flex cookbook

memcache@facebook

Marc Kwiatkowskimemcache tech leadQCon 2010

Monday, April 12, 2010

How big is facebook?

!"" million active users

400M Active

!"" million active users

#""$ #""% #"&"

400M Active

Objects▪ More than 60 million status updates posted each day

▪ 694/s

▪ More than 3 billion photos uploaded to the site each month

▪ 23/s

▪ More than 5 billion pieces of content (web links, news stories, blog posts, notes, photo albums, etc.) shared each week

▪ 8K/s

▪ Average user has 130 friends on the site

▪ 50 Billion friend graph edges

▪ Average user clicks the Like button on 9 pieces of content each month

- Infrastructure▪ Thousands of servers in several data centers in two regions

▪ Web servers

▪ DB servers

▪ Memcache Servers

▪ Other services

The scale of memcache @ facebook▪ Memcache Ops/s

▪ over 400M gets/sec

▪ over 28M sets/sec

▪ over 2T cached items

▪ over 200Tbytes

▪ Network IO

▪ peak rx 530Mpkts/s 60GB/s

▪ peak tx 500Mpkts/s 120GB/s

A typical memcache server’s P.O.V.▪ Network I/O

▪ rx 90Kpkts/s 9.7MB/s

▪ tx 94Kpkts/s 19MB/s

▪ Memcache OPS

▪ 80K gets/s

▪ 2K sets/s

▪ 200M items

All rates are 1 day moving averages

Evolution of facebook’s architecture

•When Mark Zuckerberg and his roommates started Facebook in a Harvard dorm in 2004, the put everyone on one server

•Then as Facebook grew, they could scale like a traditional site by just adding servers

•Even as the site grew beyond Harvard to Stanford, Columbia and thousands of other campuses, each was a separated network that could be served on an isolated set of servers

•But as people connected more between schools connected, the model changed--and the big change came when Facebook opened to everyone in Sept. 2006

•[For globe]: That led to people being connected everywhere around the world--not just on a single college campus.

•[For globe]: This visualization shows accepted friend requests animating from requesting friend to accepting friend

Scaling Facebook: Interconnected data

•On Facebook, the data required to serve your home page or any other page s incredibly interconnected

•Your data can’t sit on one server or cluster of servers because almost every piece of content on Facebook requires information about your network of friends

•And the average user has 130 friends

•As we scale, we have to be able to quickly pull data across all of our servers, wherever it’s stored.

Bob Brian

Bob BrianFelicia

Memcache Rules of the Game▪ GET object from memcache

▪ on miss, query database and SET object to memcache

▪ Update database row and DELETE object in memcache

▪ No derived objects in memcache

▪ Every memcache object maps to persisted data in database

Scaling memcache

Phatty Phatty Multiget

Phatty Phatty Multiget (notes)▪ PHP runtime is single threaded and synchronous

▪ To get good performance for data-parallel operations like retrieving info for all friends, it’s necessary to dispatch memcache get requests in parallel

▪ Initially we just used polling I/O in PHP.

▪ Later we switched to true asynchronous I/O in a PHP C extension

▪ In both case the result was reduced latency through parallelism.

Pools and Threads

PHP Client

sp:12345

sp:12346 sp:12347 cs:12345cs:12346 cs:12347

Different objects have different sizes and access patterns. We began creating memcache pools to segregate different kinds of objects for better cache efficiency and memory utilization.

PHP Client

sp:12345 sp:12346 sp:12347 cs:12345 cs:12346 cs:12347

PHP Client

Pools and Threads (notes)▪ Privacy objects are small but have poor hit rates

▪ User-profiles are large but have good hit rates

▪ We achieve better overall caching by segregating different classes of objects into different pools of memcache servers

▪ Memcache was originally a classic single-threaded unix daemon

▪ This meant we needed to run 4 instances with 1/4 the RAM on each memcache server

▪ 4X the number of connections to each both

▪ 4X the meta-data overhead

▪ We needed a multi-threaded service

Connections and Congestion▪ [animation]

Connections and Congestion (notes)▪ As we added web-servers the connections to each memcache box

▪ Each webserver ran 50-100 PHP processes

▪ Each memcache box has 100K+ TCP connections

▪ UDP could reduce the number of connections

▪ As we added users and features, the number of keys per-multiget increased

▪ Popular people and groups

▪ Platform and FBML

▪ We began to see incast congestion on our ToR switches.

