Views, News & more
In long-term relationships, there may come a point where things are no longer working for both parties. The current situation is adequate but slowly getting worse.
Most organisations have had a multi-decade relationship with on-premises servers. But now that cloud services have improved, it’s no longer a given that this relationship will last.
On-premises hardware and software are aging, moving ever closer to their end-of-support dates, so the decision over what to do can’t be avoided forever.
Scalability is a problem. Most on-premises servers can’t be upgraded much, and there’s limited physical space for horizontal scaling.
Resilience is also lacking. When an on-premises server fails, applications break. There’s no alternative host on standby, ready to step in. Rebuilding from backups would take many hours, if not days.
The hosting environment is suboptimal too: only one connection to the electricity grid, no backup generators and half-hearted security, with no guards around for much of the night and weekend.
Connectivity is subpar. This wasn’t a problem when users were in the same office as the server. But since COVID, staff increasingly work remotely. This has made the office’s Internet connection a bottleneck.
If your organisation has been digitising, some on-premises servers may now serve external users - customers, suppliers and partners over the Internet.
Meanwhile, security threats have grown. Amateur hackers are no longer the main threat. Professional ransomware gangs now exist, and they have no problem taking advantage of the window between patches being released and applied.
Since you first bought your on-premises servers, IT has changed.
Cloud computing has become significantly cheaper.
Web-based SaaS offerings such as Salesforce, Zendesk, Asana, and Google Workspace have grabbed market share from self-hosted apps.
Organisations are ditching Exchange Server in favour of Exchange Online. They are retiring shared network drives in favour of OneDrive or Dropbox.
And it’s not just traditional server apps that are moving. Locally-hosted PBXs (the brains of office phone systems) are being replaced with cloud-hosted versions.
It’s not just the hosting location that’s changing. The switch to the cloud is resulting in faster rollouts of new features, improved service uptime and more frequent patching.
Your organisation is probably using a growing number of cloud services. Your end users and organisational leaders have likely shown little interest in whether your organisation’s apps are hosted locally or in the cloud.
You’re not alone in embracing the cloud’s charms.
IDC found that since Q4 2021, more money has been spent on cloud infrastructure than non-cloud infrastructure. Cloud is now the default, except where complex integrations are required.
AI is going to tip the balance of that investment even further in the direction of the cloud, with mass-market server apps increasingly dependent on the cloud to power AI-related features.
It is worth considering how you might migrate away from on-prem over the coming years.
The end-of-support dates for your servers’ software and hardware are inching closer. You need to consider whether it’s worth replacing that local stack. It may make more sense to move some workloads to the cloud prior to your current stack reaching its end-of-life.
This isn’t merely a matter of replacing on-prem servers with similar servers in a data centre. An important part of that shift might be to replace some locally-hosted apps with SaaS alternatives hosted in the cloud.
Move enough server apps and you may not need to replace your physical server at all. Hosting becomes your SaaS providers’ problem. This frees you from replacing not only aging servers but also outdated operating systems, database management systems, and email software.
If you still require a hosting platform, it may be worth migrating workloads to a cloud platform, so you get additional scalability, someone else takes care of the hardware, and you don’t have to pay in advance for capacity you may not need.
If the cloud is too expensive because your systems are processor intensive, it’s worth considering a switch to colocation. That can improve the hosting environment, connectivity and physical security. Colocation may also make economic sense if you require a lot of storage – for example, dozens of Terabytes.
Dial-up broadband, floppy disks and CRT monitors served everyone well for years. But the time came to ditch them in favour of something better.
This initially involves paying more for a higher-quality alternative. Over time, that alternative often fell in price or started to deliver more value at the same price point.
Platform-as-a-service and software-as-a-service prices tend to stay broadly unchanged in real terms while delivering a stream of functionality improvements that improve value-for-money.
Infrastructure-as-a-service firms have cut their prices over many years, delivering ever more compute, memory, storage and connectivity for a given expenditure.
There have been some hiccups in recent years, with surging energy prices raising the cost of server hosting – on-premises and in the cloud.
However, over the medium-term, the cloud is becoming more affordable – and a viable option for organisations of all sizes.
Ditching your on-premises services for the cloud can still be a bit nerve-racking.
How do you ensure the migration goes smoothly? How do you avoid performance issues? How do you avoid surprisingly large cloud bills?
Well, cloud migration problems aren’t inevitable. We’ve helped many organisations migrate their workloads to the cloud from on-premises stacks. We’ve also helped firms move to the cloud from expensive managed hosting providers.
Often, on-prem systems are poorly optimised. An oversimplistic lift-n-shift could prove costly.
But there are ways to keep costs down. For example, by switching to a platform where your virtual machines (VMs) can contend for a pool of resources, just like they would on-premises.
That way, you don’t have to size every VM based on peak usage, or suffer performance issues from sizing your VMs based on their average usage levels.
The good news is you have time to get it right. You can decide when it’s time to bring your relationship with on-premises servers to a close. That doesn’t have to be today, or next month. Instead, it’s at a time of your choosing.
You just have to kick off the planning now, so when your servers reach end-of-life, you’ve no need to replace them, because you’ll have moved your workloads to the cloud by then.
You’re taking the time to find a new partner that can take care of your hosting needs over the long-term. That requires delivering the combination of value-for-money, reliability, flexibility, scalability, security and support that suits your needs.
For some organisations, public clouds such as AWS or Azure may be the right choice, offering unrivalled scalability, flexibility and security.
For other organisations, smaller clouds may be the right choice, as they offer more in the way of support and value-for-money.
The key is to find the partner that’s the best fit to YOUR needs.
If you like the idea of moving your server-based workloads to the cloud, but would like to learn more about how that might work, you should speak to hSo.
We’ve got hosting experts who can guide you through your options.
You wouldn’t be committing to anything.
We’re a cloud hosting provider in our own right, as well as an AWS and Microsoft partner able to advise on AWS and Azure migrations.
To be eligible for a free consultation, you just need to be an organisation with server-based infrastructure in the UK or East Africa.
To claim your free hosting consultation, fill in the form at the bottom of this webpage or call our UK HQ on +44(0)20 7847 4510.
020 7847 4510
We may process your personal information in order to send you information you request, measure and improve our marketing campaigns, and further our legitimate interests. For further details, see our privacy policy.