind of task the analytics engines within resource management solutions can handle effectively and expediently.

As you evaluate cloud resource optimization solutions, it’s important to keep the following in mind:

• Cloud resource optimization is closely aligned with the financial operations (FinOps) and cloud management platform (CMP) tooling categories, and solutions may lean in one of those directions with a strategy of providing a single solution.

• Private cloud, public cloud, and Kubernetes resources all require oversight and optimization, and solutions tend to be stronger in one area than another. You’ll need to determine where your resource challenges exist today and what improvements you want to have made 12 to 18 months from now.

• It can be beneficial to delegate resource and cost optimization to individual teams, with some limited central oversight. Individual teams are closely aligned with the performance needs of their applications and, if motivated properly and given the right tools, will ensure a balance is reached between cost and performance.

This is our third year evaluating the cloud resource optimization space in the context of our Key Criteria and Radar reports. This report builds on our previous analysis and considers how the market has evolved over the last year.

This GigaOm Radar report examines 11 of the top cloud resource optimization solutions in the market and compares offerings against the capabilities (table stakes, key features, and emerging features) and nonfunctional requirements (business criteria) outlined in the companion Key Criteria report. Together, these reports provide an overview of the category and its underlying technology, identify leading cloud resource optimization offerings, and help decision-makers evaluate these solutions so they can make a more informed investment decision.

The post GigaOm Radar for Cloud Resource Optimization appeared first on Gigaom.

Before the rapid adoption of public cloud infrastructures, it was common for application teams to overprovision their virtual machine (VM) resource requests, since provisioning could take a long time and they had little insight into future requirements. IT operations teams could manage the cost and resource overprovisioning issues by overallocating their infrastructure, and did so often by an order of magnitude.

Organizations today provision many more cloud resources across public clouds. These services effectively offer the opportunity to pay only for what is needed and dynamically scale up or down as the need changes. This often presents challenges with cost and resource usage. The question of precisely how much is needed becomes relevant in a new way. Application teams still overprovision their cloud requests, while public clouds charge for exactly what is provisioned, which leads to waste. Sometimes this lack of diligence is just an old habit, but even with effective processes and controls, it can be difficult to safely and precisely predict runtime requirements by hand. Any degree of overprovisioning translates into a wasted expense, but when managed by hand, it would be imprudent to not leave a comfortable margin for safety. After all, availability or performance issues would ultimately impact the bottom line more profoundly than overspending on a safety margin for resources.

Application teams deploy many more systems than they used to, as the provisioning time is instantaneous, and there might be very little oversight of their spending for cloud resources. Modern software architecture demands greater subdivision of monolithic applications into microservices, and this multiplies the potential overhead expenses incurred by safe resource provisioning. Automating resource configuration is a clear solution to this problem because it assures users that their software can effectively scale to any level without spending a dime on unneeded resources. This would be anywhere from impractical to impossible to do by hand.

Business Imperative
As organizations continue to deploy cloud resources across multiple public and private clouds, each with potentially different resource monitoring solutions, they need solutions that can provide a holistic view across all cloud resources, which will identify resource configuration optimizations that balance performance and cost. This includes the need to make specific recommendations and potentially act on those recommendations without burdening operations engineers.

Cloud resource optimization solutions have been around for years and provide reliable and mature management of traditional IaaS (such as the dynamic provisioning and configuration of VMs). For large organizations that have migrated legacy software from private data centers, this continues to be the most relevant capability. As cloud-native software development proceeds, it is important that these solutions keep pace with evolving paradigms, such as containerized deployments, and platform as a service (PaaS) architectures, such as serverless APIs, which inherently provide a more solid foundation for scale. This year’s report will take a closer look at that evolution.

Sector Adoption Score
To help executives and decision-makers assess the potential impact and value of a cloud resource optimization solution deployment to the business, this GigaOm Key Criteria report provides a structured assessment of the sector across five factors: benefit, maturity, urgency, impact, and effort. By scoring each factor based on how strongly it compels or deters adoption of a cloud resource optimization solution, we provide an overall Sector Adoption Score (Figure 1) of 4.4 out of 5, with 5 indicating the strongest possible recommendation to adopt. This indicates that a cloud resource optimization solution is a credible candidate for deployment and worthy of thoughtful consideration.

The factors contributing to the Sector Adoption Score for cloud resource optimization are explained in more detail in the Sector Brief section that follows.