Memcached

  • 3,009 views
Uploaded on

 

More in: Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
No Downloads

Views

Total Views
3,009
On Slideshare
0
From Embeds
0
Number of Embeds
0

Actions

Shares
Downloads
90
Comments
0
Likes
4

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. Memcached http://download.tangent.org/talks/Memcached%20Study.pdf Thursday, April 23, 2009
  • 2. memcached is a high-performance, distributed memory object caching system, generic in nature, but intended for use in speeding up dynamic web applications by alleviating database load. Thursday, April 23, 2009
  • 3. Who? • Facebook • Yahoo • Amazon • LiveJournal • Mixi • ... Thursday, April 23, 2009
  • 4. Why? (aka why would I...)‫‏‬ Thursday, April 23, 2009
  • 5. Thursday, April 23, 2009
  • 6. LiveJournal • Origin of memcached • 30G of cache. Terabytes of data • Writes to DB based on reads from the DB, not cache Thursday, April 23, 2009
  • 7. Mixi • Memcached on dedicated servers • 300 servers in production (reuse old MySQL Servers) • 4/8 gigs of Memory • 15,000 qps / 400Mbps throughput Thursday, April 23, 2009
  • 8. Thursday, April 23, 2009
  • 9. Patrick Lenz Eins.de http://poocs.net Thursday, April 23, 2009
  • 10. Grazr Thursday, April 23, 2009
  • 11. 100+ Nodes 2gigs a Node Processing Incoming Data Memcached Thursday, April 23, 2009
  • 12. How Thursday, April 23, 2009
  • 13. Server • Slab Allocator • Libevent based • Simple Protocol (no xml) • Server has Internal Hash Table • Servers know nothing about each other Thursday, April 23, 2009
  • 14. Clients • Client hashes Key to Server List (distribution)‫‏‬ • Serializes the Object (server just likes byte arrays)‫‏‬ • Compresses data Thursday, April 23, 2009
  • 15. Consistent Hash Thursday, April 23, 2009
  • 16. So what should I ask? • How do I dump data? • How is it redundant? * • How does it handle failover?* • How does it authenticate? Thursday, April 23, 2009
  • 17. Details on the Server? • set/get/replace/add • append/prepend • increment/decrement • cas (compare and swap atomic!)‫‏‬ • stats (detail)‫‏‬ Thursday, April 23, 2009
  • 18. Platforms • FreeBSD and Linux are top tier • Windows exists • Solaris (as of 1.2.5)‫‏‬ • OSX Good Support Thursday, April 23, 2009
  • 19. Examples Thursday, April 23, 2009
  • 20. Ruby Thursday, April 23, 2009
  • 21.  # Set up client object  require 'memcache'  servers = ['127.0.0.1:43042', '127.0.0.1:43043']  CACHE = MemCache.new(servers, :namespace => 'my_app') Thursday, April 23, 2009
  • 22. # Get; fall through to Rails' MySQL load if missing  key = "recent_posts"  # Try to get; returns nil on failure  posts = CACHE.get(key)  unless posts    # Load from DB    posts = Post.find(:all, :limit => 10, :order => 'id DESC')    # Cache the new value, which is automatically serialized    CACHE.set(key), posts, 60  end Thursday, April 23, 2009
  • 23. # Ghetto locking implementation for memcache-client  # from http://fauna.rubyforge.org/svn/interlock/trunk/lib/interlock/lock.rb  def lock(key, lock_expiry = 30, retries = 5)    retries.times do |count|      # Try to acquire the lock      response = CACHE.add("lock:#{key}", "Locked by #{Process.pid}", lock_expiry)      if response == "STOREDrn"        # We got it        begin          # Yield the current value to the closure          value = yield(CACHE.get(key))          # Set the new value returned from the closure CACHE.set(key, value)          # We're done ('ensure' block will still run)          return value        ensure          # Release the lock          CACHE.delete("lock:#{key}")        end      else        # Exponentially back off requests if the lock can't be acquired        sleep((2**count) / 2.0)      end    end    # We waited and waited but our turn never came    raise MemCacheError, "Couldn't acquire lock for #{key}"  end Thursday, April 23, 2009
