SRE Load Balancing Techniques: Data Center Load Balancing

IN THIS COURSE

• 1.  Course Overview
  1m 27s
  FREE ACCESS
• 2.  Data Center Load Balancing
  5m 41s
  In this video, you'll learn more about Data Center Load Balancing. You'll learn that the primary goal of any data center is to ingest and serve data as efficiently as possible. The video outlines three different types of Datacenter Load Balancing, including Plain Old Load Balancing or POLB, Network Load Balancing, and HTTP Load Balancing. Then, you'll learn how these load balancing algorithms can be used to improve performance for applications. FREE ACCESS
• 3.  Load Balancing Policies
  5m 46s
  Load balancing policies are used to distribute traffic among backend servers in a pool. There are two main types of load balancers, TCP load balancers and HTTP load balancers, and each has its own policies. Load balancer policies may include weighting algorithms, round-robin, least connections, and IP hash. FREE ACCESS
• 4.  Load Balancing Flow Control
  5m 58s
  In this video, you will learn how to handle unhealthy backend servers in a load balancing situation. You will learn about Google's Lame Duck approach to backend server health and how it works behind the scenes. You will also discover the benefits of a simple client-side load balancing strategy and discuss its drawbacks. FREE ACCESS
• 5.  Load Balancing Tips and Tricks
  4m 38s
  This session will discuss various tips and tricks for load balancing. First, pay attention to the type of sessions you'll be load balancing. If the sessions don't need to be processed by the same server instance, then they can be sticky or independent. However, if processing at the back-end requires knowledge of multiple requests or request history, then session persistence may be required. Additionally, when analyzing your peak load periods, it's important to know when these periods are occurring. FREE ACCESS
• 6.  Connection Pool Limits
  4m 31s
  There are two main methods used to limit the number of connections a client can make to a server: subsetting and sticky sessions. With subsetting, clients are limited to connecting to a certain number of backend servers, which frees up resources on the server. The downside is that this can be time-consuming, as each connection needs to be created and torn down. Sticky sessions don't have this limitation, and allow clients to connect to any number of backend servers. FREE ACCESS
• 7.  Load Balancing Components
  7m 3s
  In this video, we will discuss the various components that make up a load balancing architecture. We will discuss back end services, forwarding rules, health checks, and URL maps. We will also discuss how IPv6 is supported in Google Cloud and what role certificates and policies play in SSL requests. FREE ACCESS
• 8.  Internal HTTPS Load Balancing
  5m 12s
  In this video, you'll learn more about how to outline how loads can be balanced using Internal HTTPS Load Balancing. These load balancers are proxy-based, which means connections are terminated at the load balancer and new connections are made between the load balancer and backend server fronts. Being of the HTTPS protocol, these load balancers work at the Layer 7 or application layer. This means they have access to all aspects of an HTTPS request to make load balancing decisions. FREE ACCESS
• 9.  External HTTPS Load Balancing
  4m 35s
  In this video, you'll learn more about how External HTTPS Load Balancing handles web traffic. An External HTTPS Load Balancer ingests traffic from external web clients and then distributes the load among several web frontend instances that are in multiple regions. This is why when you open your web browser and go to a cloud-based website, you don't have to determine which region you should access. The load balancer does it for you. FREE ACCESS
• 10.  Internal TCP/UDP Load Balancing
  4m 55s
  In this video, you'll learn about Internal TCP/UDP Load Balancing. You'll learn that an internal TCP/UDP load balancer is meant to balance traffic internal to a region of a cloud system. The primary difference between the TCP/UDP flavor and the HTTPS flavor of internal load balancers is that TCP/UDP works at Layer 4, or the transport layer, while HTTPS works at Layer 7, or the application layer. FREE ACCESS
• 11.  External TCP/UDP Load Balancing
  4m 2s
  In this video, you will learn how external TCP/UDP load balancing works. A network load balancer is a type of load balancer that directs both TCP and UDP traffic. Google Cloud's implementation of an external TCP/UDP load balancer is called a regional load balancer. Regional load balancers are limited to one region and one virtual private cloud or VPC. FREE ACCESS
• 12.  SSL Proxy Load Balancing
  4m 7s
  SSL Proxy Load Balancing is a technology that allows for the load balancing of encrypted traffic. It is designed to improve the performance of non-HTTP traffic, and can be used with both IPv4 and IPv6. Google offers several benefits with their SSL proxy load balancer, such as automatic security patches and intelligent routing. FREE ACCESS
• 13.  TCP Proxy Load Balancing
  3m 51s
  In this video, we will cover how loads can be balanced using TCP Proxy Load Balancing. We will look at how Google implements this load balancer and some of its benefits. FREE ACCESS
• 14.  Course Summary
  1m 18s
  FREE ACCESS