How to Make Your Organization Successful in Public Cloud Tagging

By Marco Meinardi in research

This post originally appeared on the Gartner Blog Network.

If you use public cloud services at some scale, then you know you must tag your resources. You know you must use tagging so that your cost reports get nicely broken down; so that you can create resource groups to enable inventory, governance, automation and access control. You know this but, still, you are not doing it. Or maybe you think you are, but only less than half of your cloud resources are tagged in the way you want.

Unfortunately, this situation is something I hear in my inquiry calls more often than not. Why? Is it because of lack of enforcement measures? Is it because your cloud consumers are not disciplined? Or because tags are perceived just as administrative burden? The answer is all of the above. Tagging is a complex matter and some people assume that distributing a list of “mandatory” tags is enough. But that’s only the very beginning of your tagging strategy.

Common guidance on tagging usually stops at providing a list of suggested tags and describing the technical mechanisms for their implementation. While this is certainly important, such guidance often does not consider the organizational impact of metadata management in highly dynamic environments. Internal resistance within an organization is often the primary cause of the failure of tagging initiatives. That’s why I decided to publish a guidance framework for “Implementing a Tagging Strategy for Cloud IaaS and PaaS” (paywall). The framework is now available to Gartner for Technical Professionals subscribers and it is depicted in the figure below.

Defining your tagging dictionary is really just the beginning of your process, while enforcing tags is the very last step. The activities in between serve to internally promote the value of tagging and to lower the overhead for their implementation. Promoting tags requires the communication of their benefits and use cases. Implementing tags can be made easier through automation.

We developed this framework to make sure you’re successful in your tagging initiative. Cloud providers offer lots of native tools and constructs to manage your tags. However, tagging is not just a technical problem. By following the Gartner framework, you will also manage the organizational impact of tags, mitigate internal resistance and ease their implementation.

The Gartner framework is available today behind paywall with a 30+ page research note that explains its application in details. To know more about this research, you can also schedule an inquiry call with me (inquiry@gartner.com) or talk to your Gartner representative.

Lastly, feel free to follow me on Twitter (@meinardi) or connect with me on LinkedIn for further updates on my research. Looking forward to talking to you!

A Comparison of Public Cloud Cost Optimization Tools is Now Available

By Marco Meinardi in research

This post originally appeared on the Gartner Blog Network.

If you’re using public cloud infrastructure and platform services, I bet that you’ve been thinking about adopting a tool to cut down costs. You’ve been told that there is some inherent waste in your cloud spending and you want to address that. I’m also sure that a number of vendors told you that they can help with that. But to what extent? If cost management and optimization are becoming “table stakes” in the cloud management market, there can be a huge difference in the capabilities of each available solution. Some solutions may just scratch the surface and simply report for underutilized instances – when CPU have been low in usage in the last week or so. But these solutions will leave it up to you to figure out the rest. Other solutions may automatically execute precise instance rightsizing across types, families and regions, using AI-based pattern recognition and ML inference.

Both cloud providers and third-party vendors have invested in developing cost optimization capabilities for public cloud services. In this complex scenario you may wonder: which tools will allow me to truly maximize savings while minimizing performance risks? The good news is that Gartner just published research to answer this exact question and it’s available on gartner.com right now.

My colleague Brian Adler and I have just published the following two research notes, both available behind paywall:

The notes provide a side-by-side comparison of each solution based on a common set of criteria. Examples of criteria include compute instance rightsizing, block storage rightsizing, unused resource decommissioning and reservation portfolio management. For each criterion, vendors have been scored with grades such as “Low”, “Medium” or “High”.

Gartner clients can use the two research notes to understand what you can do using cloud providers’ native tools and which gaps you can fill with third-party tools. Furthermore, clients can use the provided criteria to assess the capabilities of any other public cloud cost optimization tool that hasn’t been included in this research.

This research is part of a series of Solution Comparisons that we published to assess tools in various areas of the Gartner cloud management wheel. Read the full research notes if you want to know the results of this comparative assessment. You can also schedule an inquiry call (inquiry@gartner.com) with myself or my colleague Brian Adler if you want to have private conversations about our research findings. In case you don’t have access to this research and you’d like to, I’m sure your Gartner representative will be more than happy to help.

Lastly, feel free to follow me on Twitter (@meinardi) or connect with me on LinkedIn for further updates on my research. Looking forward to talking to you!

AWS Just Made Their Management Tools Ready for Multicloud

By Marco Meinardi in events

This post originally appeared on the Gartner Blog Network.

I am just back home after spending last week at AWS re:Invent in tiresome, noisy, vibrant and excessive Las Vegas. At Gartner, I cover cloud management and governance and I was disappointed not to hear much about it in any of the keynotes. I get it, management can be sometimes perceived as a boring necessity. However, it is also opportunity to make a cloud platform simpler. And that’s something that AWS needs. Badly.

Despite the absence of highlights in the keynotes, I spotted something interesting while digging through the myriad of November announcements. What apparently got lost in the re:Invent noise is that AWS is opening up some of their key management tools to support resources outside of the AWS cloud. Specifically, AWS CloudFormation and AWS Config now support third-party resources. And that’s a big deal.

