We all know that the challenges caused by COVID-19 are far and wide and, put in that perspective, the impact on IT operations should not be over-stated. However, that’s much easier said than done when the platforms that underpin your users’ and customers’ services are your responsibility, and suddenly vulnerable. Perhaps it’s even a good distraction from the outside world? Perhaps…
What Impacts Have We Seen So Far?
Firstly, there was an impact on supply chain. There were pre-existing constraint challenges for certain chip sets and other enterprise peripherals which was compounded when Chinese factories closed in January. Clearly manufacturing sites around the world have since had similar obstacles to overcome in terms of employee travel, site access, and subsequent logistics complexity. But (touch wood) that impact has not been as severe as we initially feared it might be, and the majority of vendors have seen little impact on the lead time of mainstream storage, server, and networking products. We remain cautious and getting accurate ETAs is still difficult but, as the world adjusts to a new normal, we seem to be getting there. Again, touch wood!
The second, and very real, challenge is how to deploy and maintain infrastructure. We have spoken about some of the challenges of operating a datacentre remotely (take a look at Andrew Cochrane’s Perspective: Aloha! This one’s for you, Bob.), however clearly the issue of deploying new hardware is difficult to circumvent. If anything, there is an increase in bandwidth from a skillset perspective; the Softcat services team being one example! Clearly there is a need to use good judgement to determine what is a critical task and we advise severe caution in avoiding travel – as would any employer.
Cloud alternatives therefore seem like a logical solution. Speed of deployment and low upfront cost has made Public Cloud ideal for those times when you urgently need more capacity or need to react to a short-term project demand. For example, hosting more remote users or the sudden need for a different type of customer service.
Nothing is that easy it would seem nowadays, and the capacity challenges that Azure have seen over the last week or two are evidence of that.
The good news for some public sector organisations is that Microsoft are going to prioritise critical workloads, such as those associated with health & emergency services and critical government functions. But for all organisations this is a reminder of the need to underpin production workloads with enterprise availability; regardless of where they are hosted.
For existing workloads in Azure, you could consider turning off any automation on auto shut down and start up VM’s (scheduled VM’s), and leave them running 24/7 and on max scale out/capacity. This will ensure the VM’s will be able to be kept running, as the current restrictions and capacity issues are resulting in auto shut and start up VMs from coming back online and scale out services scaling out to maximum capacity. However, we would recommend discussing any Azure issues you have directly with us first.
Scaling Public Cloud
We anticipate these challenges will be resolved shortly, while other hosted infrastructures do not seem to be have been affected by the surge in demand. Therefore, the option of scaling up Public Cloud remains a good one, and here are some tips for those that are thinking about doing so in earnest.
Define the Use Case
When using Public Cloud, it’s important to understand what you want to use it for. Even in this time when using Cloud appears to be an easy option defining the use case is still crucial!
Ensure You Have the Right Architecture
Ensure the foundations for architecting to best practices are followed. Building services in Public Cloud can mean constructing workloads using a complex range of services, all tied together to deliver a service outcome. Softcat can provide best practice guidance from our certified Cloud solution architect teams, whilst also partnering with the respective Public Cloud vendor to obtain usable credits to help accelerate and provide a potential payment holiday.
Connect and Secure
Ensure you can set up the right connectivity with options that include IPSec VPN, AWS Direct Connect and Express Route, to name a few. It is worth noting that your existing network provider could provide options too. Ensure you understand how to secure the environment via security groups and load balancing using the native security tools, or use 3rd party tools from trusted integrated security software vendors.
Build, Operate and Run
Make sure you understand how to setup account organisational structures and on-going tagging policies to help with billing and invoicing and to simplify on-going operational management. This is critical to help with commercial decisions that could include reviewing Microsoft hybrid use benefits, reserved instances or AWS saving plans – look at potential long commitments as a way of receiving discounts. Ensure that building the environment and testing services are factored into the ongoing service deployment, which could include migrating services or integrating these from an existing source.
Review, Optimise and Automate
Continually reviewing and monitoring actual usage is critical to ensure you stay on top of the operations of deploying and running workloads. The speed to deploy can be made easier following the steps defined here, but maintaining an optimised secure environment as part of a pro-active service model is critical. Moving into creating automated runbooks can create the efficiencies required to scale up and down based on usage demands.
As organisations move from a focus on remote worker productivity to adapting to a change in customer behaviour, there will be a surge in digital initiatives.
It is also worth saying that for organisations with skills shortage, or who are just short on time, remote services capabilities are more relevant now than ever. If you would like some guidance on how to configure or review these best practices; please do get in touch. We are here to help, whether as a friendly voice to share our experience, or as part of a more formal scope of works to review or deploy.
We believe that in the coming months, as organisations move from a focus on remote worker productivity to adapting to a change in customer behaviour, there will be a surge in digital initiatives. Whether those be best served through a Public Cloud platform or an on-premises private Cloud; and whether that can be done with existing resources or needs new; we’re here to help. For Softcat’s existing customers, we believe now is our chance to support you more than ever. And this is a commitment to those organisations that haven’t worked with us before too!