Amazon has won enterprise IT management
Last week, I wrote a short comment on LinkedIn praising Systems Manager, a service within the AWS portfolio fully dedicated to IT operations and governance of your AWS infrastructure.
If you want to read it, it’s here: https://www.linkedin.com/posts/alessandroperilli_aws-systems-manager-announces-patch-policies-activity-7021201395110289408-lwgC
The post received so much attention that I thought it would be important to further articulate my thoughts and, in fact, double down:
Not only AWS has turned Systems Manager into the most advanced cloud management platform (CMP) solution on the market, but it seems obvious to me that it has won the entire enterprise IT management space.
Before I go any further in justifying this statement, some context:
In the last 20 years, first with my own virtualization.info, then for Gartner, and finally in Red Hat, I have been busy observing who sells and who buys enterprise IT management solutions. In these roles, I have talked to more IT vendors (both startups and incumbents), reviewed more presentations, tested more products, evaluated more business models, and cried over more licensing agreements than any person should ever do in a lifetime.
When I started focusing on that space, there were four market leaders: BMC, CA, IBM, and HP. They were called “the big four of ITOM” and people thought that nobody could displace them. I’ve spent a lot of time with these guys.
Then, I’ve seen the wave of enterprise IT management startups that were founded to capitalize on the emerging server virtualization market. VMware itself, Akimbi, CiRBA, Dunes Technologies, DynamicOps, Embotics, ManageIQ, Veeam, VMLogix, VMTurbo, and dozens of others.
If you are old enough and you love walking down memory lanes, here’s a portion of all the companies I worked with: https://web.archive.org/web/20170705103215/http://www.virtualization.info/en/radar
Very few of those companies still stand today. Even less thrive. The rest has been acquired by VMware, went bankrupt, or was sold in a fire sale.
Then, there was the transformative advent of cloud computing. Some of the surviving startups I mentioned above tried to evolve, and some new ones (very few, actually) emerged to take their place. But something else interesting happened: two out of the top three cloud service providers (AWS and Microsoft) started offering basic operations and management control planes as part of their clouds.
In the last 9 years in Red Hat, as I worked on the business and product strategy for their business unit dedicated to enterprise management and automation solutions, I have been one of the keenest observers of AWS within the company. I’ve spent a significant amount of time reviewing every new service launch, every new feature, and every pricing and licensing change. Well beyond the boundaries of enterprise IT management and automation.
In the meanwhile, for 20 years, I listened to thousands of customers evaluating and buying from all these companies. Countless CIOs, VPs of Technologies or I&O, IT Directors, and Chief Architects in large end-user organizations, often in the Fortune 500 and Global Fortune 2000 groups, across every industry.
These people had (and still have) incredibly complex IT operations and governance problems to solve. They were (and still are) poorly heard, if at all, by the product management and engineers working across the vendors in the enterprise IT management space.
All of this is to say that my opinion on the AWS management offering is just an opinion, yes, but it’s an opinion that is informed by a broad and hopefully in-depth perspective about what companies need and what IT vendors usually offer, developed after two decades of working with every industry participant.
Some people think that those customers’ needs have changed as the industry has expanded from using just bare metal servers and virtual machines to also using containers and serverless computing, but it’s not the case. Managing the operations of a group of objects has fundamentally similar challenges once it reaches a certain scale. The implementation might change to handle different objects, but the problems don’t. And a different implementation doesn’t equal a novel solution.
What has changed is that new generations of product managers and engineers have lost the capability to listen to what customers are asking them. Something that their job roles should define as the highest priority.
The word “opinioned” has crept into the vocabulary of modern IT professionals and, today, it’s a pillar of most IT vendors’ strategies. So, customers ended up dealing with a class of entitled bros out of a fraternity that is too loud and busy telling others how the world works to listen with humility and learn about problems that they have no experience of.
This is not an old generation vs new generation argument. I believe there are very precise cultural and technological reasons why we ended up here, but this is material for another post.
Back to AWS and its enterprise management solutions.
As I mentioned in my original post on LinkedIn, I’ve seen Systems Manager evolve into the most complete and feature-rich cloud management platform on the market today. The service now provides capabilities that customers have asked every IT vendor for years and never got.
Many of these capabilities might be perceived as boring and “unsexy”, but they are vital for IT operations managers to govern a large-scale AWS environment.
It takes enormous determination to keep working on those capabilities when other PMs and engineers get to play with the cool new technologies.
More than that, it takes superhuman determination to keep working on enterprise management capabilities for an IT platform vendor.
Platform vendors are always deficient in operations, governance, and automation tooling until their customers start complaining loudly or a competitor starts stealing their business.
