Scaling Databases - Web DevelopmentYouTube↗  IA↗(long shot)
Education  64.3K views 795 likes 7 dislikes   25 Apr 2012
Udacity 651K subscribers United States YouTube↗


Please select the problem
You must specify your email and describe the problem in detail
Describe your problem in detail if it's not obvious

Click to Play Video Click to Play Video
okay I'm going to talk for a moment about how to scale databases now this is
Quiero hablar un momento sobre cómo ordenar las bases de datos.
like databases themselves this is a very
Como las bases de datos, este es un tema muy amplio,
broad topic but there's a couple concepts I'd like to introduce because
pero hay unos conceptos de los que me gustaría hablar
they're going to affect us in this class
porque nos van a afectar en este aspecto.
there's two reasons you might need to
Hay dos razones por las que necesitarías ordenar una base de datos.
scale it database one is too much load
Una es porque haya mucha carga,
as in the database is just doing too much work you know you've got some
en una base de datos eso supone mucho trabajo.
website that is getting you know
Tienes una página web que recibe
millions of requests a day or you know thousands of requests a second and
millones de peticiones al día, miles por segundo,
you've got this one machine that's got
y tienes esta máquina que almacena tus datos
your data on it and it just can't keep up with all of the work so what you
y no puede aguantar todo el trabajo.
might do in this case is you know take
Lo que puedes hacer es tomar tu base de datos,
your database which are often represented as little cylinders and
que a menudo se representa con pequeños cilindros,
replicate it to other databases so every
y copiarla a otras bases de datos.
time we insert a piece of data to to
Así cada vez que insertes datos a esta base,
this database we we send it over to the
los enviamos a otras.
other guys so all of our database rights go into this this one master database in
Todo lo que esté en tu base de datos va a la base de datos ama.
turn get all of that data gets replicated to to the slave databases so
Todos los datos se copian a las bases de datos esclavas.
now we have three databases with all of
Ahora tenemos 3 bases de datos con todos nuestros datos
our data on it and so if you know you've got a site like Reddit that's getting thousands and thousands of requests a
si tienes un sitio como Reddit que tiene miles y miles de peticiones por segundo,
second instead of sending all of the the
en lugar de enviar todas las lecturas de la base de datos,
database reads all of the lookups to to
todas las búsquedas, a esta base,
this database we can send it you know to these databases and and that alleviates
podemos enviarlas a esas bases de datos.
a lot of this load off of this master
Eso alivia mucho la carga que tiene la base de datos ama.
database now there are much more complicated schemes where you might you
Hay muchos esquemas más complicados
know have a multi master system or a
donde puedes tener un sistema con múltiples amas
bunch of databases that are all working
o un montón de bases de datos que trabajan juntas
together but this is a fairly common common set up and a fairly common
pero este es un ajuste común
approach to you know spreading around
y una forma común de difundir la lectura.
the reads because generally in most systems you have a lot more reads than
Porque en general en la mayoría de los sistemas tienes más lecturas
you do writes so if your master can keep
que escrituras.
up with all of the writes your slaves
Si tu ama puede soportar todas las escrituras,
can handle all the reads there are some
tus esclavos pueden suportar todas las lecturas.
downsides to this one is that it doesn't
Hay algunas desventajas.
increase the speed of writes you know
Una es que no aumenta la velocidad de las escrituras.
we're still bottlenecks by this one master receiving all of this data who
La base de datos ama aún está agobiada por recibir tantos datos
has to you know in turn send it to all
y tiene que enviárselo a los esclavos.
of these slaves so if we can still fit all of our writes on this one machine
Si aún podemos soportar todas las escrituras en esta máquina, estamos en buena forma
we're in good shape and we can spread the reads over you know as many machines
y podemos difundir las lecturas por todas las máquinas que necesitemos.
as we need you know some of these slaves could even replicate to other machines
Algunos de estos esclavos pueden copiar las otras máquinas.
another downside is this notion of replication lag and this occurs when you write a piece of data to the master but
Otra desventaja es la noción de "replication lag". Esto ocurre cuando envías unos datos al master
the read hits one of these slaves before the data has propagated and that's
pero la lectura llega a uno de los esclavos antes de que los datos se hayan propagado.
called replication
Esto se llama "replication lag".
