How to Build Hybrid Cloud Confidence

This blog post is coauthored by Vamsi Chemitiganti, Chief Strategist at Platform9 Systems.

The hybrid cloud is not a new concept. Way back in 2010, AppDynamics founder Jyoti Bansal had an interesting take on hybrid cloud. The issues Jyoti discussed more than eight years ago are just as challenging today, particularly with architectures becoming more distributed and complex. Today’s enterprises must run myriad open source and commercial products. And new projects—some game-changers—keep sprouting up for companies to adopt. Vertical technologies like container orchestrators are going through rapid evolution as well. As they garner momentum, new software platforms are emerging to take advantage of these capabilities, requiring enterprises to double down on container management strategies.

Building Applications: 2008 vs. Now

When you consider how applications were built a decade ago, today is truly an amazing time to be a software engineer. From a commercial Java standpoint, software vendors in 2008 may have released only a few variations of their application—for instance, specific WAR (web archives)—to account for differences in application servers. Fast forward to today: commercial vendors must expose entirely how the sausage is made and provide a litany of formats to deploy and integrate with CI/CD and hybrid cloud infrastructures.

The 2008 point of view: commercial vendors delivering just a WAR.

Enterprises today are doing more due diligence. It’s not uncommon for enterprises to ask for source/binaries, quality/security scans, infrastructure-as-code templates, container orchestrator descriptors, and cloud vendor deployment scripts.

With enterprises needing to run a wider array of technologies, decisions must be made on where to run these workloads. The hybrid cloud allows enterprises to pick and choose which workloads and tools to run internally, and when to leverage a public cloud service.

Current Hybrid Cloud Prospective

The hybrid cloud is composed of six fundamental components, as shown in the illustration below. This diverse infrastructure spans public and private clouds to deliver a mix of VMs, containers and bare metal deployments; a SaaS-delivered management plane that guarantees 99.999% uptime; a certified catalog of applications for developers with single-click self service, and an ability to run application architectures composed of various runtimes (e.g., Tomcat, NGINX, Kafka, Istio, Spring Boot, and so on); CI/CD toolchains; line-of-business self service; and application architectures that are a mix of stateless microservices, stateful applications and serverless functions.

The six pillars of the hybrid cloud.


In a cloud-based infrastructure, the key concerns facing enterprises include complex cloud management, overall cost of infrastructure, and the lack of a single pane of glass for metrics. These lead to pain points in Day-1 and Day-2 operations, a lack of choice for developers, slow provisioning times and overly complicated management stacks.

The Key to Hybrid Cloud Success

Success depends on your ability to quickly deliver all of the above components: a lean hybrid cloud in days and weeks, not months. With an effective strategy, an enterprise can bring together all of the six component technologies, creating a hybrid cloud that “just works.” This includes the ability for both developers and infrastructure admins to deploy applications and workloads on any underlying cloud provider, a single pane of glass across all the underlying clouds, and a robust catalog of application runtimes for developers.

Incremental Success Develops Cloud Confidence

The ability to deliver a lean hybrid cloud in days or weeks is an ambitious goal. When picking workloads to run on the private or public cloud, metrics become a crucial component of your strategy. One of AppDynamics’ core strengths is its ability to compare workloads and platforms for effectiveness. When a new application works effectively in your hybrid cloud infrastructure, you gain confidence in your investment and quickly bring additional workloads into the new model. By using AppDynamics to trace a business transaction such as user conversion, IT can easily justify where and why a workload is running in a particularly segment of the hybrid cloud environment.

AppDynamics provides essential KPIs for comparing workloads—a great resource for justifying your cloud migration strategy.

The Hybrid Cloud Revolution Continues

As our cloud topology marches towards cloud infrastructure nirvana, where organizations can reallocate workloads to the most prudent infrastructure provider(s)—either on- or off-prem—confidence in your hybrid cloud is crucial. Look to AppDynamics and Platform9 to help you navigate this exciting new world!