As I’ve seen in working with many technology providers, people that are employed by these types of vendors usually don’t have experience in managing large-scale IT environments. They specialize in provisioning large-scale IT environments and, perhaps, deprovisioning them, but what happens in between (what people call “Day 2 Operations”) is not their business. They don’t care, and it certainly doesn’t sound like a big deal to them.
IT platform professionals have one mission: maximize the number of logical units (VMs, containers, etc.) customers create and use on their platforms. The more workloads, the stickier the platform.
These people have no empathy for the challenges faced by those IT management professionals that have to operate and govern that ever-growing fleet of workloads.
I am not familiar with the culture and politics inside AWS, and they might well be an exception. But my experience tells me that working in a BU focused on enterprise management solutions within an IT platform vendor is a phenomenally hard and thankless job.
(you can tell, among other things, from the number of sessions dedicated to enterprise IT management in the catalogue of the flagship conference of any IT platform vendor)
If that’s the case in AWS, too, the Systems Manager team deserves double the credit for what they have built in the last few years.
Now. What that team has built is not just a powerful management tool. But an asset that can make a dramatic difference in the evaluation of a cloud service provider for large customers.
If you are an end-user organization and you have to manage tens of thousands of workloads, your first concern is (or should be) what you are going to do with those workloads once they have been provisioned by your users or your customers.
The answer to this question deeply influences (or should influence) your strategic decision about the cloud provider to adopt.
If the answer of the IT platform vendor (a cloud service provider or otherwise) is “I don’t know, it’s not my job”, then you know you are talking to people that have never been in your shoes. And if they didn’t fill their knowledge gap so far, it’s very unlikely that they will in the future.
If their answer is “immutable infrastructure”, you know that they simply don’t have the faintest clue of what they are talking about and they are blind to reality. Another post for another day.
If their answer is “we have a rich ecosystem of partners to help you with that”, you might want to reply “which one?”
This is one of the key points I want to highlight with this post: differently from 20 years ago, there is no rich ecosystem of enterprise IT management players that can effectively augment or replace the control plane of an IT platform vendor, especially the one of a cloud provider.
If your platform provider is doing a bad job at that, you have nowhere to go.
What you have is a sparse landscape of vendors that have repurposed their 10+ years old codebase so many times that their offering for X is now pure slideware.
The competitive pressure in enterprise IT management has long gone. And so it’s all the more wondrous that AWS has continued to deliver capabilities in such a market.
Most IT vendors are reactive: they introduce new features only when more aggressive competitors do that first and product managers get possessed by intense FOMO.
AWS owns its cloud platform. They have little competition from Microsoft and zero from Google in terms of control plane. And they have no competitive pressure from enterprise IT management startups. And yet, the company has kept building Systems Manager, quarter after quarter, delivering many of the capabilities that customers have always asked for and rarely got from other IT vendors.
And it’s not just about the capabilities. It’s also about the integrated and coherent experience that Systems Manager offers. It’s about the frictionless experience of deploying your workloads and having them ready to be managed in Systems Manager.
Going through the capabilities and experience of Systems Manager is beyond the purpose of this post, but the people who built it and keep building it are some of the unsung heroes of the company. If you have not followed the evolution of this service, you should pay attention.
Of course, System Manager is tightly integrated with the rest of the AWS infrastructure. However, imagine a theoretical scenario: if Amazon could flip a switch and offer the service as a stand-alone offering to manage any asset in any data centre and in any cloud, Systems Manager would be enormously more credible than any other enterprise IT management tool out there.
That is not to say that this is a strategy worth pursuing. It’s to say that Systems Manager is significantly better than anything else on the market today.
Is the solution perfect? No. Is there room for improvement? Significant.
On top of that, there are immense opportunities to revolutionize IT operations through artificial intelligence. Use cases and capabilities that I’ve not seen explored by any vendor, including AWS, yet.
Now notice that Systems Manager is just a portion of the assets that AWS has built for IT operations and governance. There are a million things to say about AWS’s efforts in many other areas, like software automation, event-driven orchestration, security and compliance management, IoT management, and an ocean of other subdisciplines.
Going through all of them is beyond the purpose of this post, but this is the bottom line:
The ones that were supposed to pay attention have done a poor job in the last decade in tracking and highlighting the immense investments that AWS has made in the enterprise IT management space.
If you are an enterprise organization, you should dedicate (your own) quality time to review the services that the company has developed. Understanding why these assets will make such a difference when your cloud investment will turn into a large-scale computing environment is key to long-term success.
If, instead, you are a startup: what are you doing??