As we look back on the past 10 years, it’s pretty remarkable how much has changed in the world of cloud computing. On January 1st, 2010, AWS was a mere four years old with an estimated revenue between $500 million and $700 million, representing less than 3% of Amazon’s annual revenue (source). Microsoft Azure, Google Cloud Platform, IBM Cloud, Oracle Cloud, and Alicloud were not yet in the market, nor were containers or serverless technologies as we know them today. The National Institute for Standards in Technology (NIST) hadn’t even written a definition yet for cloud computing. In April of 2010, the New York Times would publish an article marveling that a “DVD-by-mail company” would decide to move most of its “web technology” to Amazon.com.
Let’s take a quick dive into the archives and explore the top news of the past 10 years and remember how exactly we got here.
2010: Microsoft launches Azure to go head to head with Amazon
In 2008, Microsoft announced they were working on Windows Azure, “a cloud services operating system that serves as the development, service hosting, and service management environment.” However, it wasn’t until February 2010 that the service became generally available. At launch, Azure cost a whopping 12 cents per hour for compute (only one VM option) and 15 cents per GB per month for storage. Azure was in many ways ahead of its time, offering more than just utility compute and storage, focusing services more on PaaS and developer services. Since then, Azure’s market share in the cloud computing space has grown steadily, reaching just shy of 20% of the market by 2020.
2011: AWS failure takes down half the internet
While there have been many cloud outages over the past decade (although thankfully they’re becoming fewer and further between), there was one major and widespread outage that stands out from the rest. On Thursday, April 21, a large number of EBS volumes went down in US East bringing down some notable web presences at the time, including Reddit, Quora, and FourSquare. It wasn’t until late Sunday night that service was fully restored. This outage served as a wakeup call for many companies that still relied on a single AWS Region or even a single Availability Zone.
2012: The cloud price wars peak
For years, one of the main ways public cloud providers competed was on price. Eventually, as prices dropped and baseline services commoditized, the big three found other ways to compete. While there were many years of price wars, 2012 felt like the true peak when AWS, Azure, and Google were firing shots back and forth and continuously dropping prices. It was also the year we saw major price reductions off of a broad array of core services (compute, storage, relational databases). Price cuts would of course continue, but the outright warfare between the providers started to fade after 2012.
2013: Google and IBM enter the cloud computing market
In 2012, Google announced the limited preview of Google Compute Engine, which then became generally available in December 2013 (an uncharacteristically short beta period, by Google’s standards). Similarly to Amazon, Google’s cloud platform was less of a net new offering and more of a commercialization of the back end that powers Google’s commercial offerings in the market. On the opposite end of the spectrum, 2013 was also the year that IBM entered the cloud computing market with their acquisition of Softlayer. IBM used Softlayer as the basis for their cloud offering, Blue Mix, which was launched the next year (rebranded as just IBM Cloud in 2017).
2014: AWS announces Lambda, launching serverless computing
Sometimes the biggest news of the year is only truly apparent in hindsight, and for 2014, when AWS launched Lambda, this was definitely the case. Announced at re:Invent 2014, it was called an “event-driven computing service for dynamic applications.” At the time, it hardly made a massive impression—I could only find three news articles written about the announcement at the time! However, this was the beginning of serverless becoming a major movement in cloud computing, and today Forrester reports that one in four global developers report that their firm uses serverless architecture, with 24% more planning to do so in the next 12 months (Forrester, Serverless Development Best Practices, October 2, 2019, By Jeffrey S. Hammond, John R. Rymer.)
2015: Oracle throws in their hat as HP throws in the towel
After years of bashing cloud computing Oracle got into the cloud computing game in 2015. This came just a few months after HP execs told the New York Times that they couldn’t compete with the big guys (they formally announced they would sunset their HP Helion Public Cloud later in 2015). Initially, it appeared that Oracle was seeing good traction with its offering—in 2015 they claimed to have earned $1.5B in cloud revenue—until it emerged that they may have been cooking the books. The scandal involved a lawsuit from a former Oracle CPA who claims she was fired after refusing to add millions of dollars in accruals to cloud financial reports.