  • 24. PHP Thursday, April 23, 2009
  • 25. /** * Initalize Memcache object and add local server 127.0.0.1 using port 11211 */ $cache = new Memcache; $cache->addServer("127.0.0.1",11211); Thursday, April 23, 2009
  • 26. /** Look in cache, if not found, set object (misses null) **/ if (!$user = $cache->get($user_key)) { /** * Set object with expire of 1 hour and no compression */ $cache->set($user_key,$value,NULL, $expire); $user = $value; } Thursday, April 23, 2009
  • 27. /* Get user counter value (does not handle add fail) */ if (!$counter = $cache->get($counter_key)) { /* No counter, set to 1 */ $cache->add($counter_key,1); } else { /* Increment by 1 */ $cache->increment($counter_key); } Thursday, April 23, 2009
  • 28. /* Print out stats from memcached server */ print_r($cache->getStats()); /* Print out extended stats from memcached server */ print_r($cache->getExtendedStats()); /* Flush memcached (bad idea in production)*/ var_dump($cache->flush()); Thursday, April 23, 2009
  • 29. C Thursday, April 23, 2009
  • 30. libmemcached • C/C++ (many language wrappers) • Multiget support • Async/Sync Modes (including buffered) • Supports Binary Protocol (TCP/UDP) • Read Through Cache Support Thursday, April 23, 2009
  • 31. memcached_st *memc; memcached_return rc; memc= memcached_create(NULL); ...do stuff... memcached_free(memc); Thursday, April 23, 2009
  • 32. memcached_server_st *servers; memcached_st *memc= memcached_create(NULL); char servername[]= "0.example.com"; servers= memcached_server_list_append(NULL, servername, 400, &rc); for (x= 0; x < 20; x++) { char buffer[SMALL_STRING_LEN]; snprintf(buffer, SMALL_STRING_LEN, "%u.example.com", 400+x); servers= memcached_server_list_append(servers, buffer, 401, &rc); } rc= memcached_server_push(memc, servers); memcached_server_free(servers); memcached_free(memc); Thursday, April 23, 2009
  • 33. char *key= "foo"; char *value; size_t value_length= 8191; unsigned int x; value = (char*)malloc(value_length); for (x= 0; x < value_length; x++) value[x] = (char) (x % 127); for (x= 0; x < 1; x++) { rc= memcached_set(memc, key, strlen(key), value, value_length, (time_t)0, (uint32_t)0); assert(rc == MEMCACHED_SUCCESS); } free(value); Thursday, April 23, 2009
  • 34. memcached_return rc; char *keys[]= {"fudge", "son", "food"}; size_t key_length[]= {5, 3, 4}; uint32_t flags; char return_key[MEMCACHED_MAX_KEY]; size_t return_key_length; char *return_value; size_t return_value_length; rc= memcached_mget(memc, keys, key_length, 3); while ((return_value= memcached_fetch(memc, return_key, &return_key_length, &return_value_length, &flags, &rc))) { free(return_value); } Thursday, April 23, 2009
  • 35. Memcached foo; char *value_set= "Data for server A"; char *master_key_a= "server-a"; char *master_key_b= "server-b"; char *key= "xyz"; char *value; size_t value_length; foo.set_by_key(master_key_a, key, value_set, strlen(value_set)); value= foo.get_by_key(master_key_a, key, &value_length); assert((memcmp(value, value_set, value_length) == 0)); value= foo.get_by_key(master_key_b, key, &value_length); assert((memcmp(value, value_set, value_length) == 0)); return 0; Thursday, April 23, 2009
  • 36. Application Integration Thursday, April 23, 2009
  • 37. lighttpd/mod_memcache • Cache files from disk • Specify mime/types • Create Expire Times Thursday, April 23, 2009
  • 38. Apache (mod_memcached)‫‏‬ • CAS operations exposed • GET/PUT/DELETE operations • Still Alpha • (pandoraport.com!) Thursday, April 23, 2009
  • 39. NGINX • Support variable Time outs • Based on Perl • http://wiki.codemongers.com/ NginxHttpMemcachedModule • http://www.igvita.com/2008/02/11/nginx- and-memcached-a-400-boost/ Thursday, April 23, 2009