The Lost Announcements

The CloudFormation announcement reports that AWS has changed the tool’s architecture to implement resource providers, much in line with what Hashicorp Terraform is also doing. Each resource provider is an independent piece of code that enables support in CloudFormation for a specific resource type and API. A resource provider can be developed independently from CloudFormation itself and by nonAWS developers.

AWS plans to promote resource providers through the open source model and has certainly the ability to grow a healthy community around them. The announcement also says that a number of resource providers will be shortly available for third-party solutions. Upcoming solutions include Atlassian, Datadog, Densify, Dynatrace, Fortinet, New Relic and Spotinst. AWS is implementing this capability also for native AWS resources such as EC2 instances or S3 buckets, hinting that this capability may not be just an exception, but a major architectural change.

In the same way, AWS Config now also supports third-party resources. The same resource providers used by CloudFormation enable AWS Config to manage inventory, but also define rules to check for compliance and create conformation packs (a.k.a. collections of rules). All of this also for nonAWS resources.

Why is This a Big Deal?

With this launch, AWS addresses one of the major shortcomings of its management tools: being limited to a single platform – the AWS cloud. From today, anyone could develop resource providers for Microsoft Azure or Google Cloud Platform resources. This possibility makes AWS CloudFormation and AWS Config de facto ready to become multicloud management tools. And we all know what AWS thinks about multicloud, don’t we?

Furthermore, AWS is now challenging the third-party management market, at least within the provisioning and orchestration, inventory and classification and governance domains (see this Gartner framework for reference). AWS CloudFormation now incorporates more capabilities of HashiCorp Terraform. It also can be used to model and execute complex orchestration workflows that organizations normally handle with platforms like ServiceNow. AWS Config can now aim to become a universal CMDB that can keep track of resource inventory and configuration history from anywhere.

Both AWS CloudFormation and AWS Config are widely-adopted tools. Customers could be incented to extend their use beyond AWS instead of selecting a new third-party tool that would require a new contract to sign and new vendor to manage. Does this mean that AWS has issued a death sentence to the third-party management market that makes much of its ecosystem? Certainly not. But these announcements speak to the greater ambition of AWS and will force third-party vendors to find new ways to continue to add value in the long term. Maybe the resource provider ecosystem will not develop, and customers will continue to prefer independent management vendors. Or maybe not.

In conclusion, it was disappointing not to hear this message loud and clear at re:Invent this year, especially compared to the amount of noise we heard around the launches of Google Anthos and Azure Arc. But there is certainly a trend for which all the major providers are preparing their management tools to stretch out of their respective domains. How far they want to go is yet to be determined.

What Blockchain and Cloud Computing Have in Common

By Marco Meinardi in research

This post originally appeared on the Gartner Blog Network.

Blockchain technologies provide ledger databases whose records are immutable and cryptographically-signed using a distributed consensus or validation protocol. These characteristics contributed to the popularity of blockchain to power transaction execution in multiparty business environments. With blockchain, multiple parties can agree on transaction details while still guaranteeing correctness and prevent tampering, without having to rely on a trusted centralized authority.

To provide such functionality and just like any other database, blockchain technologies are built around platforms, infrastructure, APIs and management tools. Cloud computing is a well-oiled model that provides easy access to all these technology components, in addition to services and capabilities for application development and integration. While cloud computing can certainly help accelerate the execution of blockchain projects, it is also a heavily centralized model, specifically around few hyperscale megavendors. Conversely, the effectiveness of blockchain relies on decentralization as one of its core principles .

Full decentralization is especially important for public blockchains (such as Bitcoin) where anybody is free to participate and transact. Conversely, enterprise blockchains may accept to trade aspects of decentralization (such as a single technology provider) in exchange of easier access to technologies and a lower management overhead.

All hyperscale cloud providers have launched blockchain cloud services in the last 18 months to help organization with their blockchain projects. These services build on the strength of each provider (in terms of infrastructure, platform and application development capabilities) but also aim to facilitate the use of open-source DLT frameworks such as Ethereum, Hyperledger Fabric and Quorum.

On my recently published research “Solution Comparison for Blockchain Cloud Services From Leading Public Cloud Providers” (paywall), I have assessed and compared the blockchain-related cloud services offered by:

  • Alibaba Cloud
  • Amazon Web Services
  • Google
  • IBM
  • Microsoft
  • Oracle

The research provides a heatmap of the capabilities provided by each vendor, allowing Gartner clients to quickly assess their strengths and weaknesses in this space. The research also provides all the details behind the attributed scores for those technical professionals who want to dig deeper into each vendor’s offering. Some example of the comparison criteria include:

  • Number of Supported DLTs
  • Blockchain Community Involvement
  • Infrastructure Supported
  • Fully Managed Ledger Service
  • Smart Contract Management

Like most blockchain technologies, also blockchain cloud services are still immature, especially in light of the rapidly evolving landscape of DLT frameworks. As a demonstration of that, many of the assessed cloud services have been launched during the conduction of this research, which required multiple re-assessements of the vendor offerings. Some vendors also launched additional services and features after the publication of this research, for example:

To know more about this topic or if you would like to discuss further, you can read the research note at “Solution Comparison for Blockchain Cloud Services From Leading Public Cloud Providers” (paywall). You can also reach out to your Gartner representative to schedule an inquiry call with me. Looking forward to hearing your comments!