Photo by Abigail Keenan on Unsplash
Growth in the various cloud platforms has become a dinner party conversation staple of those in the tech industry, in much the same way that house price appreciation was in the mid-2000’s. It’s interesting, everyone has an opinion about cloud computing growth statistics and it’s not entirely clear how it ends.
So let’s start with some industry projections. According to Gartner, the global infrastructure as a service (IaaS) market will grow by 39% in 2017 to reach $35 billion by the end of the year. IaaS growth shows no sign of slowing down and is expected to reach $72 billion by 2021, a CAGR of 30%. This revenue is principally split by the big-four players in public cloud: Amazon Web Services (AWS), Microsoft Azure (Azure), Google Compute Platform (GC) and IBM.
The approximate market share of these four public cloud platforms at the end of the first quarter of 2017 can be seen in the Canalys chart below. The reasons these numbers are only approximate is that each of these vendors include (or exclude) different facets of their cloud business and each seek to ensure their growth remains opaque to the investor community.
However, Amazon reported their earning in April 2017 and showed revenue growing 43 percent in the quarter to $3.66 billion, an annualized run rate of some $14.6 billion. Meanwhile Microsoft reported their cloud earnings in July 2017 and that its annualized revenue run rate was just under $19 billion. However, this includes a lot more than just IaaS and once non-IaaS is removed, analysts suggest that revenue is likely at the $6 billion run rate. Google cloud business is even harder to separate but its cloud revenue was estimated to be some $1 billion at the end of 2015, and although they seem to have hit their stride in the last year or so they clearly have a lot of ground to make. Current estimates are for approximately $2.5 billion in 2017. Lastly, IBM are estimated to be of a similar size to Google but appear to have a lot less momentum than the others, and certainly based on the requests we hear from our customer base IBM is not very often, if ever, referenced.
OK so other than guessing on the winners and losers why does this matter? In our humble opinion, it matters because this scenario creates increased competition and competition is good for consumers. It’s also relevant as companies have a choice, and many are looking at more than one cloud platform, even if they have not yet done anything about it. But what is really interesting, and what keeps us awake at night, is how much of this consumption is being wasted. We think and talk about this waste in terms of three buckets:
1) Always on means always paying – 44% of workloads are classified as non-production (i.e., test, development etc.) and don’t need to run 24×7
2) Over provisioning – 55% of all public cloud resources are not correctly sized for their workloads
3) Inventory Waste – 15% of spend on paying resources which are no longer used.
Combine these three buckets and by our reckoning you are looking at some estimated $6 billion in wasted cloud spend cloud in 2016, growing to $20 billion by 2020. Now that is something to really care about.
Few tools exist to actively monitor and manage this waste, and today there is not a cloud waste management industry per se, and currently tech analysts tend to lump everything under ‘cloud management’. We think that will change in the near future as cloud cost control becomes top-of-mind and the industry is able to leverage cloud computing growth statistics to calculate the scale of this industry problem.If you are in the cloud this is definitely something you should know about. Maybe you should consider optimizing your cloud spend now (before your CTO, CIO or CFO ask you to do so).
We sat down with Brian Park from Real Time Cases to talk to him about his company, how he uses AWS, and AWS free credits. We found out that the AWS startup package is a crucial part of making his business run.
Can you start by telling us about Real Time Cases and what you guys do?
So Real Time Cases is an education tech startup that is a new generation experiential learning platform. The new form of learning for today’s student is “learning by doing” and not just learning by reading antiquated textbooks. So Real Time Cases, through our partners, approach high level executives and say: “if you can hire 70-80 students to solve any problem in your department, then what would it be?” This forms the foundation for a “Real Time Case”. We film and document the issue, and professors can use that to drive concepts, theories and frameworks that they are trying to teach in the classroom, and use current, real life examples. Our cases are ongoing and happen “in real time” so they are like mini projects. This also opens the door for students to pitch some of these ideas to local business executives, which is exciting.
What is your role in the company?