2016: The tipping point for enterprise adoption of cloud
It wasn’t apparent at the time, but in hindsight, 2016 was a watershed year for large enterprise adoption of public cloud. It started with Capital One announcing they were going all-in on cloud, stating that they needed to start acting and thinking more like a tech company than a bank. Since then, Capital One has become a poster child for leading-edge adoption of cloud technologies. Along similar lines, Salesforce.com, another notable enterprise, announced they were going to the public cloud. As one of the largest SaaS companies in the world that grew up in the days before cloud computing, they became a role model for other SaaS organizations that were still operating their own data centers.
2017: Kubernetes wins the container orchestrator war
For several years there was a small war waging in the container orchestration world: who would come out on top? Mesos? Kubernetes? (Docker Swarm, Openstack Magnum, and VMware Photon were also in the mix, but weren’t serious contenders). It’s hard to say exactly when the tides shifted in Kubernetes’ favor, but by the end of 2017, the container wars had a clear winner. By the end of 2017, native Kubernetes was supported on all the major public clouds: Google Kubernetes Engine (GKE), Amazon Container Service for Kubernetes (EKS), and Microsoft Azure Kubernetes Service (AKS).
2018: Leaky S3 buckets force a cloud security reckoning
An unfortunate trend started to take shape in 2017 and came to a head in 2018: leaky S3 buckets. In other words, S3 buckets with improper permissions that resulted in the exposure of critical or sensitive data. The worst exposure was actually in late 2017 when an open AWS S3 bucket exposed sensitive Experian and census info on 123 million U.S. households. For several months, barely a week went by without the news of another security breach. Eventually, the storm calmed down, helped somewhat by AWS releasing the capability to block public access to all objects in an account with a few clicks. This is more of a brute force approach, but it can be effective. In 2019, AWS released the ability to do a more in-depth analysis of S3 bucket access. I should note that this problem is not unique to AWS, a user on any public cloud could easily leave an object storage bucket open and accessible to the public, but as the market leader, most of the stories featured Amazon S3 buckets.
See our article here on how to prevent AWS S3 bucket misconfigurations.
2019: Microsoft wins the JEDI contract
After nearly a year and a half, two lawsuits, and a lot of drama, the pentagon announced that they were awarding the JEDI contract to Microsoft. For some (myself included), this was a major surprise as the contract seemed destined from the start to be for AWS. But given the political climate and Trump’s dislike of Jeff Bezos, others were not at all shocked to see the contract go to Microsoft. I believe we will look back at this moment and realize it was a turning point in Microsoft’s pursuit of AWS as the top public cloud provider.
2020: COVID-19 accelerates cloud adoption
The COVID-19 outbreak impacted (and is continuing to impact) the lives of people around the world, and forcing organizations to make significant changes to business operations. To accommodate remote workforces and enable scalable infrastructure for digital products and services, businesses are looking to the power of public cloud infrastructure and container technology.
In fact, the latest analyst research indicates that 70% of organizations using cloud services plan on increasing their cloud spending in the wake of COVID-19—a trend that isn’t expected to slow down any time soon. With more organizations investing in cloud technology, we’re also seeing a greater need for cloud management and optimization solutions to help maximize their investment.
To better understand the impact of COVID-19, we conducted our own analysis of nearly 500 organizations’ actual public cloud spend from January through September 2020. Based on these insights, we can understand how different businesses used the public cloud in response to the realities of the COVID-19 pandemic. You can download a copy of the report for yourself here.
The past ten years have been quite a ride and I’ve been privileged to have a front-row seat for it. There were so many watershed events I didn’t even touch on, you’ll notice I largely left out acquisitions (notably RedHat by IBM and GitHub by Microsoft), and personnel changes (notable changes at Google with Diane Greene and Thomas Kurian). I can’t wait to see what the next 10 years will bring!