Well, this blog post has been 3 months in the works – in fact a draft has been sitting in WordPress since September when I collaborated with Kim Bottu on his article about Integrating CloudHealth and vROps, the plan was for me to write a companion blog which show-cased the same capabilities in CloudHealth that he mentioned in his vROps blog.
2020 has been a profoundly difficult and odd year for everyone, and I found myself not wanting to do anything after a busy day of zoom meetings and home schooling. The motivation to write a blog just wasn’t there, and after a busy day all I just wanted to do was chill and relax in the evenings.
There’s been a fine-line in everyone’s work-life balance this year, and everyone needs to find that little bit of time each day to just shut off and unwind (usually when the kids are in bed)!
Anyways, the Christmas holidays and having time off work has given me the opportunity to sit back down and finish the blog (plus Kim was saying I should publish it in order to help my vExpert application for 2021… hahahahha… lol…. – btw, you have till the 9th January 2021 to submit!)
What is CloudHealth?
I guess the best place to start this blog is to give a quick overview of what CloudHealth actually is, so here’s the elevator pitch I always give….
“The more organisations invest in public cloud, the more important it is to have a cloud management strategy for their success, and this is where CloudHealth can assist.
CloudHealth is a multi-cloud management platform designed to provide full visibility into your cloud environment – helping you to identify opportunities for cost savings and usage optimisation. We help you to easily analyse and control cloud costs, security, performance and governance all from one single platform.
We give you insight into your data centre, hybrid and public cloud spend – aligning costs and usage to users, lines of business or even projects and business initiatives.We help make cloud management simple.”
Sooooo, what does that actually mean I hear you ask!?!
In a nutshell, CloudHealth takes your cloud billing and usage data, process and presents it in reports that help you visualise your costs and usage. In addition, one of their USPs is the ability to create perspectives to help you categorise and filter your data.
Currently CloudHealth supports Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform (GCP), Oracle Cloud Infrastructure (OCI) and on-premise VMware environments. They also have a beta-program for VMware Cloud on AWS support.
CloudHealth is the clear leader in multi-cloud management, they’re the largest player in the market with 10,000+ customers and 230+ partners globally, managing over $11+bn in annual cloud spend.
CloudHealth has continued to be named a Leader in The Forrester Wave: Cloud Cost Management and Optimization Report.
What is a Perspective and How are They Used in Reports?
The most common way to describe a CloudHealth Perspective is that they are “lenses” through which you want to view your infrastructure. Each role within an organisation measures and evaluates the business from different viewpoints or ‘Perspectives’.
You can create Perspectives to view and group cloud assets together in order to align them with business objectives.
They provide a framework for categorising all the assets within your cloud infrastructure. For example, you could create a Perspective to group assets into Environment, Application, Department, Function, Project, or even Cost Centre.
You can build Perspectives dynamically using cloud tags or statically using the search capabilities.
For Example, the default view of a Cost History Report within CloudHealth is to show 13 months of cost data categorised by Service type (this is an example of an AWS report):
We can then take that default view and change the categorisation to a Perspective built to show Owners (this could help identify those users who spend all the company’s money on cloud!):
Or we can even change the view to categorise by a Perspective built to show Environment (IT Operation Managers are constantly looking for ways to show how much different Infrastructure Environments cost the business):
Finally, we can combine a number of Perspectives together to drill down further into our costs. In this example we’re filtering to look at just the Production Environment Perspective group, and categorising by the Owner Perspective (so helping to identify who spends the most in Production!):
Chart Types for a report can also be changed from Bar to Line – in this example we’re looking at the Cost History Report categorised by the Perspective ‘Line of Business’:
Another great Chart Type to use is the Pie Chart – as this is only 2 dimensional you will need to filter to a specific time period (eg. November 2020) and change the X-axis away from time interval (in this example I’ve used the ‘Line of Business’ Perspective):
Using CloudHealth to Generate Alerts.
Now the basics of Reporting and Perspectives are out of the way…. Let’s take a look at replicating within CloudHealth what Kim configured in vROps.
In Kim’s blog, he looked at how vROps can be configured to generate alerts based on Month to Date Cloud Spend for certain assets.
We’ll take look at how the Policy Engine works in CloudHealth to generate Alerts, and the actions that can be taken by a Policy.