I am the Director of Product. We have a platform that hosts the cases, videos are the primary content, due to the fact that most students would prefer to watch, rather than to read – think YouTube and Netflix. I am responsible for overseeing the technical team, both developers and designers. Amazon Web Services (AWS) is our cloud provider of choice and our entire infrastructure is hosted there
Why AWS over others?
We chose AWS because of the startup package, we get $10,000 of AWS free credits to use as we wish – compute, databases, and storage all for free! As with any startups, we have to bootstrap operations by keeping costs as low as possible and in addition AWS services are easy to use and access. If we had launched this company 10 years ago, we couldn’t operate at this cost point. So the credits and service offerings were very important to getting us successfully off the the ground and to market quickly and in a cost effective way. We have both domestic and international customers, and we can host and publish content for any university in the cloud at negligible cost which translates into affordable price points for students, and at our current cloud burn we can further sustain our operations for many months to come.
What technologies do you use in AWS?
We don’t have an official DevOps team, but we use Github for our code repository, Jira for agile processes, and Slack for communication . These low cost, SaaS tools plus AWS have been very productive for us. We are able to push code out in either 1 or 2 week cycles depending on the size of our stories. Our output used to be a 2 week sprint, but is now a 1 week sprint due to improved tools and processes. We follow agile development practices, participate in scrums and try to utilize the latest DevOps tools. As we have a distributed development and QA team. It’s best to use a tool like Jira to coordinate over time zones and accomplish the harder logistical tasks. We don’t have an overly complex architecture in AWS and use EC2, RDS and S3. S3 is used to store and host the video content we create for the professors and students.
Do you have any cost control measures in place for AWS?
Right now, no. When our AWS free credits expire we don’t expect our costs to be very high, but as a startup being able to leverage cost control tools like ParkMyCloud, to save 20-30% will be important – every dollar counts in a startup. We have been using AWS since our inception and haven’t had to move into the paid area yet – Bezos has created a truly disruptive business model that enables the startup community to rapidly prototype and test their thesis by quickly and inexpensively getting to market.
Perhaps you heard that Microsoft recently acquired Cloudyn in order to manage Microsoft Azure cloud resources, along with of course Amazon Web Services (AWS), Google Cloud Platform (GCP), and others. Why? Well the IT landscape is becoming more and more a multi-cloud landscape. Originally this multi-cloud (or hybrid cloud) approach was about private and public cloud, but as we recently wrote here the strategy as we talk to large enterprises is becoming more about leveraging multiple public clouds for a variety of reasons – risk management, vendor lock in, and workload optimization seem to be the three main reasons.
That said, according to TechCrunch and quotes from Microsoft executives the acquisition is meant to provide Microsoft a cloud billing and management solution that provides it with an advantage over competitors (particularly AWS and GCP) as companies continue to pursue, drum roll please … a multi-cloud strategy. Additional, benefits for Microsoft include visibility into usage patterns, adoption rates, and other cloud-related data points that they can leverage in the ‘great cloud war’ to come … GOT reference of course.
Why are we writing about this – a couple reasons. One of course is that this a relevant event in the cloud management platform (CMP) space, as this is really the first big cloud visibility and governance acquisition to date. The other acquisitions by Dell (Enstratius), Cisco (Cliqr), and CSC (ServiceMesh) for example were more orchestration and infrastructure platforms than reporting tools. Second, this points to the focus enterprises have on cost visibility, cost management and governance as they look to optimize their spend and usage as one does with any utility. And third, this proves that a ‘pushback’ from enterprises to more widely adopt Azure has been, “I am already using AWS, I don’t want to manage through yet another screen / console”, and that multi-cloud visibility and governance helps solve that problem.
Now, taking this one step farther: the visibility, recommendations, and reporting are all well and good, but what about the actions that must be taken off those reports, and integration into enterprise Devops processes for automation and continuous cost control? That’s where something like Cloudyn falls short, and where a platform like ParkMyCloud kicks in:
- Multi-cloud Visibility and Governance- check
- Single-Sign On (SSO) – check
- REST API for DevOps Automation – check
- Policy Engine for Automated Actions (parking) – check
- Real-time Usage and Savings data – check
- Manage Microsoft Azure (AWS + GCP) – check
The next step in cloud cost control is automation and action, not just visibility and reporting. Let technology automate these tasks for you instead of just telling you about it.
