From the course: Cloud Computing Essential Training: Deployment Models, Operations, and Migration Foundations
Introduce deployment strategies
From the course: Cloud Computing Essential Training: Deployment Models, Operations, and Migration Foundations
Introduce deployment strategies
When we talk about deployment strategies, we're deciding where Brightlane retails applications and data lives, who operates the infrastructure, and how those environments are connected. Brightlane doesn't pick just one model. It assembles a mix of private, public, hybrid, and multicloud to match different needs. Note that multicloud and hybrid cloud have similar patterns. We'll get into these later in the course. Brightlane runs its payment processing and HR systems in a private cloud hosted on a co-location facility. The environment is dedicated to Brightlane with strict network isolation and compliance controls. They use virtualization and automation to get cloud-like self-service and elasticity, but keep sensitive data in tightly governed footprints. This is their high control, high assurance zone. For customer-facing digital channels, Brightlane uses a public cloud provider. The e-commerce site, mobile APIs, and marketing landing pages run on shared infrastructure logically isolated from other customers. Brightlane benefits from rapid scaling during holiday peaks, global content delivery, and easy access to manage databases and message queues without buying hardware. Here, speed and innovation win over deep infrastructure control. Over time, Bright Lane connects these worlds into a hybrid cloud. Order data flows from the public cloud e-commerce front end into back end systems running in the private cloud. Secure VPNs and identity federation let services talk to each other as if they were in one environment. The rule of thumb, keep regulated or latency sensitive components in the private cloud and use the public cloud for elasticity, experimentation, and the new digital experiences. As Brightlane matures, it adopts multi-cloud, one provider for e-commerce and APIs, and another for advanced analytics and AI. Transactional data is streamed from one provider A into provider B, where data teams run large-scale analytics and recommendation models. Results are exposed back to the e-commerce stack via APIs. This lets Brightlane use the best capabilities for each provider while accepting the extra complexity of managing more than one cloud provider. So what's the difference between multicloud and hybrid cloud computing? Hybrid cloud and multicloud both involve using more than one environment, but they describe different patterns. Hybrid cloud means combining at least one private environment, which is on-premises, data centers, things like that, with at least one public cloud provider connected so data and workloads can flow between them. For example, Brightlane keeps payments and HR in a private cloud while running e-commerce in a public cloud linked with secure networking and shared identity. That's hybrid, even if they only use one public provider. Multi-cloud means using more than one public cloud provider, whether or not a private environment is involved. Brightlane might run e-commerce on provider A and analytics and AI on provider B, choosing each for its strengths. That's multi-cloud, even if nothing runs on-prem. You can be hybrid without multi-cloud, multi-cloud without hybrid are both at the same time, depending on how you mix private and multiple public clouds.