Policies at its most basic is a set of rules that allow you to govern various aspects of your cloud infrastructure, such as cost, availability, security, performance, and usage.
The Policy Engine in CloudHealth is pretty powerful, it’s not just used to track cloud spend, for example:
- you can track the launch of new resources
- you can identify and terminate unused or underutilized assets
- you can track unexpected cost spikes
- you can track changes across the cloud infrastructure
- you can identify resources that have been created out of compliance with specific rules (ie region location, OS type, etc)
At the core of each policy is a rule, which monitors for one or more conditions and, optionally, responds with an action. Actions could be to send an email to notify that a policy has been triggered, or to power off an EC2 instance or VM.
Creating a Policy to alert on Month to Date (MTD) Cloud Spend
One of the most common policies created by CloudHealth customers is a policy to identify increasing cloud costs over a set time period. When overall costs in your cloud environment increase suddenly, it could be an indicator of a larger problem – for example, a compromised cloud account where attackers have spun up a large number of EC2 instances and VMs.
You can create a policy that alerts someone via email whenever the Total Cost of your cloud bill increases by more than a certain percentage:
Or even by a fixed amount:
You even have the granularity to set the conditions to focus on a single Account (in this example ‘Test account name’):
Whilst these examples have a time interval of 1 day, this can be changed to 1 week or 1 month to suit your requirements.
Most Policies allow you to filter the rule condition to focus on a specific account (eg. Test Account name), a specific service/asset type (eg. EC2 Compute), a specific Region, or even by a Perspective you’ve created (eg. Environment = Production):
Alternatively, you can create a policy for a specific resource type you may want to focus on, in the following example we’re just looking at EC2 Instances and want to be alerted if the total costs increased by 10% over 1 month, we could then take a number of different actions – email, delete EC2 instance, stop EC2 instance, etc:
CloudHealth vs vRealize Operations
Having used both CloudHealth and vROps, I would say it’s far easier to create reports, policies and alerts within CloudHealth compared to vROps – but I might be a little biased here… =)
The Cost and Usage reports are far better in CloudHealth – the added feature of being able to use filters, categorisations and Perspectives to change the viewpoint of the report visualisation is something that stands us apart from other tools! Not to mention that changing the visualisation occurs instantly, there’s no need to wait for processing to occur to rebuild the graphical data.
Within CloudHealth you also have far greater granularity to customise the policy conditions by using the filter capabilities.
One thing I constantly get asked is whether CloudHealth and vRealize overlap each other and perform the same functions.
They’re actually complementary management solutions as they are two different products providing information for different use cases within an organisation!
vRealize offers operational efficiency and automation and CloudHealth brings collaboration, governance, and optimization.
- vRealize focuses on driving efficient operations (i.e., provisioning, troubleshooting, capacity planning, automation) in the private and hybrid clouds. Providing Consistent infrastructure and operations, from the data center to the cloud.
- CloudHealth focuses on driving improved business outcomes (i.e., governance, optimization, visibility, chargeback) in the public and hybrid clouds. Breaking down public cloud silos and streamline cost, compliance and analytics operations.
It’s also worth noting that the starting point for the journey to multi-cloud can originate in the enterprise data centre or from the public cloud. Whether an enterprise is looking to expand its data centre to public or vice versa.
In the data centre, infrastructure/operation teams require tools for configuration, provisioning, automation, capacity planning and governance for all their data centre assets (ie Day 2 operations). It’s also very Capex-intensive and costs are somewhat stable and predictable. This is the perfect scenario for vRealize.
In the public and multi-cloud world, developers and lines-of-business users provision resources directly themselves. It’s very Opex-intensive and resource usage can be dynamic and unpredictable. The management disciplines needed for cloud-centric, de-centralized IT include ways to govern usage, optimise costs and deal with cloud security threats and vulnerabilities. This is where CloudHealth comes into the fore.
For example, vRealize can be used to help perform capacity planning assessments and ‘What If’ scenario modelling. CloudHealth can be used to model the cost of migrations from private to public cloud.
Anyways, I’ve realised that this has been a super long post so I’m going to end here. I hope it’s been useful reading…. I’m also hoping that I’ll get the chance to blog more often on CloudHealth and its features in the coming year! =)
For now, I hope you all have a Happy New Year! Let’s pray that 2021 will bring back some normality to the world!