Google SRE Book: Load Balancing in the Datacenter

Ideas you will meet

Load Balancer

Spread traffic across servers with L4/L7 balancing, health checks and algorithms.

What we already know about this

Our own lessons on these ideas. Read them here before you start, so it makes sense.

• System Design Foundations · 10 minFrom one server to many: scaling and load balancingVertical versus horizontal scaling, why stateless servers matter, and how a load balancer spreads traffic.Read hereClose

  Two ways to grow

  When one server is not enough you have two options.

  Image: Derrick Coetzee · CC0 · Source · Wikimedia Commons: Technician with laptop working on server rack at NERSC

  • Vertical scaling ("scale up"): use a bigger machine with more CPU, memory or disk. It is simple because nothing in the design changes.
  • Horizontal scaling ("scale out"): add more machines and share the work between them. It needs more design effort but has no hard ceiling.

  	Vertical	Horizontal
  Effort	Low	Higher
  Ceiling	Hardware limit	Practically none
  Failure	One machine is a single point of failure	Survives losing a machine
  Cost curve	Gets steep for the largest machines	Roughly linear with commodity machines

  Real systems use both: scale up until it gets expensive or risky, then scale out.

  100%

  Figure SDA-FND-02-1 · Scale up versus scale out · © forcsguy.com

  What a load balancer does

  Once you have several servers, something must decide which one handles each request. That is the load balancer.

  100%

  Figure SDA-FND-02-2 · Load balancer in front of a server pool · © forcsguy.com

  It does three jobs:

  1. Distributes requests so no server is overloaded.
  2. Checks health. It probes each server regularly and stops sending traffic to ones that fail.
  3. Hides the pool. Clients talk to one address, and servers can be added or removed without clients noticing.

  Common strategies

  • Round robin: take turns. Simple and fine when requests are similar.
  • Least connections: send to the server with the fewest active requests. Better when request cost varies.
  • Hashing (for example on user id or IP): the same key goes to the same server. Useful when a server keeps a local cache for that key.
  • Weighted: bigger machines get a bigger share.

  Load balancers work at different layers. A layer 4 balancer routes by IP and port and is very fast. A layer 7 balancer reads the HTTP request, so it can route by URL path or header and terminate TLS.

  A security benefit comes with this layout: only the balancer needs a public address. The servers behind it use private addresses, reachable from inside the network but not from the internet, so clients cannot bypass the balancer or attack a server directly. If one server goes offline, the balancer sends its traffic to the others and a replacement can join the pool.

  Before: one server

  100%

  After: load balancer and several servers

  100%

  What changed

  • Traffic is shared across servers
  • One server failing no longer takes the site down
  • You can add capacity without downtime

  Stateless servers make scaling easy

  A server is stateless when it keeps nothing about a user between requests. Everything it needs comes from the request or from a shared store.

  100%

  Figure SDA-FND-02-3 · Stateful versus stateless servers · © forcsguy.com

  With stateless servers any request can go to any server, so you can add servers at will, replace a failed one without losing data, and run auto-scaling. The state does not disappear. It moves to a database, a cache or a session store, which you then have to scale and protect, which is what the next lessons are about.

  Putting it together

  A typical first scalable layout:

  1. Users reach a load balancer.
  2. It spreads requests over a pool of stateless application servers.
  3. Servers read and write a shared database (and later a cache).
  4. Static files such as images and scripts are served from a CDN so the servers do not carry that traffic (lesson 9).

  This design handles a surprisingly large amount of traffic. The first thing to give way is usually the database, which is where we go next.

  A worked example: how many servers do you need?

  Capacity planning is arithmetic you can do on a napkin. Say your product has 5 million daily users and each makes about 20 requests a day.

  1. Requests per day: 5,000,000 x 20 = 100 million.
  2. Per second on average: 100,000,000 / 86,400 = about 1,160 requests per second.
  3. Traffic is not flat. A common rule of thumb is that the busiest second is 2 to 3 times the average, so plan for about 3,500 per second.
  4. One server handles about 400 requests per second flat out, but you never want servers at 100%: queues build and latency explodes near full load. Plan for roughly 70% utilisation, so 280 requests per second each.
  5. 3,500 / 280 = 12.5, so 13 servers carry the peak.
  6. Add spare capacity so you survive failures: with 13 servers, losing 2 leaves 11, which carry only 3,080 requests per second. Plan for N+2: 15 servers.