  • 40. Memcached Functions for MySQL (and Drizzle) Thursday, April 23, 2009
  • 41. Overview... • Uses UDF API and libmemcached • Manage memcached Cluster via SQL • Read through Cache • Write through Cache Thursday, April 23, 2009
  • 42. Thursday, April 23, 2009
  • 43. Installation • CREATE FUNCTION memc_servers_add RETURNS INT SONAME "libmemcached_functions_mysql.so"; • CREATE FUNCTION memc_set RETURNS INT SONAME "libmemcached_functions_mysql.so"; • CREATE FUNCTION memc_get RETURNS STRING SONAME "libmemcached_functions_mysql.so"; Thursday, April 23, 2009
  • 44. Functions Available • memc_servers_set(); • memc_servers_behavior_set() • memc_set() • memc_get() • memc_append() • memc_prepend() Thursday, April 23, 2009
  • 45. Setting via SELECT select id, url, memc_set(concat('feeds', md5(url)), url) from feeds; select memc_get(concat('feeds', md5(url))) from feeds; +-------------------------------------------------------+ | memc_get(concat('feeds', md5(url))) | +-------------------------------------------------------+ | http://feeds.feedburner.com/littlegreenfootballs/Ilds | | http://del.icio.us/rss/jacomien | | http://del.icio.us/rss/tags/rmj20 | | http://www.jihadwatch.org/index.rdf | | http://www.connotea.org/rss/user/rmj20 | | http://devblog.grazr.com/?feed=rss2x | | http://feeds.feedburner.com/littlegreenfootballs/kyeH | +-------------------------------------------------------+ Thursday, April 23, 2009
  • 46. Trigger DROP TRIGGER IF EXISTS feed_insert; CREATE TRIGGER feed_insert BEFORE INSERT ON feeds FOR EACH ROW BEGIN SET @mm= memc_set(concat('feeds:',md5(NEW.url)), NEW.url);END | insert into feeds (url) values ('http://grazr.com/feedlist.xml'); select memc_get(concat('feeds:', md5('http://grazr.com/feedlist.xml'))); +------------------------------------------------------------------+| memc_get(concat('feeds:', md5('http://grazr.com/feedlist.xml'))) | +------------------------------------------------------------------+| http://grazr.com/feedlist.xml | +------------------------------------------------------------------+ Thursday, April 23, 2009
  • 47. Memcached Replication via MySQL INSERT INTO table_a VALUES (“key”, “value”, “values”); INSERT INTO blackhole_table VALUES (memc_delete(“key”)); Thursday, April 23, 2009
  • 48. Implementation Thursday, April 23, 2009
  • 49. Limits • Key Size (250 bytes) (1.2 <)‫‏‬ • Data Size (under 1 megabyte)‫‏‬ • 32bit/64bit (maximum size of the process)‫‏‬ • Maxbytes (limits item cache, not everything!)‫‏‬ Thursday, April 23, 2009
  • 50. LRU • Least recently accessed items are up for eviction • One LRU exists per “slab class” • LRU evictions don't need to be common • Can be common if you set expiration to 0 • Monitor evictions via 'stats items' command Thursday, April 23, 2009
  • 51. Threads • No single-threading in versions past 1.2 • Great for large instances (16G+) • Also great for large multiget requests • Improvements in 1.3+ • Don't set too many threads • One per CPU • No more than 8 total for 1.2 Thursday, April 23, 2009
  • 52. Slab Allocator • Memory permanently allocated from OS • Classes created by chunk size • Cannot (presently) reassign slab pages • Carefully use 'stats sizes' to test efficiency Thursday, April 23, 2009
  • 53. Binary Protocol 24 HEADER Required bytes EXTRA FIELD (Command Specific) Sizes are defined in KEY As needed the header VALUE * Values must be in network byte order. Thursday, April 23, 2009
  • 54. Binary Protocol MAGIC Opcode Key Length (1 byte) (1 byte) (2 bytes) Extra Length Data Type Reserved (1 byte) (1 byte) (2 bytes) Total Body Length (4 bytes) Opaque (4 bytes) CAS (Compare and Swap) (8 bytes) Thursday, April 23, 2009
