Who are the best cloud platform providers for your multi-cloud?
I attended the Gartner Catalyst Conference 2018 in London recently. Every year the conference is a good place to network with technical professionals on the frontlines of digital transformation and learn about key initiatives happening within enterprise organizations. Being interested in strategy around cloud infrastructure and multi-cloud, I joined a session presented by Elias Khnaser, Research Director at Gartner, that compared the major cloud platform providers. In this blog post I’ll share some of what I learned.
Who is the best cloud provider?
I was interested to hear that Gartner’s customers constantly ask them who the best cloud provider is, as KEMP is often asked the same question.
It turns out there’s no simple answer, of course. KEMP’s main customers — enterprises— are looking for different things in cloud infrastructure as a service (IaaS), and Gartner emphasizes that. The analyst firm has put together an in-depth assessment that encompasses the realities you face when you’re evaluating cloud providers.
For example, almost every enterprise wants compute, storage and networking in their mix of cloud services, so you would naturally evaluate those as required services. However, almost every provider offers value-added services that aren’t necessarily required but that will make your life easier, like securing access and managing operations. Think of those as preferred services in your cloud evaluation scheme. Then there are optional services that differentiate vendors even more finely, like compliance and documentation. If you’re still new to cloud, you may not be ready to think about those yet, but as they grow in importance and as cloud computing matures in your organization, you’ll begin to place more value on them.
It won’t surprise you that Gartner’s in-depth assessment covers the best-known public cloud providers: Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform and Oracle Cloud Infrastructure, with IBM Cloud, Alibaba Cloud and others in the future. In the session I attended, Gartner measured cloud offerings strictly on whether the cloud provider includes a given technical service or not. Gartner’s Magic Quadrant, on the other hand, is a superset of technical and non-technical characteristics of both the provider and the offering, such as sales strategy, execution, market share and business model.
Models for choosing a cloud provider
Gartner offers two Excel spreadsheet models for evaluating cloud vendors: a blank model to send out with requests for proposals, and a pre-populated, in-depth assessment Gartner has made for each provider. The models cover more than 200 criteria and sub-criteria that go into your decision about cloud vendors. Criteria are constantly changing too, as the market changes and lines begin to blur between, say, IaaS and some of the services offered as part of platform as a service (PaaS).
Applying these criteria you’ll see how each provider’s relative strengths and weaknesses surface. As a technical professional, you may choose AWS’s strong, enduring market share, the stability of its services at scale and the third-party vendors in its marketplace. But you’ll have to put up with horizontal-only autoscaling, lack of backup and no VM console access. What’s your priority?
Or, you may find it more important that Microsoft has been gradually building its software products to run both on premises and in the Azure cloud, and that its vast installed base gives it the inside track on identity and access management (IAM). But are those features more important to you than Azure’s limited support for availability zones per region, slower spin-up of VMs and the lack of metrics-driven load balancing?
Some technical professionals like Google Cloud Platform for its strong reputation for security, its optimized network and its financial discounts for sustained usage. But at least for the time being, they’ll have to put up with manual-only autoscaling, a lack of encryption for LAN traffic and lack of support for enterprise-grade databases.
And, if you’re an Oracle shop, then you may see Oracle Cloud Infrastructure as a logical next step on the road to IaaS. But it’s a relatively new offering, meaning that it doesn’t yet deliver features like autoscaling, object versioning, backup and global load balancing.
And the winner (for the moment) is . . .
Bottom line, the scores Gartner gave in this Cloud Platform comparison session were: AWS, 81%; Azure, 79%; Google Cloud Platform, 69%; Oracle Cloud Infrastructure, 48% (higher scores are better.) Gartner bases those overall scores on a mixture of how vendors can support required, preferred and optional services. Your own evaluation could yield different scores.
Multi-cloud means multiple eggs in multiple baskets
For me, the biggest conclusion I drew from the session is that it makes sense that an increasing number of enterprises are developing a strong multi-cloud strategy that doesn’t adopt one provider to the exclusion of the others.
People start out asking, “Who is the best cloud provider?” But as they spend more time in cloud computing, they switch to asking, “Which providers are best suited to the particular applications and workloads I need to run in the cloud at the moment?”