  100%

  Figure SDA-FND-02-4 · From users to a server count · © forcsguy.com

  The same method works for any tier. Only the "requests per server" number changes, and you should measure it with a load test rather than guess.

  Choosing a balancing method with real numbers

  Round robin is fair in the number of requests, not in the work. Imagine three servers and a stream where most requests take 10 ms but every fifth takes 1 second (a report export, say). Round robin gives each server the same count, but by chance one server collects several slow requests at once, its queue grows, and its users wait while the other two are nearly idle.

  Least connections notices: the server with the most open requests gets fewer new ones, so the slow requests spread out as they arrive. It costs the balancer a little bookkeeping and helps most when request costs vary a lot. Use plain round robin when requests are similar and servers identical; use least connections, or "power of two choices" (pick two servers at random and send to the less busy one), when they are not. For caches and stateful workloads, consistent hashing keeps the same key on the same server so its memory stays useful.

  Health checks and draining

  A balancer is only as good as its knowledge of which servers are alive.

  • Active health checks call a lightweight endpoint (for example /health) every few seconds. After two or three failures in a row, the server leaves rotation; after a few successes it returns. Make the check reflect real readiness: a server that is up but cannot reach its database should fail the check.
  • Passive checks watch real traffic: too many errors or timeouts from one server and it is ejected.
  • Connection draining matters at deploys. Before stopping a server, tell the balancer to send it no new requests, let in-flight requests finish (typically 30 to 60 seconds), then stop it. Without draining, every deploy drops a few user requests.

  100%

  Figure SDA-FND-02-5 · Rolling out a new version without dropping requests · © forcsguy.com

  Mistakes people make

  • Treating the balancer as magic. It is a server too and can be a single point of failure. Run at least two, or use a managed one.
  • Adding servers behind a single database. App servers scale out easily; if the database is the bottleneck, more app servers only make it fail faster.
  • Keeping state on the server. Sessions, uploads and caches held in one server's memory break the moment a request lands elsewhere. Move them to a shared store.
  • Sizing for the average. The system fails at peak, which is when it matters.
  • Never testing failure. Kill a server during a load test. If users notice, you have found a gap before they did.

  Try it yourself

  Open the System Builder challenge "One server is not enough". Run the test on one server and watch it fail, add a load balancer without adding servers, then add servers and see where the numbers settle. Or skip the challenge and open a load-balanced system on the free canvas, then change the wiring yourself.

  More on this site

  Related reading

  • Maglev: A Fast and Reliable Software Network Load Balancer (paper)Eisenbud et al. · GooglePapers
  • NGINX HTTP Load BalancingNGINX · F5 / NGINXDocumentation
  • Google SRE Book: Load Balancing in the DatacenterGoogle SRE team · GoogleBlogs
  • Scaling Load Balancing Infra to Support 1.3 Billion Users at Facebook - Patrick Shuff, Production Engineer at FacebookFacebookYouTube

  Show 6 moreShow fewer

  • Load Balancing infrastructure to support more than 1.3 billion users at FacebookFacebook engineers · FacebookCase Studies
  • Implementing health checksAmazon Builders' Library · Amazon Web ServicesBlogs
  • Awesome System Design ResourcesAshish Pratap Singh · ashishps1GitHub
  • Load Balancing at GithubGithub engineers · GithubCase Studies
  • Google Cloud Architecture FrameworkGoogle · Google CloudDocumentation
  • HAProxyHAProxy Technologies · HAProxyDocumentation

  Open this lesson with its quiz and exercise

• Advanced Distributed Systems · 12 minLoad balancing: layer 4, layer 7 and gatewaysWhat a load balancer can see and do at each layer, the algorithms behind it, and how API gateways add routing, aggregation and policy.Read hereClose

  What a load balancer sees

  A load balancer spreads incoming work across servers, removes failed servers and makes scaling invisible to clients. What it can do depends on which layer it works at.