  • 55. Response Header MAGIC Opcode Key Length (1 byte) (1 byte) (2 bytes) Extra Length Data Type Status (1 byte) (1 byte) (2 bytes) Total Body Length (4 bytes) Opaque (4 bytes) CAS (Compare and Swap) (8 bytes) Thursday, April 23, 2009
  • 56. Tools Thursday, April 23, 2009
  • 57. Protocol • Telnet’able (see doc/protocol.txt)‫‏‬ • Store commands are binary (no escaping)‫‏‬ • WireShark support for Binary Protocol Thursday, April 23, 2009
  • 58. Thursday, April 23, 2009
  • 59. memcached-tool • Example minimal monitoring tool • Shows details of server stats, slabs • memcached-tool 10.0.0.1 display • memcached-tool 10.0.0.1 stats Thursday, April 23, 2009
  • 60. libmemcached Tools • memcp • memrm • memstat • memslap (hahahhaha)‫‏‬ Thursday, April 23, 2009
  • 61. MRTG • Plenty of interfaces monitored • Old technology, been around forever Thursday, April 23, 2009
  • 62. Thursday, April 23, 2009
  • 63. ...and others • Cacti • http://dealnews.com/developers/cacti/ memcached.html • Ganglia • http://ganglia.wiki.sourceforge.net/ Thursday, April 23, 2009
  • 64. Current Connections Thursday, April 23, 2009
  • 65. Requests Per Second Thursday, April 23, 2009
  • 66. DTrace • User static DTrace probes to probe application logic • Hot keys • hash efficiency • lock contention Thursday, April 23, 2009
  • 67. Tuning Thursday, April 23, 2009
  • 68. Connections • Don't fear setting -c (max conns) too high! • Watch 'listen_disabled_num' (1.2.8+) • Audit any firewalls on the system (can run out of firewall states!) • Watch for connection overload. Probably an application bug Thursday, April 23, 2009
  • 69. Memory • Don't set -m (max bytes) too high • -m only controls memory for stored values • Other memory is used for connection handling, LRU tracking, the hash table, etc • Especially if running under 32bit (-m 2048 bad!) • Careful of -k. Could force server to swap • Monitor server well! If you are swapping, memcached will be slow Thursday, April 23, 2009
  • 70. Things to remember! • Memcached is great for scaling, but abuse will slow down your page rendering time • Fetching 10,000 separate keys to build a page will not be fast • Use persistent connections if possible • Don't fetch more than you need. Takes time to serialize/deserialize 100k+ of data Thursday, April 23, 2009
  • 71. Shrink data! • Especially if you use multiget, try to use shorter keys • “fooooooooooooooooooooooo-$MD5” will take more data packets than “foo1” • Try using smaller keys anyway, saves memory if your data is small too Thursday, April 23, 2009
  • 72. Future Solutions! • Evaluate the binary protocol in 1.3 • New tricks will be possible, such as smashing many commands into fewer packets • 'noreply' commands. Don't wait for the response if you don't have to • ('noreply' is in 1.2, but not recommended for use) Thursday, April 23, 2009
  • 73. 1.2.8 • Bugfixes • New stats • listen_disabled_num • cmd_flush (are you flushing your cache and not realizing it? :) ) • evicted_time under 'stats items' • Only bugfixes and minor features planned for 1.2 series Thursday, April 23, 2009
  • 74. 1.3.* Beta • Scalability improvements • New stats (per-slab get/miss/etc counters) • Binary protocol • Memory optimizations (disable CAS support) Thursday, April 23, 2009
  • 75. ASCII Protocol vs Binary Protocol (memcached-1.3.1 Development Branch) task completion time (secs) ASCII Protocol Binary Protocol concurrent connections Binary Protocol Thursday, April 23, 2009
  • 76. Future • 1.4.0 stable • Multiple Engine Support • Durable • Highly Threaded • New Slab Features Thursday, April 23, 2009
  • 77. More Resources... • http://danga.com/memcached/ • #memcached on Freenode • http://tangent.org/552/libmemcached.html • http://groups.google.com/groups/ memcached • http://dev.mysql.com/doc/refman/5.1/en/ha- memcached.html Thursday, April 23, 2009