The Top 5 Advantages of SaaS-based Application Performance Management

Software-as-a-Service (SaaS) has received a lot of success and adoption in the past five years, but not as much in the field of application performance management (APM) than it has in other markets. With Cloud computing gaining momentum, you’re likely to see SaaS APM adoption increase significantly as more applications are deployed to the Cloud. Gartner also recently made SaaS a mandatory requirement for APM vendors to be included in their 2012 APM Magic Quadrant, so SaaS-based APM is definitely becoming hot right now!

Here’s the top 5 advantages that SaaS-based APM can offer:


SaaS-based APM can be deployed within your organization in the time it takes you to read this article. Think about that for a second – you get to experience the full benefits of APM in just a few minutes with no interaction from sales people or technical consultants. All you need to do is sign up for an account, take a free trial, and evaluate whether APM can meet your needs or solve your problems.

Many cloud providers are now actively partnering with APM vendors to embed agents within the servers they provision for customer applications. I personally know of a company that solved a 6 month production issue within an hour of deploying SaaS-based APM. How about that for ROI and time to value!


Simply put, subscription-based licenses are cheaper, more flexible and less risk than owning perpetual licenses. Annual maintenance is included in the subscription, as is the cost of managing and supporting the APM infrastructure required to monitor your applications. You don’t need to buy hardware to run your APM management server, and you also don’t need to pay someone to manage it either – you simply deploy your agents and you’re all done. There’s now no need to sign up to a multi-million dollar 3 year APM ELA agreement with a vendor; rather, you can pay as you go. If the APM software rocks, you renew your subscription. If the APM software sucks, you go elsewhere.


When a customer signs up for a SaaS account and evaluates APM for the first time, there is no pre-sales or technical consultant sitting next to them to configure or demo the solution. The experience from account registration to application monitoring is a journey taken alone by the customer.

First impressions are everything with SaaS. Therefore, the learning curve of APM in this context must be faster and easier, so the APM solution can sell itself to the customer.

SaaS-based APM solutions are also much younger than traditional on-premise software, meaning the technology, UI design principles, and concepts applied are more superior and interactive for the user. Try comparing the UI of an iPhone with a Nokia phone from 5 years ago and you’ll see my point.

First generation APM solutions were typically written for developers by developers. Today the value of APM touches many different user skill sets. It is therefore no surprise that SaaS-based APM can appeal to and be adopted by development, operations and business users.


When an APM vendor announces a new release of its software with lots of cool features, it’s normally down to the customers themselves to migrate to the new release. If things go well, they might spend several days or perhaps a few weeks performing the migration. If things go badly, they might end up spending several weeks working hand in hand with the vendor to complete the migration.

With SaaS-based APM, the vendors themselves are responsible for the migration. Customers simply login and they get the latest version and features automatically. They get to harness APM innovation as soon as it’s ready, rather than having to wait weeks or months to find the time to migrate by themselves. If anything goes wrong, then it’s the vendor who spends the time and money to fix it, rather than the customer.

Customers today will typically upgrade their APM software once a year because of the time and effort. With SaaS-based APM, they can receive multiple upgrades and always be on the latest version.


Enterprises and Cloud providers can manage lots of applications, which can span several thousand servers. It is one thing for a customer to deploy APM across two applications and a hundred servers in their organization. It is another thing to deploy it across fifty applications and a thousand servers.

Scaling APM has never been easy. The more agents you deploy, the more management servers you need to collect, process, and manage the data. How quickly can you purchase, provision, and maintain the APM management infrastructure when you’ve got hundreds of applications you want to monitor?

With SaaS-based APM, you let the vendor take care of that for you. I know of a SaaS-based APM user that monitors over 6,000 servers in their organization. Compare that with the largest APM on-premise deployment you know of and you can see why SaaS-based APM is a better scalability option.

So there you have it–five compelling reasons why you should consider SaaS-based APM in your organization. SaaS-based APM isn’t for everyone, though. I typically see less adoption in financial services customers where data privacy and security controls are much tighter.