Freedom from vendor lock-in is as urgent as ever for enterprise clouds. This is a crucial insight to glean from Gartner’s Magic Quadrant for Cloud Infrastructure and Platform Services, published in July. In the report, analysts Raj Bala and Bob Gill named questionable, often overly aggressive sales tactics among the top three problems they found with market leaders Amazon Web Services, Google, and Microsoft. Such sales behavior points to the need for enterprises to adopt an effective multi-cloud strategy to help lessen their vulnerability to cloud vendors and, ultimately, ensure freedom from vendor lock-in.
Negative Experiences with Sales
Imagine that you’re an AWS customer, and it’s time to renew your subscription. You’re not ready to expand your cloud; you just want to renew the previous contract. However, when you discuss the renewal with your AWS sales rep, they keep pushing you to spend significantly more, hinting that otherwise the renewal may be at risk. You feel annoyed and offended by this threat, but having already deployed critical workloads on AWS and trained your staff in AWS deployment tooling, you feel your hand is forced into agreeing to spend more.
According to the Gartner report, this is what’s happening at AWS. The report highlights “challenging renewals” as the first caution to consider when evaluating the public cloud giant. “Dozens of Gartner clients across multiple geographies have reported unexpected pressure from AWS sales, which has sharply accelerated over the past year, to increase annual spend commitments by 20% to renew existing contracts,” the authors wrote. Additionally, “Since these customers typically have significant dependence on the platform, they may feel as if they have limited recourse” but to cave in to the pressure.
To be fair, the report states that “the pressure to increase spend is not AWS’s policy and will be eliminated if the customer escalates,” but with dozens of Gartner clients reporting such pressure during renewals, this should give pause to any enterprise considering AWS as a strategic cloud partner.
What’s more, AWS is not the only public cloud leader with problematic sales tactics. According to the report, “Microsoft has very complex licensing and contracting, and a complex account management structure with uneven cloud skills in the field. Further, Microsoft sales pressures to grow overall account revenue prevent it from effectively deploying Azure to bring down a customer’s total Microsoft costs.” So if you’re implementing Azure in hopes of cutting your Windows Server costs, lo and behold — you may also end up increasing your total Microsoft bill.
As far as Google is concerned, Gartner didn’t find unwarranted sales pressure but instead criticized the company for providing “limited incentives.” Currently, Google is undercutting rivals with “aggressive pricing,” but with GCP operating at a financial loss, analysts expect that “discounting is likely to eventually taper off as the company continues to mature with respect to revenue and customers.”
So how can you avoid getting the short end of the stick from your cloud vendors?
Gaining Leverage through a Multi-Cloud Strategy
Multi-cloud architecture is about distributing workloads across two or more cloud infrastructure platforms. In multi-cloud deployments, workloads are portable, so development teams can move application components from one cloud platform to another as needed.
By deploying your workloads on multiple cloud providers, whether you choose multiple public clouds or a mix of public and private clouds, you acquire leverage to negotiate the best pricing and service levels. Don’t get stuck in a situation where you feel you have “limited recourse” and are unable to push back on vendors trying to take advantage of your reliance on their services.
Even if questionable sales tactics were not a problem, employing a multi-cloud strategy gives you the freedom to take advantage of the best pricing, capabilities and geographic availability of each provider. Cloud provider pricing can be hard to predict and change with little notice, so if you get surprised by a bill that is higher than expected, being able to instantly move between multiple providers helps to mitigate risk with a Plan B. Additionally, being able to access multiple clouds lets you continually optimize price/performance in terms of workload hosting, network egress costs, latency and other factors.
Using multiple providers also mitigates risk when a vendor decides to change their offerings. For example, AWS just announced the discontinuation of EC2-Classic, giving any customers still using the service just one year to migrate their workloads.
Plus, every cloud provider has its strengths and weaknesses in terms of enterprise features, level of innovation, availability zones and many other factors, and going multi-cloud enables you to choose the best cloud provider for specific application requirements. For example, if you have workloads for customers in China or Southeast Asia, your best bet may be to go with Alibaba or Tencent for those applications due to their leadership in the region, while still keeping other workloads on American or European providers, or even on your own private cloud.
As reported recently in the Wall Street Journal, whereas IT managers previously deemed Amazon and Microsoft as the only cloud computing options, now businesses and governments ranging from Experian and AT&T to the Central Intelligence Agency are “signing up for a mix of providers, cherry-picking features and playing the vendors against each other to keep costs down.”
That said, cloud technologies such as Kubernetes and OpenStack are complicated enough on their own. How can you harness all the benefits of multi-cloud without getting mired in the additional complexity?
A Standard API across Multiple Clouds
In order to successfully implement a multi-cloud strategy, it is essential to employ a standard API across all your cloud providers. Otherwise you have the burden of maintaining multiple provisioning stacks unique to each infrastructure or provider. Additionally, having a common management API enables you to monitor usage, behavior, and performance across all your clouds with a single pane of glass.
For example, products such as Mirantis Container Cloud provide one set of APIs to deploy, manage and observe Kubernetes clusters on multiple infrastructures, including public cloud, private cloud, or bare metal. Some also provide a user-friendly web interface for developers, enabling them to create new clusters in minutes on a number of different providers.
Mirantis Container Cloud self-service web UI
Ultimately, the thing to remember is that a multi-cloud strategy is about having control over the cloud infrastructure your business depends on. You don’t want to put all your eggs in one basket, and you don’t want cloud vendors to be able to take advantage of you. With a multi-cloud architecture and a common management API, you can make multi-cloud a successful strategy for your organization.
If you would like to experience the flexibility of a standard API for multi-cloud, you can test drive Mirantis Container Cloud with a free trial.