Virtualization Technology News and Information
5 Steps To Resolve Any Cloud Computing Issue


Regular IT system diagnostics is remarkably different from the troubleshooting work you'll need to do with cloud-based systems.

Why? Because cloud services, like Amazon Web Services, Microsoft Azure and the many others, all function differently than internal networks and platforms most IT professionals typically working with. Not to mention, many believe it's a black box - a system or device that you can view in terms of its inputs and outputs, even with no working knowledge of the internal hardware. That is certainly not true.

Cloud platforms do have an added benefit worthy of the "black box" moniker, as many of the tools and APIs available are incredibly easy to use. But that doesn't help much in the middle of a meltdown because you have to understand first what tools to use, and second, how to even use them.

Furthermore, cloud systems, cloud communications and the networks facilitating these connections are more complicated than even the most skilled IT professional may realize. With local systems, it's as easy as fixing the problem then and there, or making a few changes. The IT team must review the remote systems first from the perspective of the host, then the client. Then you can make the proper adjustments system-wide, sometimes across an entire network of machines.

Considering that cloud computing is likely to become the default corporate policy by 2020, it's time we address these issues.

That doesn't mean it's impossible to resolve cloud computing issues, quite the contrary. It simply means that there's a different science to it if you will.

1. Start With the Infrastructure and Your Service Provider

The first thing you'll want to rule out is whether or not your service provider is keeping up their end of the partnership. If you chose a vendor with the appropriate infrastructure and technical expertise, you'll likely come away from this step unscathed - and so will they.

But even when a third party provides the remote hardware, there are monitoring tools you can use to determine CPU, storage and memory utilization by various systems and servers. This includes both on the client and server side of the equation.

More importantly, you need to ensure that the provider can scale according to your needs and that they know to do so. Don't just assume that a system will scale up to meet an increasing performance demand. Check with your cloud provider beforehand, and be sure the infrastructure in place is ready to handle any growth.

2. Check Your Apps and Software

As mentioned in the previous step, there are a variety of apps and software tools you can use to peer into the entire network. Use them, as often as possible, and be sure to conduct regular audits of your system. Amazon CloudWatch is an excellent example for AWS, while AppDynamics is another offered by popular firm Cisco Systems. Then there's CA Technologies, Datadog, BMC, New Relic, AppNeta and many others.

You never know what your monitoring tools will uncover. For example, you may find that a single, prominent application is the cause of severe performance issues, and all your employees are using it currently. Or, you may discover that the software in question simply wasn't designed for use with a cloud-based solution or environment. If you didn't know any better, you'd pin the blame on the infrastructure when it's the applications moonlighting as the main culprit.

3. Factor in Outliers

Step three means you've ruled out the major contributors to performance and platform issues. Now, it's time to review the outliers, or components that could directly affect network and system performance. This includes the security system, authentication tools, encryption, firewalls and more.

Sometimes, even the very tools you use to measure performance can hit a snag and lock up the entire network. Therefore, you should never rule out a potential component here. It's possible application, software or operating system updates are to blame. Maybe there's a bug in the latest version of your security suite?

The only way to know for sure is, again, to audit the various outliers you have connected to your network and system.

4. Review Your Network

When things go awry, the first to take the blame is always the cloud provider or cloud service in question, which may not have anything to do with the problem. Unfortunately, with cloud-based platforms, there are a lot of parties involved. You have your provider and their platform or software, but there are also your clients and users, any and all associated hardware and, of course, the network. That network has its own set of reliant parties attached, such as an ISP, or internet service provider.

Because a cloud solution requires internet access, and a great deal of bandwidth, that means if there are problems with your network or access point you'll encounter problems with your system too. Maybe there are too many devices connected to your internal network, so the total bandwidth is all used up. Perhaps your ISP is experiencing hardware issues, and a ping test is in order.

5. Don't Forget Your Users (Client-Side)

One of the easiest and most common oversights to make when working with cloud-based systems is to forget there are two sides of the equation. That is, you have the local host equipment, servers, and storage platforms, but you also have the client/user hardware to contend with. Just as the people inside your company can be the most prominent cybersecurity threats, so can the users and customers involved with your system. In addition to being security threats, they can cause software, hardware, and performance issues, primarily due to lack of knowledge and experience.

Even something as simple as intrusive malware on a user's computer can cause significant issues with cloud-based applications. Or they may be using an incompatible browser or an older version of the operating system that needs to be updated.

In fact, you may find that issues are caused primarily due to user-error as opposed the systems themselves. That's why it's always a good idea to double check the platform or devices your clients are using before ruling out a potential problem.


About the Author

Kayla Matthews is a tech-loving blogger who writes and edits Follow her on Twitter to read all of her latest posts!
Published Thursday, November 30, 2017 8:12 AM by David Marshall
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<November 2017>