Designing Scalable Architectures With MySQL Proxy

Designing Scalable
Architectures
with MySQL Proxy
Giuseppe Maxia
MySQL Community Team Leader - Sun Microsystems
John Loehrer
Data Architect, Gaia Online
Jimmy Guerrero
Sr Product Marketing Manager - Sun Microsystems, Database Group
Copyright 2008 MySQL AB The World’s Most Popular Open Source Database 1
Agenda
• Proxy concepts
• MySQL Proxy architecture
• Proxy with a single back-end
• Proxy with multiple back-ends
• broken?
• missing feature?
• not flexible?
This work is licensed under the Creative Commons Attribution-Share Alike 3.0 Unported License.
Solving database problems
 traditional way
 traditional way
1. file a bug report
 traditional way

2. wait
 traditional way

2. wait
• Open source way
source modify
code
new
compile source
code
• creative (shortsighted) way
bring the logic

at application
level
• creative (enlightened) way

set the logic at
server level
(stored
routines)
• creative (more enlightened) way
set the logic at

protocol level
(proxy)
what can you do with MySQL Proxy
• create new commands
• filter queries (deny specific queries)
• collect statistics on usage
• implement usage quotas
• execute shell commands
• create customized logs
• implement server-side pivot tables
• start/stop a MySQL server remotely
• play movies (seriously!)
• make coffee (really?)
• sharding
• load balancing servers
Basic principles
Basic principles
PROXY
CORE Lua script
connection function
function
hook function
read query function

hook
read result function

hook
??
Lua
{
Perl ?
PHP?
Why not ...
Javascript?
[whatever]?
Lua
• SMALL ( < 200 KB)

• DESIGNED for
EMBEDDED systems
• Widely used (lighttpd)
lighttpd, like MySQL
Proxy, was created by
Jan Kneschke
Lua
Very popular among
game writers
Proxy overview
global Lua script
context
connect_server
session
context
session read_handshake
context
session
read_auth
context
session
context
session
context
session
context
session
read_auth_result
context
read_query
read_query_result
disconnect_client
Proxy overview
/usr/local/sbin/mysql-proxy \
--proxy-lua-script=/path/name.lua
IMPORTANT!
THE SCRIPT DOES NOT START UNTIL THE FIRST
CLIENT CONNECTION
read_query and read_query_result
client
query
result
function
function MySQL read_query_result

read_query Proxy
query if a query is passed directly to the

result server, its result is NOT evaluated
by read_query_result
SERVER
read_query and read_query_result
client
query
result
function
read_query_result
function MySQL
read_query Proxy only if a query is
query queue
added to the query
query
query
query queue, its result is
result
evaluated by
read_query_result
SERVER
An example: user quotas
• You want to limit user access
• When the user reaches a given amount of data,
further queries should be rejected
• the quota must work across sessions
(code online)
http://forge.mysql.com/tools/tool.php?id=111
User quotas
-- some global variables
-- proxy.global.bandwidth
-- will sum up the used bytes
proxy.global.bandwidth
= proxy.global.bandwidth or {}
-- session_user will identify the user

-- throughout the session
local session_user
User quotas
-- session_user needs to be initialized
-- when the user information is passed
-- (during the authentication)
function read_auth( auth )

session_user = auth.username
proxy.global.bandwidth[session_user]
= proxy.global.bandwidth[session_user]
or 0
end
BREAK (a handy Lua idiom)
-- simple assignment
a = a or 0
-- corresponds to:
if a == nil if a ~= nil
then then
a = 0 a = a
else else
a = a a = 0
end end
BREAK (a handy Proxy Lua function )
-- returns an error to the client
function error_result (msg)

proxy.response = {
type =
proxy.MYSQLD_PACKET_ERR,
errmsg = msg,
errcode = 7777,
sqlstate = 'X7777',
}
return proxy.PROXY_SEND_RESULT
end
User quotas
-- read_query (1)
-- checking if the quota has been
-- exceeded
function read_query (packet )
if proxy.global.bandwidth[session_user]
> 10000
and session_user ~= 'root'
then
return error_result(
'you have exceeded your query quota')
end
-- ...
User quotas
-- read_query (2)
-- adding to the totalizer
-- ...
proxy.global.bandwidth[session_user ]
=
+ packet:len()
proxy.queries:append(1, packet )
return proxy.PROXY_SEND_QUERY
end
User quotas
-- read_query_result (1)
-- adding row headers to the totalizer
function read_query_result(inj)
local fields = inj.resultset.fields
local rows = inj.resultset.rows
if fields then
for i = 1, #fields do
=
+
(fields[i] and fields[i].name:len() or 0)
end
User quotas
-- adding rows contents to the totalizer
if rows then
for row in rows do
for i = 1, #fields do
=
+
(row[i] and row[i]:len() or 0)
end
end
end
end
User quotas
-- displaying the current bandwidth
print (session_user .. ' -> ' ..