and you can sometimes get some funny behavior and the reason I bring it up
A veces puedes tener un comportamiento divertido.
now is because in the database we'll be
La razón por la que saco esto ahora es porque en la base que usaremos,
using this class you can occasionally see symptoms of this type of behavior
puedes ver síntomas de este comportamiento.
it's not necessarily owing to a
No es necesario tener un ajusto amo/esclavo,
master/slave set up but this this is the the general concept now what happens if
pero es el concepto general.
you have too much data for one machine
¿Qué pasa si tienes muchos datos para una máquina?
the replication hand handles the case
La copia se da en los casos
where you have too much load basically
en los que tienes mucha carga.
you've got to do you know too many reads so you can replicate your database to
Básicamente, tienes que hacer muchas lecturas, así que puedes copiar tu base de datos.
scale this up what if you just have too much data you know you master database
¿Qué pasa si tienes muchos datos,
or any one of these databases can't even hold all of your data you know it
tu base de datos ama, o cualquier otra, no puede soportar los datos,
doesn't fit in memory or even in disk
no cabe en la memoria o el disco,
it's just there's just too much one
si es demasiado?
things you can do is you can shard the database and this is a fairly simple
Una de las cosas que puedes hacer es fragmentar la base de datos.
approach where instead of having one
Esta es una opción simple, en la que en lugar de tener una ama,
master you might have a couple that are all the same size despite my drawing so
puedes tener un par que tengan el mismo tamaño, a pesar de mi dibujo.
let's say we're storing on links
Digamos que vamos a almacenar nuestra base de datos de links.
database won't approach the sharding might be to store links you know 1 through 100 here actually more like 1
Una opción para fragmentar puede ser almacenar los links 1-100 aquí
through a billion here and 101 to 200
en realidad, de uno a mil millones aquí, y 101-200 aquí
here and you know 201 to 300 here and
y 201-300 aquí
basically you you store some of your
y así almacenas algunos links aquí y otros aquí
links here and some of your links here and some of your links here in this
y algunos aquí en esta base de datos.
database obviously you know 1 through 100 probably isn't the correct approach
Obviamente, 1-100 no tiene por qué ser la opción correcta.
you could probably use a hashing approach as if we had a hash table and
Puedes usar enfocarlo como si tuviéramos una tabla hash
these were all cells in our hash table
y estos fueran celdas en nuestra tabla hash
and you you know hash on some particular
y tu hash en una llave específica.
key in this case I'm referring to linked IDs you know the idea of whatever you're storing to figure out which database you
En este caso, me refiero a los ID de los links, el ID de lo que estés almacenando,
want to store it this is cool because
para averiguar qué base de datos quieres usar para almacenarlos.
now if we triple our write load from
Esto es interesante porque ahora
from this scenario over here we have 3
si triplicamos nuestra carga de escritura en este escenario de aquí,
database machines to to handle it
tenemos tres bases de datos que usar.
likewise we already have 3 machines to
También tenemos tres máquinas con las que soportar la carga de lectura.
handle the read load and of course you can replicate these machines as well and
Por supuesto, puedes copiar esas máquinas también.
a lot of systems both shard and
Muchos sistemas fragmentan y copian las máquinas
replicate your machines if you really want to get fancy but there are some
si quieres ir a más.
downsides to sharding as well you know
Pero hay algunas desventajas en fragmentar también.
one of the downsides is the queries get much more complex you know what if I
Una de las desventajas es que las consultas son mucho más complejas.
said that this helps the case of find me
¿Qué pasa si dijese que esto ayuda al caso
the link whose ID is 150 you know we
de buscar el link con el ID 150?
just say ok which machine is 150 go to
Vale, ¿dónde va el 150?
and you know what hashes you know to
Sabes los hash que van en esta máquina,
this machine and we do our read and
y hacemos nuestra lectura y eso es todo.
that's that what if I said get me all the links sorted by hotness for example
¿Qué pasa si quiero que me consiga todos los links ordenados por popularidad?
well we don't know
No sabes dónde ese link puede estar,
where that link may reside especially if
especialmente si usamos un algoritmo hash,
we're using a hashing algorithm that's you know more sophisticated than you
eso es más sofisticado que 1-100.
know 1 through 100 that's called a range query and a range
Eso se llama un "range query",
query might have to hit all of your
que puede tener que llegar a todas las máquinas.
