Cloud is everywhere, the recent events worldwide have led to many cloud implementations being accelerated / expanded to facilitate home working or make an organization more flexible in a changing environment.
This article aims to outline the support plan options available from the cloud providers themselves (What are the Azure support plans? / Whats are the AWS support plans?) and their partners (Who are the partners? What do they offer?). Then ask the questions you have, in how it relates to you and your needs for the short or long term.
Maybe recent events have highlighted gaps in your capability to react to the unexpected or you have operational issues with your cloud implementation ranging from performance to cost to support and don’t know where to go next…
If you are reading this then you already know the question. This paper will aim to give you the options, so you can get some answers.
This outline aims to provide clarity about the choices facing your organisation for supporting your cloud deployments. The document focuses on Azure and AWS, but the points and background are relevant for any cloud platform that provides infrastructure as a service.
There is an assumption that your organisation already has cloud infrastructure deployed in a production setting. As this infrastructure has become more embedded within the organisation the platform has become business critical and for some reason the support options have become a concern.
There are often several drivers for the support options being reviewed and these could include:
Every organisation has some combination of the above and some that are unique to your business / sector. This document tries to break down the pros and cons of the different support options at the different levels and how they address the above.
I hope you find this outline useful and any feedback, good or bad, is welcomed. If there is anything you want to discuss further or areas where we have not answered your questions, please get in touch with me personally. (Steve.Rastall@igcloudops.com | 0151 332 3839)
There is a tendency with IT related areas to overcomplicate things and get bogged down in the technology. If you are reading this, you are probably working at a senior level within your organisation and not necessarily in an IT role. Over many years of working with customers I have found it helpful to use an analogy to explain the technology /players and how it fits together.
For infrastructure as a service, I have found the comparison to a national utility company to be one of the best. Also, this is one that almost everyone will be familiar with and can be used to explain how it all fits together for a technical as well as a non-technical audience at any level within an organisation.
The table below shows an example breakdown of the different players and their roles. The example below uses the electricity grid, but another example would be water or telecommunication utilities.
|
Infrastructure as a service |
Example utility (Power Grid) |
Notes |
|
Azure (Microsoft) / AWS (Amazon) |
National Grid infrastructure then power companies e.g. British Gas |
The utility company provide the infrastructure service but apart from billing and an online presence / helpline their direct interactions with the majority of customers are limited |
|
Partners |
Contractors |
These range in size from one-man bands to large companies with a direct relationship with the utility company almost acting as them for their customers |
Both Microsoft (Azure) and Amazon (AWS) offer direct support packages in a very similar manner. Each offer a free level of support which is led by online forums and wikis. Then there are one or two levels of support for development / production deployments which enable troubleshooting and tickets. Then there are the more enterprise levels which are either not offered unless your organisation is a certain size / profile (Microsoft) or are priced highly (Amazon).
The packages available from a cursory examination tick all the right boxes once you get to a middle tier. They mention all of the points that you might look for in a standard support agreement:
This appears to tick all the boxes. However, these mid-level packages have to balance what they will take on and provide boundaries where their responsibility ends.
In the case of a system failing, your team could bring it back up and log a ticket when it doesn’t work as expected and then wait for a diagnosis. This then comes back confirming that all the infrastructure is working as expected or that there are errors being thrown by the software you have installed. At this point the ticket is resolved.
Another scenario is that you have a deployment which is experiencing performance issues and needs to be restarted periodically or it will fail. A ticket is logged and the infrastructure is checked and no errors are found. At this point the ticket is resolved with the feedback.
In both the scenarios above the provider has met their obligation and ensured that the infrastructure as a service is running as expected. There might even be a series of recommendations, but these types of tickets will drag on and there is often no clear way forward.
In my opinion the support available from the providers (the utility companies) is limited unless you are large enough or prepared to pay enough. This point is made by AWS on the front page of their support overview with the pricing of the plans only Enterprise has access to a concierge team and the cost example starts at around a minimum of $15k per month.
The support is also inflexible by its very nature. Let’s assume that both AWS and Azure each have hundreds of thousands of deployments across hundreds of thousands of customers. The support levels offered by the providers have to service all of these customers and not only that they need to be scalable so that they can be offered. By its very nature, these constraints mean that it will be limited to what can be found as common ground by the providers to offer across all of these customers.
