Effective Data Center Design Techniques: Data Center Topologies and Control Planes
Data Center Topologies and Control Planes
Data centers are the foundation of the cloud, whether private, public, on the edge, or in the center of the network. This training will focus on topologies and control planes, including scale, performance, and centralization. This training is important for network designers and operators who want to understand the elements of data center design that apply across all hardware and software types.
What you'll learn-and how you can apply it
- You will learn the fundamental concepts of spine and leaf topologies, including the Clos, the butterfly, and the pod and fabric designs.
- You will learn about the design of path vector and link state protocols on a densely meshed topology, including how to control convergence time.
- You will learn how to think about the performance factors of a fabric, and drivers for traffic engineering.
- You will be able to apply these lessons to real world data center design across a wide array of sizes and purposes.
This training course is for you because...
This course is for you if you want to learn:
- data center spine and leaf topologies, including variants, scaling, and theory
- how to optimally use BGP as a data center control plane
- link state protocols for data center fabrics
- performance metrics for data center fabrics
- telemetry challenges for data center fabrics
- centralized control planes and southbound interfaces used in data center fabrics
- tradeoffs around centralized and decentralized control planes in data center fabrics
- Basic understanding of the principles of routing
- A basic idea of the various protocols used in an IPv4/IPv6 network
About your instructor
Russ White began working with computers in the mid-1980's, and computer networks in 1990. He has experience in designing, deploying, breaking, and troubleshooting large scale networks, and is a strong communicator from the white board to the board room. Across that time, he has co-authored more than forty software patents, participated in the development of several Internet standards, helped develop the CCDE and the CCAr, and worked in Internet governance with the Internet Society. Russ has a background covering a broad spectrum of topics, including radio frequency engineering and graphic design, and is an active student of philosophy and culture.
Russ is a co-host at the Network Collective, serves on the Routing Area Directorate at the IETF, co-chairs the BABEL working group, serves on the Technical Services Council/as a maintainer on the open source FR Routing project, and serves on the Linux Foundation (Networking) board. His most recent works are Computer Networking Problems and Solutions, The Art of Network Architecture, Navigating Network Complexity, and the Intermediate System to Intermediate System LiveLesson.
MSIT Capella University, MACM Shepherds Theological Seminary, PhD (in progress) Southeastern Baptist Theological Seminary CCIE #2635, CCDE 2007::1, CCAr
The timeframes are only estimates and may vary according to how the class is progressing
Segment 1: Data Center Topologies
Length: 50 minutes
- Scale out versus scale up
- Fabrics versus other network topologies
- Spine and leaf topologies: Clos and butterfly
- Oversubscription and performance parameters
- Connection into the spine and leaf
- Chassis versus 1ru
10 Minute Break
Segment 2: Distributed Data Center Control Planes
Length: 50 minutes
- BGP in the data center fabric: methods to control hunt
- IS-IS in the data center fabric
- Modified IS-IS (openfabric)
- Centralized flooding solutions
- Other solutions to link state scale in a dense topology
Segment 3: Centralized Control Plane Tradeoffs and Southbound Interfaces
Length: 50 minutes
- Tradeoffs in centralilzed control planes
- Southbound interfaces: PCEP, YANG (I2RS), gRPC, and others
Q&A (10 mins)