“Good morning, IT staffer. Your mission, should you choose to accept it, is to migrate all of our enterprise applications to the cloud.”* If you feel like you and your team have been handed an objective that rivals a Mission Impossible episode, you’re not alone.
From the television series Mission: Impossible
which aired 1966-1973, starring Peter Graves, et al.
According to a survey of 400 IT professionals conducted by iLand, 43% of application migrations to Microsoft Azure and 57% of migrations to Amazon Web Services stalled or failed. That’s a staggering failure rate. On average, cloud migrations are a 50-50 proposition at best. Most businesses would agree. An initiative with only a 1 in 2 chance of success is incredibly risky.
But yet, according to Gartner, “The worldwide public cloud services market is projected to grow 17.5 percent in 2019 to total $214.3 billion, up from $182.4 billion in 2018.” Business Wire cites that “the global hybrid cloud market was valued at USD 40.62 billion in 2017, and is expected to reach a value of USD 138.63 billion by 2023, at a CAGR of 22.7%, during the forecast period (2018-2023). That’s an amazing growth rate.
Which leads to an obvious question: “If it’s so risky, why are enterprises flocking to the cloud in droves?”
The answer is simple. Mobile access has permanently changed IT foreverand new technologies, such as the Internet of Things (IoT) and 5G networks will change it even more. To respond to the dynamic and sporadic application access demand patterns generated by mobile access, IT infrastructure has to be equally dynamic.
Cloud infrastructure is the embodiment of dynamic. It’s built to be elastic so that enterprises can add IT capacity during peak demand periods and then relinquish it when it’s not needed. Accomplishing that with fixed on-premise infrastructure is next to impossible. There would either be too little capacity for peak demand or excess capacity that is idle except during peak periods. Not a cost effective solution and why elastic cloud options are a huge benefit.
But then, there is that pesky migration failure rate problem.
The biggest reason why cloud migrations fail so often is because the migrated applications don’t perform as well from the cloud as they do on-premise. You might ask, “How can that be? The cloud infrastructure is the same as what we have on-premise.”
True, although the infrastructure is comparable, the ability to see app performance like you can on-premise is occluded in the cloud. On-premise performance monitoring is facilitated by real user experience monitoring (RUM) and infrastructure monitoring (network taps and spans) tools. The latter enables you to monitor server to server interaction performance and you can add as many of those as needed to ensure optimal application performance.
In the cloud, it’s not possible to add physical network taps and spans to the cloud infrastructure. And, even if it were possible, it’s not practical because fixed hardware devices aren’t elastic. That makes cloud server to server interactions invisible because RUM tools don’t show server to server interaction performance. If that’s where the problem is, there’s no way to know until users are complaining.
In order to provide cloud performance monitoring capability, some vendors are offering a new technology – virtual taps and spans – to provide server to server monitoring in the cloud. It can capture the network traffic between servers, which is a benefit but also a problem. The reason is because the captured data needs to be sent to a capture device for performance analysis. For a 10Gbps link, additional 10Gbps is sent over the cloud network to the capture device.
You have to pay for all network traffic in the cloud meaning that you’ll be charged for transferring the same data twice. Not to mention the negative impact that amount of network traffic can have on cloud app performance.
That means that the traditional methods, even when adapted for the cloud, aren’t viable or cost effective for monitoring cloud app performance. A cloud monitoring solution not only has to provide 100% performance visibility – before, during, and after the cloud migration – but also has to be elastic and cost-effective.
Accedian SkyLIGHT NAPM Solution
That’s why we created SkyLIGHT PVX – an integrated, software-only network and application performance monitoring (NAPM) solution specifically designed for monitoring cloud migrations. SkyLIGHT PVX uses lightweight, passive traffic capture points that forward a thin stream of metadata to a server-based capture store. The metadata is between 0.2 and 0.5 percent. For a 10 Gbps link, that’s only about 2 to 5 Mbps as compared to the full 10Gbps captured.
Accedian SkyLIGHT PVX provides:
- 360o visibility of all network traffic throughout the entire migration process
- Complete end user experience status for all applications
- 100% of application transactions visibility
- 100% of application backend performance information
These capabilities make SkyLIGHT PVX an ideal solution for ensuring that your cloud migration is a ‘Mission Possible’, or better yet a ‘Mission Accomplished’!
Download our Cloud Migration solution brief to learn how you can improve your migration success by following our six step cloud migration guidelines.
* From the television series Mission: Impossible which aired 1966-1973, starring Peter Graves, et al. which inspired a series of films by the same name starring Tom Cruise.