This post was authored by Maryam Sanglaji, Product Marketing Principal Nutanix
As Yoda would say “Wake we must.” More and more applications are being built for cloud. Why? Because cloud is agile, scalable, and simple. It is driven by constant innovation, has a simple consumption model, and it just works. These are all really attractive features for many enterprise customers. However, due to various reasons such as legal, security, or costs - many enterprises decide to use public cloud for only a small portion of their workloads leaving them with complex IT environments. Thus, management gets fragmented and specialized teams are required to manage these different technology stacks.
It is evident that enterprises want to create a cloud-like experience for their IT customers across their entire IT environment (e.g. branch offices, remote offices, main datacenters). These enterprises also want to avoid hefty IT costs and lock-in. Since each cloud vendor has its own unique software stack, once an application is built for a certain cloud vendor it cannot be moved to another without extensive changes. Just like the old days with Windows and Linux operating systems, today’s cloud war is happening for enterprise apps. It is unclear who will be “The Last Jedi,” but Nutanix Enterprise Cloud OS is built for all clouds - eliminating lock-in, providing automation and simplicity, and encompassing a unified solution for all IT environments.
If you’re wondering which combination of clouds is best for your organization, you need to consider machine learning (ML) and artificial intelligence (AI). Native integration of these technologies into the cloud software stack directly influences the resulting performance and experience.
Patched services will never realize the true potential of ML and AI in traditional IT infrastructures, especially when it comes to delivering simplicity. Simplicity of IT operations is a true north for Nutanix. It was understood from the early days that to deliver true simplicity through one-click operations requires native integration of ML and AI into the core software. And this is exactly what Nutanix does.
Not every enterprise has the luxury of having an army of specialized IT resources to address its infrastructure needs. Things like simplified risk management, optimization, and ease of operations are not possible without ML and AI technologies. This is why Nutanix Enterprise Cloud OS is different from other cloud vendors. By natively integrating ML and AI into our core software, systems can make intelligent decisions and deliver a one-click operational experience. Let’s look at just a few examples of where Nutanix is leveraging these technologies.
Anomaly Detection to maintain cluster health
This year on black Friday, both Macy’s and Lowe’s had technical IT glitches that slowed customers’ purchasing activity. In this highly-competitive market, business continuity is key to business success; slow application response times can cause businesses to suffer and lose millions. Datacenters need to maintain health across their clusters even when various hardware or software issues create node degradation, causing the whole cluster performance to drastically suffer.
Every scale out IT architecture should be designed to maintain health and availability at all times for this particular situation. Nutanix leverages a clustering ML algorithm along with a distributed set of degraded node monitors to identify the degraded node. Once the degraded node is flagged, an alert is generated and the leadership and critical services will not be hosted on that node. As a result, technology ensures cluster health and high availability. Therefore, business revenues won’t suffer due to node degradation.
Optimization & Proactive Placement
Maintaining consistent performance is both critical to the business and important for meetting application SLAs. Nutanix AHV includes Acropolis Dynamic Scheduler (ADS) which leverages an AI algorithm to improve VM placement while maintaining a set of constraints such as VM affinity or anti-affinity rules. ADS is always on, responds to hotspots proactively, and enables full resource utilization without compromise. Different VMs have predictable performance while their individual constraint rules are maintained.
VM Behavior Learning
I am sure you can think of a scenario where the VM, along with specific resources, was created for a user who didn’t end up using it. This is one of many examples of datacenter inefficiencies where hardware resources are not fully utilized - resulting in resources silos.
The Nutanix X-Fit engine within Prism Central eliminates these inefficiencies using time-series analysis algorithms that identify VM behavior patterns. This learnt behavior helps the system detect anomalies, generate smart alerts and optimize resource utilization.
Smart Planning & What if Analysis
Guesswork and spreadsheets! How many different management consoles do you need to monitor before making a critical and expensive decision for your datacenter operations? How can you avoid the cost and inefficiency of over provisioning? Nutanix X-Fit engine not only helps with the VM behavioral analysis described above, but also provides accurate forecasting. These forecasts assist customers in estimating their true resource needs, empowering them to optimally size hardware resources for specific workloads, and in making their datacenter operations more efficient.
“Smaller in number are we, but larger in mind.” Machine learning and artificial intelligence allows our datacenters to be “larger in mind,” making smart decisions on behalf of IT. This allows enterprises to deliver a cloud-like experience across all of their environments - operated by IT generalists. As Nutanix President Sudheesh Nair nicely put it at .NEXT conference in Nice, this is how “Nutanix brings delight in the datacenters. Nutanix architecture is built with a purpose of simplifying and making the infrastructure invisible.” Invisible simply means no need for an army of IT resources.
Follow us at Nutanix Enterprise Cloud and learn more about how we utilize ML and AI technology in an upcoming post called ‘The Cloud OS Awakens : A Deeper Dive’ - stay tuned.