proxy.global.bandwidth[session_user])
end
User quotas - what the user sees
mysql> select repeat('a', 10000);
ERROR 7777 (X7777): you have exceeded your
query quota
User quotas (VERY advanced)
-- you can create another module
-- to be loaded at run time
-- (this is really advanced)
-- and in such module you define
-- a SHOW QUOTAS command
-- load_multi
mysql> pload show_quotas.lua;
+--------------------------------+
| info |
+--------------------------------+
| module "show_quota.lua" loaded |
+--------------------------------+
mysql> show quotas;
+----------+-------+
| name | quota |
+----------+-------+
| simple | 3578 |
| root | 2111 |
| msandbox | 102 |
+----------+-------+
Designing Scalable
Architectures
with MySQL Proxy
Giuseppe Maxia
John Loehrer
Data Architect, Gaia Online
Jimmy Guerrero
quick-start guide to mysql muxing
ME
online hangout
for teens
5 million unique users per month
1.3 billion forum entries
How does Gaia use the proxy?
connection multiplexer
turning 7,500 connections ...
into 100 MySQL connections
MySQL Proxy
7,500 connections
100 connections
MySQL
PHP Server
Why MUX?
a little history ...
Gaia’s data center in 2003
our server farm...
as our site grew ...
some scaling issues
added more servers ...
site explodes
diagnose the problem
GAIA, under the hood
typical LAMP stack
Linux
Share-Nothing Architecture
each apache server = 50 connections
Parent
Process
httpd
worker thread
child httpd
worker thread
child httpd ...
listener thread
...
50 connections * 150 servers = stampede
close connections each time
setup / tear-down is expensive
connecting to MySQL
read write read
connect challenge auth auth OK
read write read
connect challenge auth auth OK write query read result
75% 19% 6%
read write read

connect challenge auth auth OK write query read result close
MySQL Thread spawning
Client
Client MySQL Server
connect
Client MySQL Server
connect
Client MySQL Server
Thread
connect
new
Client MySQL Server
Thread
connect
new
accept
threads stack up
MySQL I/O
listen
accept
listen
close
accept
end program
NEW THREAD
listen
close
accept
end program
NEW THREAD
blocking read
listen
close
accept
end program
NEW THREAD
blocking read
data
blocking write
listen
close
accept
end program
NEW THREAD
blocking read
eof data
close blocking write
close thread
talking to multiple databases ...
Forum Page Session DB User DB Forum DB
start session
get userdata
get topic
get related users
save session
idle threads
we need a connection pool
ideally ...
Event Queue Thread Pool
Thread
allocate thread
process query
free
release the thread after each query
it’s like a limo ...
... vs. a taxi
decouple threads from connections
Libevent’s callback I/O very sweet!
listen
register callback
event dispatch
active descriptor Trigger callback

epoll/kqueue rulez! (proxy uses it)
MySQL 6.0 taking the right steps ...
In the meantime ...
by default, mysql-proxy isn’t a pooler
lua hooks are awesome!
hack some lua scripts ...
goal: recycle back-end connections
keep clients connected
proxy stays transparent
how to twist the proxy into a pooler
connecting: find an idle back-end
Proxy
client
Proxy
client
connect ahead idle pool

authentication
Client Proxy Back-End Pool
MySQL Server
MySQL Server
connect MySQL Server


allocate

allocate

allocate
com-change-user

allocate
com-change-user
challenge

allocate
com-change-user
challenge
authenticate

allocate
com-change-user
challenge
authenticate
OK

allocate
com-change-user
challenge
authenticate
OK
free back-end
run the query
don’t be afraid to let go
Client Proxy
query
query
allocate
query MySQL Server

allocate
process query
query MySQL Server

allocate
process query
free back-end
connections scale!
Designing Scalable
Architectures
with MySQL Proxy
Giuseppe Maxia
John Loehrer
Data Architect, Gaiaonline
Jimmy Guerrero
MySQL Load Balancer
MySQL software which can route database queries to
the appropriate database in a scale-out environment.
• Intelligently route reads across slaves

• Use database least behind, least loaded
• Remove latent slaves from read rotation
• Other distribution algorithms
• Improves/scales throughput of reads for online applications

• Helps customers reduce cost of adding slaves
MySQL Query Analyzer
MySQL Enterprise Monitor feature that allows users
to trace, monitor, and analyze MySQL query activity
for specific servers, users, and applications.
• Adaptive “Evil” query collection/tracing

• Historical browsing/analysis
• “Needle in a haystack” identification of worst
queries
• Worst execution times, # of execs, etc.
SQL code is the #2 cause of performance issues

97% of those surveyed will use this
MySQL Query Analyzer
Next Steps
MySQL Proxy Forum
http://forums.mysql.com/list.php?146
Documentation – MySQL Proxy

http://dev.mysql.com/doc/refman/5.1/en/mysql-proxy.html
Documentation – MySQL Load Balancer

http://dev.mysql.com/doc/refman/5.1/en/load-balancer.html
Beta Testing MySQL Proxy-Enabled Products

enterprise-beta@mysql.com
MySQL Proxy Online Poll
http://dev.mysql.com/tech-resources/quickpolls/
Questions?
Giuseppe Maxia
John Loehrer
Data Architect, Gaiaonline
Jimmy Guerrero

Designing Scalable Architectures With MySQL Proxy

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Designing Scalable Architectures With MySQL Proxy

Transféré par

Droits d'auteur :

Formats disponibles

Designing Scalable

1. file a bug report

1. file a bug report

1. file a bug report

• creative (shortsighted) way

bring the logic

• creative (enlightened) way

• creative (more enlightened) way

set the logic at

read query function

read result function

• SMALL ( < 200 KB)

function MySQL read_query_result

query if a query is passed directly to the

-- session_user will identify the user

function read_auth( auth )

function error_result (msg)

print (session_user .. ' -> ' ..

child httpd ...

read write read

close blocking write

get related users

active descriptor Trigger callback

connect ahead idle pool

connect MySQL Server

connect MySQL Server

connect MySQL Server

connect MySQL Server

connect MySQL Server

connect MySQL Server

connect MySQL Server

connect MySQL Server

query MySQL Server

query MySQL Server

• Intelligently route reads across slaves

• Improves/scales throughput of reads for online applications

• Adaptive “Evil” query collection/tracing

SQL code is the #2 cause of performance issues

Documentation – MySQL Proxy

Documentation – MySQL Load Balancer

Beta Testing MySQL Proxy-Enabled Products

MySQL Proxy Online Poll

Vous aimerez peut-être aussi