Serialization and Compression▪ We noticed our short profiles weren’t so short

▪ 1K PHP serialized object

▪ fb-serialization

▪ based on thrift wire format

▪ 3X faster

▪ 30% smaller

▪ gzcompress serialized strings

Multiple Datacenters

SC Memcache

SC Web

SC MySQL

SF Web

SF Memcache

SC Memcache

SC Web

SC MySQL

SF Web

SF Memcache

SC Memcache

SC Web

SC MySQL

Memcache ProxyMemcache Proxy

▪Multiple Datacenters (notes)▪ In the early days we had two data-centers

▪ The one we were about to turn off

▪ The one we were about to turn on

▪ Eventually we outgrew a single data-center

▪ Still only one master database tier

▪ Rules of the game require that after an update we need to broadcast deletes to all tiers

▪ The mcproxy era begins

Multiple Regions

SC Memcache

SC Web

SC MySQL

East Coast

VA MySQL

VA Web

VA Memcache

Memcache Proxy

West Coast

Multiple Regions

SF Web

SF Memcache

SC Memcache

SC Web

SC MySQL

East Coast

VA MySQL

VA Web

VA Memcache

Memcache Proxy

West Coast

Multiple Regions

SF Web

SF Memcache

SC Memcache

SC Web

SC MySQL

MySql replication

East Coast

VA MySQL

VA Web

VA Memcache

Memcache Proxy

West Coast

▪ Multiple Regions (notes)

▪ Latency to east coast and European users was/is terrible.

▪ So we deployed a slave DB tier in Ashburn VA

▪ Slave DB tracks syncs with master via MySQL binlog

▪ This introduces a race condition

▪ mcproxy to the rescue again

▪ Add memcache delete pramga to MySQL update and insert ops

▪ Added thread to slave mysqld to dispatch deletes in east coast via mcpro

Replicated Keys

PHP Client PHP ClientPHP Client

Memcache MemcacheMemcache

Replicated Keys

key key key

Memcache MemcacheMemcacheMemcache

Replicated Keys

key key key

Replicated Keys

MemcacheMemcache Memcache

Replicated Keys

key#" key#& key#'

MemcacheMemcache Memcache

Replicated Keys (notes)▪ Viral groups and applications cause hot keys

▪ More gets than a single memcache server can process

▪ (Remember the rules of the game!)

▪ That means more queries than a single DB server can process

▪ That means that group or application is effectively down

▪ Creating key aliases allows us to add server capacity.

▪ Hot keys are published to all web-servers

▪ Each web-server picks an alias for gets

▪ get key:xxx => get key:xxx#N

▪ Each web-server deletes all aliases

Memcache Rules of the Game▪ New Rule

▪ If a key is hot, pick an alias and fetch that for reads

▪ Delete all aliases on updates

Mirrored Pools

General pool with wide fanout

Shard & Shard #

Specialized Replica !

Shard & Shard #

Shard & Shard # Shard ' Shard n

Specialized Replica "

Mirrored Pools (notes)▪ As our memcache tier grows the ratio of keys/packet decreases

▪ 100 keys/1 server = 1 packet

▪ 100 keys/100 server = 100 packets

▪ More network traffic

▪ More memcache server kernel interrupts per request

▪ Confirmed Info - critical account meta-data

▪ Have you confirmed your account?

▪ Are you a minor?

▪ Pulled from large user-profile objects

▪ Since we just need a few bytes of data for many users

Hot Misses▪ [animation]

Hot Misses (notes)▪ Remember the rules of the game

▪ update and delete

▪ miss, query, and set

▪ When the object is very, very popular, that query rate can kill a database server

▪ We need flow control!

Memcache Rules of the Game▪ For hot keys, on miss grab a mutex before issuing db query

▪ memcache-add a per-object mutex

▪ key:xxx => key:xxx#mutex

▪ If add succeeds do the query

▪ If add fails (because mutex already exists) back-off and try again

▪ After set delete mutex

Hot Deletes▪ [hot groups graphics]

Hot Deletes (notes) ▪ We’re not out of the woods yet

▪ Cache mutex doesn’t work for frequently updated objects

▪ like membership lists and walls for viral groups and applications.

