The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
Amazon Web Services entered 2017 as the public-cloud infrastructure leader and finished it still accelerating. AWS reported roughly $17.5 billion in sales for the year, up about 43% from 2016, while its catalog, customer base and global infrastructure continued to expand. At the same time, Kubernetes became the cloud-native ecosystem’s center of gravity.
Those trends were not contradictory. AWS was gaining commercial scale, while Kubernetes gave developers a more portable orchestration layer and made vendor-specific container strategies less inevitable. AWS’s response—announcing a managed Kubernetes service, Amazon Elastic Container Service for Kubernetes (EKS), in preview at the end of November—showed how the company usually handles strategic threats: absorb them into the platform.
What “the cloud” meant in 2017
2017 cloud comparisons often mixed infrastructure as a service, platform services, software as a service, hybrid-cloud offerings and vendor revenue run rates. Those measures are not interchangeable. Contemporary estimates put AWS’s cloud-infrastructure run rate near $18 billion and the overall market’s growth near 38%, but those were estimates rather than a single audited market-share measure (contemporary market coverage).
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
The safest conclusion is narrower: AWS led public-cloud infrastructure by a substantial margin. Azure and Google Cloud were growing faster in percentage terms from smaller bases, but faster growth did not erase AWS’s absolute lead.
#1 Best Overall
Why AWS remained so difficult to catch
AWS’s advantage was cumulative. It had introduced core services such as EC2 and S3 in the mid-2000s, giving it years to build:
- a large installed customer base and operating history;
- a broad global infrastructure footprint;
- mature APIs, developer tools and documentation;
- partner, consulting and training ecosystems;
- economies of scale in computing, storage and networking; and
- a rapid cadence of incremental product launches.
By 2017 AWS was no longer merely renting virtual machines. Managed databases, analytics, security, Internet of Things services, machine learning, serverless computing and specialized hardware let customers place more of an application stack inside one commercial relationship. Lambda, for example, reduced server-management work while tying applications to AWS’s event model and APIs.
That breadth created value and lock-in simultaneously. Moving a basic virtual machine is one problem; replacing databases, identity, monitoring, networking, data pipelines and operational knowledge is another. The concern reported at the time was not that AWS was collapsing, but that its expanding platform made exit increasingly expensive (GeekWire’s 2017 analysis).
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →The numbers behind the “no signs of slowing” claim
Contemporaneous reporting put AWS’s 2017 sales at approximately $17.5 billion, with year-over-year growth of about 43% (CIO Dive). The phrase “no signs of slowing” should be read in that context: AWS was still expanding rapidly at enormous scale. It does not mean AWS had the highest percentage growth in every quarter or that competitors were standing still.
Azure and Google Cloud were important strategic challengers. Their growth rates benefited from smaller starting points, while AWS benefited from scale, service breadth and customer familiarity. For executives, the practical distinction was between absolute scale, where AWS remained ahead, and strategic pressure, which was increasing as customers sought negotiating leverage and alternatives.
Why Kubernetes became 2017’s defining infrastructure story
Kubernetes is an open-source system for scheduling and operating containers across a cluster. It handles placement, scaling, service discovery and recovery, and supplies a common API for microservices deployments. It can run on a public cloud, private infrastructure or bare metal.
Its significance in 2017 was ecosystem momentum. The Cloud Native Computing Foundation (CNCF) reported growth from 63 members and four projects at the start of the year to 170 members and 14 projects by year-end. Microsoft and AWS both became CNCF platinum members. KubeCon + CloudNativeCon North America drew 4,212 registrations, 106 sponsors and attendees from 51 countries (CNCF’s 2017 annual report).
Kubernetes 1.9 added a stable core workloads API and beta support for Windows Server containers. CNCF also added adjacent projects including containerd, CoreDNS, Envoy, gRPC and Jaeger. Training, certification and the Kubernetes Certified Service Provider program turned a project into a labor market, services market and operating discipline.
Rank #3
Calling Kubernetes “the standard” would overstate the evidence. Docker Swarm, Mesos/DC/OS, AWS ECS, OpenShift and other approaches remained active. A more accurate description is that Kubernetes had become the de facto leading orchestration project and the ecosystem’s center of gravity by late 2017.
How the major cloud providers responded
Google Cloud
Google had originated Kubernetes and already offered Google Kubernetes Engine. That gave it technical credibility and an early managed service, but technical leadership did not automatically translate into AWS-level commercial share.
