From Next-Generation Data Center
Building Next-Generation Data Center is a 6 week highly interactive online course. You'll study the background technical material on your own, solve design assignments together with your peers, and participate in online discussions and live design sessions.
The course is split in six sections:
Collecting the Requirements
In the initial session we’ll focus on the big picture:
- When to use private infrastructure or public cloud
- How to identify services that have to be offered by the data center infrastructure
- Identifying the reliability and recovery requirements
- Multi-DC considerations
- The strategy for integrating with public cloud
- DC placement (on-premises or colocation)
- How to collect the infrastructure requirements
- The approach of top-down design (start with orchestration system)
Automation and Orchestration
The automation and orchestration topics are usually discussed as an afterthought, but we’ll start with them because they are one of the most crucial aspects of a successful data center deployment.
We'll figure out what our single source of truth is, what data storage mechanism to use, and how to use the network and services state to provision infrastructure and services, and validate network operations. We'll also discuss various network automation tools, and select the best one for the job that needs to be done.
Design Compute and Storage Infrastructure
You can’t start designing or building a data center infrastructure from the bottom up. Before designing the network you need to know the workloads the network has to support. In this section we’ll talk about:
- Server virtualization and impact of high-density virtualization
- Bare-metal servers, virtual machines and containers
- Traditional storage solutions, distributed storage and software-defined storage
- Hyperconverged solutions
Additional session: Server Virtualization Explained
We’ll focus on various compute virtualization methods, including server virtualization (VMs) and containers. This session will help you understand the underlying assumptions (example: you can’t migrate Docker containers) and networking requirements of individual virtualization solutions.
Design the Network Services and Infrastructure
Finally it’s time to get our hands dirty. We’ll talk about virtual networks needed by application workloads, focus on data center fabrics, design one, and deploy network services on top of it.
This part of the course focuses on high-availability aspects of modern data center designs and deployments:
- How to define high availability
- Dispelling the high availability myths
- Identifying real business needs and their viability
- Deciding what’s feasible to do based on application architecture, and what makes sense to deploy;
Putting It All Together
In the last section of the course we’ll yet again refocus on the big picture: how to combine all the pieces we’ve discussed during the previous weeks into an optimized data center infrastructure.
Some of the topics we’ll cover include:
- Modular pod-based design;
- Inter-pod connectivity (transport core within and across data centers);
- WAN connectivity;
- Integration with public cloud;