  	Layer 4 (transport)	Layer 7 (application)
  Sees	IP addresses, ports, TCP/UDP connections	HTTP method, URL, headers, cookies, body
  Decides by	Connection details	Request content
  TLS	Passes through, or terminates in some products	Usually terminates TLS
  Speed and cost	Very fast, low overhead	More work per request
  Typical features	Spread connections, health checks	Path routing, header rules, sticky sessions, rewrites, caching, WAF rules
  Examples	Network load balancers, LVS, HAProxy in TCP mode	NGINX, Envoy, HAProxy HTTP mode, cloud application balancers

  100%

  Figure SDA-ADS-05-1 · Layers of balancing in front of an app · © forcsguy.com

  Many designs use both: an L4 tier for raw scale and an L7 tier for smart routing.

  Before: layer 4: ports and addresses

  100%

  After: layer 7: reads the request

  100%

  What changed

  • Layer 7 can route by path, header or cookie
  • Layer 7 can terminate TLS and retry safely
  • Layer 4 is simpler and faster, with less to configure

  Choosing a target: algorithms

  • Round robin / weighted round robin: simple rotation; weights favour bigger machines.
  • Least connections (and weighted): send to the server with fewest active requests; good when requests vary in length.
  • Least response time: combine active load with measured latency.
  • IP hash / consistent hashing: the same client or key goes to the same server, useful for caches and sticky state (see consistent hashing).
  • Resource-based: servers report load and the balancer follows.

  Health checks (active probes and passive error tracking) remove unhealthy servers; connection draining lets in-flight requests finish before a server is removed during a deploy.

  Global and DNS balancing

  Across regions, DNS-based or global server load balancing directs users to the nearest or healthiest region (see the DNS and CDN lessons). DNS answers are cached, so failover takes as long as the TTL; anycast and global balancers react faster.

  API gateway: policy at the edge

  An API gateway is an L7 component specialised for APIs. Beyond routing it commonly provides:

  • Authentication and authorisation (validate tokens, apply scopes),
  • Rate limiting and quotas (see the rate-limiter lesson),
  • Request and response transformation, versioning, and protocol translation,
  • Aggregation: one client call fans out to several services and merges results, saving round trips for mobile clients,
  • Caching, logging, metrics and tracing in one place.

  100%

  Figure SDA-ADS-05-2 · Gateway routing and aggregation · © forcsguy.com

  Risks: the gateway is a single entry point (run several instances, keep it stateless) and can become a bottleneck or a place where business logic piles up. Keep it to cross-cutting concerns. A variation is a backend for frontend: one gateway per client type (web, mobile) tailored to its needs.

  Inside the cluster

  Between services, balancing is done by client-side libraries or sidecar proxies that know the healthy instances from service discovery, so no central balancer is needed for east-west traffic. See the microservices lessons for service mesh ideas.

  Failure habits

  • Run at least two of every balancer and gateway instance across zones.
  • Keep timeouts and retry budgets (see the resilience lesson) so the balancer does not amplify an outage.
  • Use slow start for new instances so they warm up before taking a full share.

  Real-world reading

  Related reading

  • Maglev: A Fast and Reliable Software Network Load Balancer (paper)Eisenbud et al. · GooglePapers
  • NGINX HTTP Load BalancingNGINX · F5 / NGINXDocumentation
  • Scaling Your API with Rate LimitersPaul Tarjan · StripeBlogs
  • Active-Active for Multi-Regional ResiliencyNetflix Engineering · NetflixCase Studies

  Show 6 moreShow fewer

  • ByteByteGo: System Design Video SeriesAlex Xu · ByteByteGoYouTube
  • Rate Limiting with NGINXNGINX · F5 / NGINXBlogs
  • Implementing health checksAmazon Builders' Library · Amazon Web ServicesBlogs
  • Google SRE Book: Load Balancing in the DatacenterGoogle SRE team · GoogleBlogs
  • Traffic Steering using Rum DNS at LinkedInLinkedIn engineers · LinkedInCase Studies
  • Google SRE BookBetsy Beyer et al. · GoogleCourses

  Open this lesson with its quiz and exercise

• Advanced Distributed Systems · 12 minCloud architecture principlesRegions, zones, identity, autoscaling and a checklist of design pillars for reviewing a cloud architecture, from reliability to cost.Read hereClose