The providers get around this problem by providing a basic level of support that will scale so that there is something there for customers to fall back on but then having an extensive network of partners who then provide tailored services to their customers.
This is a consistent approach with all large technology providers like CISCO / Apple not just Microsoft / Amazon. It’s an approach that works which is why Amazon implemented it when they started to offer infrastructure as a service.
Both Amazon AWS and Microsoft Azure have a partner programme, which vets to different levels the partner organisations and individuals within them to help give some guidance to customers choosing them.
To go back to the analogy of a utility company the partners come in all shapes and sizes and offer a variety of services.
The utility company is not interested in fixing a lightbulb or installing a car charging point. This is the bread and butter of the partners and it’s an area that the utility providers have no interest in providing.
When looking at the market it’s useful to understand how the partners are structured. This influences the services they provide and how they will interact with your organisation.
The partners generally fall into the following categories:
If you run a google search for AWS support or Azure support, you will be bombarded with pay per click ads from different partners offering their services. They will be a range from all of the partner types discussed previously, and they will all be offering ‘support’ but what does this actually mean for you?
Many of the partners are offering support services which amount to a traditional IT support contract. Along the lines of if it breaks or you have a problem raise a ticket and we will respond. They also provide additional services for consultancy or projects which can then be resourced as required.
A few are offering some form of network monitoring, but this is not hooking into the feeds that the platforms provide. The partners offering this more sophisticated service are using traditional network monitoring so they are only looking at one level of the platform, they are getting none of the underlying information. They usually have a networks background so this can look attractive but it is only skin deep.
As an example, if you only monitor a virtual machine with network software you will know the windows version for example and if an update needs to be applied, which is good. However, you will not know any information about the VM from the platform and what it costs / when it needs to patch etc…
This will only get you so far with cloud infrastructure because as discussed in the introduction to this article the range of areas that support needs to cover and provide feedback on is too wide ranging.
Something a bit different is needed to work with your cloud infrastructure and provide the support you need.
IG CloudOps are cloud support and management specialists and have worked with cloud infrastructure since its initial release. We are partners with both Amazon and Microsoft and can provide support on both your own subscription and host you on ours.
We offer something a bit different; we have hooked into the feeds the providers have so we can give you insights about your infrastructure. Initially we speak to many customers about cost as this can be an important driver in our initial discussions. Making sure that you are not overpaying for the infrastructure you have or running something at a higher spec than needed for your usage.
We tick all the traditional support boxes and then have our own technology CloudOps which takes things further. We have tech and people, a UK based team providing support and monitoring.
CloudOps addresses all of the areas discussed in the introduction.
For more background and the core areas CloudOps covers please see the following information about CloudOps core features.
If you review the market it seems like there is a lot of choice at different levels and price points. However, 99% of the market is offering a variation on a traditional support contract. Even the providers are doing this just with more inflexibility and a higher price point.
This might be a fit for your organisation's needs or part of them for a while but from our experience at IG CloudOps this will only cover part of the problem.
Cloud infrastructure touches all areas of your organisation, so why would you only look at one area when choosing a support partner. Its no good only being able to fix the cost if you cant get the performance.
We are proud of our CloudOps offering and believe it will help you get the best out of your cloud infrastructure so please get in touch and we can show you how we can help.
In conclusion I hope this outline has provided some useful background which will enable you to make choices around what your organisation needs when it comes to cloud infrastructure support based on what’s being offered.
If you take nothing else from this document, I hope it has provided some background on what the options are and how the partners fit together with the providers.
CloudOps is designed to solve these exact problems, based on our team’s experience in supporting and managing cloud environments since 2010.
CloudOps offers highly advanced performance and cost monitoring tools, support, and administration through one single portal to manage your cloud for you. This includes:
CloudOps is a proven way of reducing your cloud spend and optimising performance. The system can be implemented across any new or existing AWS and Azure cloud environment, allowing you to focus on delivering applications and services, not managing your cloud infrastructure.
Unlike a traditional cloud support service, CloudOps already includes all the features and functionality your business needs without any hidden costs or surprise charges.
CloudOps will:
If need help with any of the challenges explored and would like some additional guidance, please do get in touch with us and a member of our team will be happy to talk through your needs.