REQUEST A QUOTE
Best Infrastructure as a Service
In many ways, it’s a simple choice. CapEx? Or OpEx? Do you want to lay out your dollars in advance, to buy all the servers, storage devices, monitors, cooling plant and backup facilities you’re going to need at times of peak workload? Or would you rather buy it all in as a service, and pay for it as you use it?
If you prefer the second choice, then IaaS is for you.
That’s a choice that the finance professionals in your company are going to make. But there are also good reasons why the CTO should prefer to focus on the applications that keep the company on stream and on target, and leave the infrastructure to someone else.
Here are the things that define IaaS:
- It’s self-service, and it’s on demand. Once you’ve signed an SLA, there should be no haggling – when one of your users needs access to a resource, they log on and use it. When they’ve finished, they stop.
- It doesn’t matter whether your user is using a smart phone or a workstation – access to the network is there, and from the user’s point of view there is no difference
- It’s scalable. There is a reason why that word appears in almost everything you ever read about the cloud; one of the biggest advantages of IaaS is that you can handle monster rises (and nerve-wracking falls) in workload without feeling any pain – and without risking a sudden inability to deliver
Those are the reasons for using IaaS – but how can you tell the good from the bad? It’s about performance and security. Here are some of the things you should expect from a best-in-class IaaS provider:
- The additional option for SaaS and PaaS from the same provider
- Fast processing. HDD is fine for secondary and archival storage, but you really want to see SSD up front
- Service level agreements (SLA) will tell you what the provider is committing to – and are therefore essential – but testimonials from other users will tell you what they actually deliver. And, when you read the SLA you’re first offered, check exactly what the availability clause says. Is there any kind of downtime that will not be considered a violation of the SLA?
And here’s a very strong recommendation. Would you consider buying a software package that had never been tested? Then don’t commit to an IaaS agreement that hasn’t been tested – by you. Of course, you have to pay for this, but it’s worth it for peace of mind. Run your applications on the platform you’re proposing to use. Run them several times, not all on the same day, and not all at the same time. And have the IT equivalent of a stopwatch on the job, because you want to check exactly what performance you’re getting.