  The building blocks

  • Region: a geographic area with several data-centre groups. Choose by user latency, data-residency rules, service availability and price.
  • Availability zone (AZ): an isolated failure domain inside a region (own power, cooling and network), connected to others by fast links. Spread an app across at least two zones.
  • Edge locations: points of presence for CDN and DNS close to users.
  • Identity and access management (IAM): who or what may do which action on which resource. Users, groups, roles for services, and policies that allow or deny actions.
  • Compute choices: virtual machines, containers (managed Kubernetes or container services), and functions (serverless). See the cloud models and serverless lessons.
  • Storage choices: block, file and object (see the storage lesson), managed databases, caches and queues.
  • Networking: private networks, subnets, routing, security groups and gateways; keep databases in private subnets and expose only load balancers.

  100%

  Figure SDA-ADS-08-1 · A typical multi-zone web architecture · © forcsguy.com

  Design pillars (a review checklist)

  Cloud providers publish similar frameworks. In plain words, ask these questions of any design:

  1. Operational excellence: can we deploy safely and often, see what the system is doing and learn from incidents? (Automation, infrastructure as code, runbooks, dashboards.)
  2. Security: who can do what? Is data encrypted in transit and at rest, are secrets managed, are permissions minimal, is activity logged?
  3. Reliability: what fails, and what happens then? (Redundancy across zones, health checks, backups and tested restores, graceful degradation, recovery time and data loss targets.)
  4. Performance efficiency: are we using the right resource types and scaling to demand? (Caching, right-sizing, choosing the right data store, measuring.)
  5. Cost optimisation: do we know where money goes and can we cut waste? (Tags and budgets, right-sizing, autoscaling, reserved or committed use for steady load, lifecycle rules for storage.)
  6. Sustainability: are we avoiding idle capacity and unneeded data and compute?

  Treat the pillars as trade-offs, not a score: more reliability usually costs more; more security can slow delivery. Write down the choice and the reason.

  Practical rules

  • Design for failure: assume instances, zones and dependencies fail. Prefer stateless services and managed, replicated state.
  • Disaster recovery is chosen by two numbers: how much data you can lose (RPO) and how long you can be down (RTO). Strategies range from backup and restore (cheap, slow) through pilot light and warm standby to active-active across regions (costly, fast).
  • Autoscale carefully: scale on a metric tied to load, set minimum and maximum, allow warm-up, and test scale-in.
  • Least privilege and no long-lived keys: use roles and short-lived credentials; rotate and audit.
  • Everything as code so environments are reproducible and reviewable.
  • Observe: metrics, logs and traces from day one, with alerts on user-facing symptoms.
  • Know the shared responsibility split: the provider secures the cloud; you secure what you put in it.
  • Avoid unnecessary lock-in where it is cheap, but do not contort a design to be portable at the cost of using good managed services.

  Real-world reading

  Related reading

  • Load Balancing infrastructure to support more than 1.3 billion users at FacebookFacebook engineers · FacebookCase Studies
  • Scaling Load Balancing Infra to Support 1.3 Billion Users at Facebook - Patrick Shuff, Production Engineer at FacebookFacebookYouTube
  • Static stability using Availability ZonesAmazon Builders' Library · Amazon Web ServicesBlogs
  • MySQL High Availability at GitHubGitHub Engineering · GitHubCase Studies

  Show 6 moreShow fewer

  • Google SRE Book: Load Balancing in the DatacenterGoogle SRE team · GoogleBlogs
  • GitHub Availability Report: June 2023GitHub Engineering · GitHubCase Studies
  • Scaling Backend at Uber - Matt Ranney, Chief Systems Architect at UberUberYouTube
  • Performance Optimization for the Greater China Region at Salesforce - Jeff Cheng, Enterprise Architect at SalesforceSalesforce engineers · SalesforceCase Studies
  • Scaling Data Infrastructure at Spotify - Matti (Lepistö) Pehrs, SpotifySpotify - MattiYouTube
  • Scaling Infrastructure at Etsy - Bethany Macri, Engineering Manager at EtsyEtsyYouTube

  Open this lesson with its quiz and exercise