Microsoft Azure
Microsoft made Kubernetes central to its open-source and multi-cloud repositioning. Kubernetes became generally available as an orchestrator option in Azure Container Service in February 2017, and Microsoft joined CNCF as a platinum member in July (Azure announcement; Microsoft’s CNCF announcement).
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsDo not retroactively describe every 2017 Azure Kubernetes announcement as mature AKS availability. The February announcement concerned Kubernetes support in Azure Container Service; product naming and service maturity evolved later.
AWS
AWS already had a container strategy. Amazon Elastic Container Service (ECS), launched in 2014, offered AWS-native orchestration, while AWS announced Fargate in 2017 to abstract server management for containers. Kubernetes’ momentum nevertheless made an AWS-native-only posture risky.
On November 29, 2017, AWS announced Amazon Elastic Container Service for Kubernetes (EKS) as a preview. AWS described a managed Kubernetes control plane with three masters across three Availability Zones, automated replacement of unhealthy masters, upgrades and patching, and integrations with IAM, VPC, Elastic Load Balancing, PrivateLink and CloudTrail (AWS announcement). EKS did not become generally available until June 2018 (AWS GA announcement).
Did Kubernetes threaten AWS?
Yes—but indirectly. Kubernetes reduced the risk of choosing one cloud for the orchestration layer, weakened ECS’s claim to be the default container control plane, and improved customers’ negotiating position. It also supported hybrid and multi-cloud deployment strategies.
But Kubernetes did not make a complete application cloud-neutral. A workload can use a portable Kubernetes API while depending on:
- Amazon Aurora, DynamoDB or another cloud database;
- provider-specific identity and permissions;
- cloud load balancers, networking and storage classes;
- monitoring, logging and security integrations;
- serverless APIs and proprietary messaging; and
- large datasets whose egress costs make migration uneconomic.
Stateless containers are generally easier to move than stateful systems. Persistent volumes, database replication, identity models and operational tooling vary by provider. Multi-cloud can improve resilience or bargaining power, but it can also duplicate staff, monitoring, security controls and incident procedures.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.EKS was a concession—and a retention strategy
AWS had three broad choices: continue emphasizing ECS, build a proprietary Kubernetes-compatible system, or offer upstream Kubernetes as a managed AWS service. EKS chose the third path.
That let AWS meet customer demand while continuing to sell the surrounding infrastructure: compute, storage, networking, identity, load balancing, logging and support. Kubernetes increased portability at the orchestration layer; EKS made that portability available without requiring customers to leave AWS. The result was not Kubernetes defeating AWS, but AWS monetizing Kubernetes adoption.
Choosing among the 2017 approaches
| Approach | Best fit | Main trade-off |
|---|---|---|
| Kubernetes | Hybrid or multi-cloud teams needing a common orchestration model | Operational complexity and incomplete application portability |
| AWS ECS | AWS-centric teams prioritizing simplicity and native integration | No Kubernetes API compatibility |
| Fargate | Teams wanting containers without managing servers | Less host control and potentially different economics |
| Google Kubernetes Engine | Organizations invested in Google Cloud and Kubernetes-native operations | Potential dependence on Google services |
| Azure’s Kubernetes offerings | Microsoft-heavy enterprises and Windows workloads | Azure identity and platform dependencies |
| OpenShift or Mesos/DC/OS | Enterprise governance or established alternative platforms | Smaller ecosystem momentum than Kubernetes by late 2017 |
The other bets shaping the cloud
Kubernetes was not the only important 2017 trend. Serverless computing promised less infrastructure management but increased dependence on provider APIs. Machine-learning services moved specialized infrastructure into managed platforms. IoT and edge computing raised demand for local processing where latency, bandwidth or resilience made a distant region unsuitable. Hybrid and multi-cloud strategies reflected the reality that many enterprises operated across data centers and several providers rather than making a single, clean migration.
What the year actually proved
2017 demonstrated that commercial dominance and technical influence could move in different directions. AWS retained a powerful lead through scale, service velocity and customer familiarity. Kubernetes became the most credible cross-cloud orchestration layer without eliminating provider lock-in. AWS’s EKS preview captured the relationship: customer demand made Kubernetes strategically necessary, but AWS could incorporate it into its own platform instead of surrendering the customer relationship.
For historical analysis, that is the durable lesson. Kubernetes changed the bargaining and architecture conversation; it did not, by itself, change where the money was being spent.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