Today we’re happy to announce a new chatbot for AWS Slack integration that allows you to fully interact with ParkMyCloud without having to access the GUI. Combined with the recent addition of Notifications in ParkMyCloud, you can manage your continuous cost control from the Slack channels you live in every day!
Developers and operations engineers are increasingly utilizing ChatOps to manipulate their environments and help users self-manage the servers and databases they require for their work. There’s a few different chat systems and bot platforms available, but the most common used today is Slack. By setting up the SlackBot to interact with your ParkMyCloud account, you can allow users to assign schedules, temporarily override parked instances, or toggle instances to turn off or on as needed.
Combine this with notifications from ParkMyCloud, and you can have full visibility into your cost control initiatives right from your standard Slack chat channels. Notifications allow you to have ParkMyCloud post messages for things like schedule changes or instances that are being turned off automatically. Now, with the new ParkMyCloud Slackbot, you can reply back to those notifications to snooze the schedule, turn a system back on temporarily, or assign a new schedule.
The chatbot is open-source, so you can feel free to modify the bot as necessary to fit your environment or use cases. It’s written in Python using the slackclient library, but even if you’re not a Python expert, you’ll find it easy to modify to suit your needs. We’d love to have you send your ideas and modifications back to us for rapid improvement.
If you haven’t already signed up for ParkMyCloud, then start a free trial and get the Slackbot hooked up for easy AWS Slack integration. You’ll find that ParkMyCloud can make continuous cost control easy and help reduce your cloud spend, all while integrating with your favorite DevOps tools!
As of today, you can now connect to Okta for Single Sign-On (SSO) through the Okta App Network (OAN). This simplifies SSO configuration using SAML 2.0.
Using Okta for Single Sign-On allows administrators to easily add and govern their existing internal users in ParkMyCloud. It also reduces the number of passwords that users need to remember and use.
If you are an Okta customer, it is straightforward to connect to your ParkMyCloud account. First, run your account in admin mode and search for ParkMyCloud on the OAN. All you need from ParkMyCloud is an identifier string, provided in your account settings. Once configured, your users will automatically be added to ParkMyCloud, to the team you specify, after they have been authenticated through Okta.
This makes it extremely simple to get your enterprise users started with parking and saving in ParkMyCloud.
For more details about connecting Okta to your ParkMyCloud account, please see our knowledge base article on the subject.
We also recently added support for OneLogin for SSO, which joined Ping, Google Apps, and Azure Active Directory as SSO options for your ParkMyCloud account.
As of today, you can now connect ParkMyCloud to OneLogin for Single Sign-On. ParkMyCloud is now integrated with OneLogin’s App Catalog marketplace to simplify Single Sign-On configuration using SAML 2.0.
Using OneLogin for Single Sign-On (SSO) simplifies processes for users by reducing the number of passwords they need to track. It also simplifies the burden for administrators, by allowing them to control user access in one place (for example, through OneLogin) so they don’t have to manage access separately for each user for all applications.
Through OneLogin, ParkMyCloud supports just-in-time provisioning of new users, which means that as soon as a user is authenticated in OneLogin, he or she is automatically created in ParkMyCloud. All the administrator needs to do is email users the organization’s unique ParkMyCloud login link, which can be found in the ParkMyCloud management console.
Once you have a ParkMyCloud account that you would like to integrate with OneLogin, go to the App Catalog and search for “ParkMyCloud”. Once on the ParkMyCloud page, configure the SSO options and accessibility to users, then connect the metadata to ParkMyCloud. That’s it! Now you can invite your OneLogin users to ParkMyCloud using your unique link, and they will be able to log in to ParkMyCloud in one step with their OneLogin info.
For more information about configuring OneLogin SSO for your ParkMyCloud account please refer to this article.
By the way, ParkMyCloud also supports SSO via Ping, Okta, Google Apps, and Azure Active Directory – see more about our support for these providers.