▪ Each process that acquires a mutex finds that the object has been deleted again

▪ ...and again

Rules of the Game: Caching Intent▪ Each memcache server is in the perfect position to detect and

mitigate contention

▪ Record misses

▪ Record deletes

▪ Serve stale data

▪ Serve lease-ids

▪ Don’t allow updates without a valid lease id

Next Steps

Shaping Memcache Traffic▪ mcproxy as router

▪ admission control

▪ tunneling inter-datacenter traffic

Cache Hierarchies▪ Warming up Cold Clusters

▪ Proxies for Cacheless Clusters

Big Low Latency Clusters▪ Bigger Clusters are Better

▪ Low Latency is Better

▪ L2.5

▪ UDP

▪ Proxy Facebook Architecture

Worse IS better▪ Richard Gabriel’s famous essay contrasted

▪ ITS and Unix

▪ LISP and C

▪ MIT and New Jersey

http://www.jwz.org/doc/worse-is-better.html

Why Memcache Works▪ Uniform, low latency with partial results is a better user

experience

▪ memcache provides a few robust primitives

▪ key-to-server mapping

▪ parallel I/O

▪ flow-control

▪ traffic shaping

▪ that allow ad hoc solutions to a wide range of scaling issues

We started with simple, obvious improvements.As we grew we deployed less obvious improvements...But they’ve remained pretty simple

flex cookbook

Documents

Transcript of flex cookbook

Flex 14/Flex 14D/Flex 15/Flex 15D UserGuide · 2 Kapitel 1. Übersicht über den Computer Flex 15/Flex 15D Hinweis:Die mit gestrichelten Linien gekennzeichneten Bereiche sind von

K-FLEX CataLoguE K-FLEX

BAĞLANTI HORTUMLARI U-FLEX / V-FLEX / UW-FLEX

Lenovo IdeaPad Flex 2-14/Flex 2-14D Flex 2-15/Flex …Lenovo Flex 2-14/Flex 2-14D/Flex 2-15/Flex 2-15D Hardware Main-tenance Manual 4 † Always look carefully for possible hazards

Lenovo IdeaPad Flex 14/Flex 14D Flex 15/Flex 15D · Lenovo Flex 14/Flex 14D/Flex 15/Flex 15D Hardware Maintenance Manual 2 Follow these rules below to ensure general safety: † Observe

Make Your Own Cookbook App with Cookbook Cafe

Lenovo IdeaPad Flex 14/Flex 14M Flex 15/Flex 15Mcdn.cnetcontent.com/aa/94/aa94f26b-d102-4dcd-a749...Lenovo Flex 14/Flex 14M/Flex 15/Flex 15M Hardware Maintenance Manual 2 Follow these

HP Virtual Connect Ethernet Cookbook: Single and … Virtual Connect Ethernet Cookbook: Single and Multi Enclosure Domain (Stacked) Scenarios ... Flex-10 Scenario ...

Designing Flex and Rigid Flex Applicaitons - …circuits-corp.com/.../03/Designing-Flex-and-Rigid-Flex-Applications... · Designing Flex and Rigid Flex Products for High Reliability

CPAP specifications guide - RespShop specifications guide ... Fixed CPAP Auto CPAP C-Flex C-Flex+ A-Flex CPAP C-Flex C-Flex+ A-Flex Standard Standard Adjustable Automatic Legend …

Lenovo IdeaPad Flex 2-14/Flex 2-14D Flex 2-15/Flex 2-15D€¦ · · 2014-06-12Lenovo IdeaPad Flex 2-14/Flex 2-14D Flex 2-15/Flex 2-15D Hardware ... has hazardous voltages. ... Be

Www.samsung.com in Microsite Cookbook CookBook English Hindi CookBook English Hindi

IP MIGRATION MADE SIMPLE NVT PHYBRIDGE FLEX ......IP MIGRATION MADE SIMPLE FLEX ADAPTERS & EXTENDERS NVT PHYBRIDGE FLEX-Link FLEX-C FLEX-Base DATA SHEET FLEX Adapters & Extenders The

Flex your Database on 12c's Flex ASM Flex Cluster

Lenovo IdeaPad Flex 14/Flex 14D Flex 15/Flex 15Dcdn.billiger.com/dynimg/gnZ0W6UgR4HX90aFpmTDmHt4H9... · Lenovo IdeaPad Flex 14/Flex 14D Flex 15/Flex 15D Hardware Maintenance Manual

Cisco UCS Director Open Automation Cookbook, Release 6 · Cisco UCS Director Open Automation Cookbook, Release 6.5 1. Cisco UCS Director Open Automation Cookbook, ... flex-messaging-common.jar

FGR FLEX / FLEX E

Lenovo IdeaPad Flex 2-14/Flex 2-14D Flex 2-15/Flex 2-15D · 2016-10-19 · Lenovo Flex 2-14/Flex 2-14D/Flex 2-15/Flex 2-15D Hardware Main-tenance Manual 2 Follow these rules below

2008-Flex Air Cookbook Feb2008

Cooking RIA CRUD with Flex and BlazeDS - · PDF fileCooking RIA CRUD with Flex and BlazeDS In today’s cooking class you’ll add to your cookbook a delicious recipe. It’s quick