machines and you know then then we've lost our you know one of our advantages
Hemos perdido una de nuestras ventajas para usar esto.
to using this now you could replicate these to spread that around but you had
Puedes copiar estos para difundirlos,
all these machines and then merge the
pero tienes que enviarlos a todas las máquinas e incorporar los resultados
results probably and you know do the sort again in memory and so you know
y ordenarlos de nuevo y guardar.
some queries become very very complex
Algunas consultas se vuelven muy, muy complejas.
another downside is that joins become difficult or impossible that makes sense
Otra desventaja es que esa unión se vuelve difícil o imposible.
you know so if we have this master
Eso tiene sentido.
database that has multiple tables on it
Si tenemos esta base de datos enorme con múltiples tablas, podemos hacer uniones.
you know we can do joints but if we start charting our database up you know
Pero si fragmentamos la base de datos,
we have one database doesn't fit on one machine you know so if we can't fit one
tenemos una base de datos que cabe en una máquina.
database on our machine how are we going
Si no podemos encajar una base de datos en nuestra máquina
to fit you know multiple databases this
¿cómo vamos a encajar varias bases de datos?
notion that you have all of your tables
Esta idea de que tienes todas las tablas en el mismo lugar para unirlas,
in the same place to do a joint to do those fancy joint sequel queries becomes
para hacer más cosas con las consultas SQL,
a lot more difficult now there are
se vuelve mucho más difícil.
there's a lot of research going on of
Hay mucha investigación sobre
building systems that can overcome these
la construcción de sistemas que superen esas desventajas,
downsides but if you take the naive approach to replicating and charting
pero si tomas una opción débil de copias y fragmentos,
these are some of the things you have to deal with now I generally the benefits
tendrás que lidiar con ese tipo de cosas.
outweigh the the downsides because if
En general, los beneficios superan a las desventajas,
you design your data in such a way that you don't need to do joints or you don't
porque si diseñas tus datos de forma que no necesites hacer uniones
need to do complex queries you're in
o no necesites hacer consultas complejas,
very good shape with both of these or
estarás preparado para esto.
maybe you have you know multiple setups you know once setup for handling your
Quizás tengas varios ajustes:
general load and another setup for handling your complex queries you know
uno para tu carga general y otro para las consultas complejas.
that's ok and that's actually very
No pasa nada, es algo muy común.
common now I wanted to introduce these
Quería presentaros estos conceptos
concepts because the database we'll be using in this class Google the App
porque la base de datos que usaremos,
Engine data store actually has some of these limitations you know there are no
el App Engine Datastore de Google, tiene alguna de estas limitaciones.
joins allowed even though it does provide a sequel interface and a lot of
No se permiten las uniones, aunque ofrezca una interfaz SQL.
more complex queries that you could do in a general sequel database now you
Más consultas complejas de las que puedes hacer en una base de datos SQL general,
can't do in Google datastore but that's ok you know you get this really nice
no las puedes hacer en Google Datastore, pero no pasa nada.
benefit of having a database that is
Consigues este beneficio tan interesante de tener una base de datos
charted and replicated to you know wazoo
completamente fragmentada y copiada.
and back it's actually very fast and reliable you know we were probably not
Es algo muy rápido y fiable.
going to have to worry about systems
Probablemente no tendrás que preocuparte de que se caigan los sitemas.
crashing something that you know replicating a database can help with and
Es algo con lo que copiar una base de datos te puede ayudar
charting a database can help with ok so
al igual que fragmentar una base de datos.
just wanted to introduce those concepts
Bien, solo quería presentaros esos conceptos.
let's have a quick quiz on it before we move along
Hagamos una pregunta rápida antes de seguir.
which is an appropriate technique for increasing the read speed from a
¿Cuál es una técnica apropiada para aumentar la velocidad de lectura de una base de datos?
database get a faster machine replicate
Conseguir una máquina más rápida. Copiar la base de datos.
the database store less data or press the turbo button
Almacenar menos datos. Pulsar el botón turbo.
Similar videos:

Memcached - Web Development
Udacity
13.5K views 115 likes 3 dislikes 23 Feb 2015100% match

Dynamic Content - Web Development
Udacity
6.1K views 2 likes 0 dislikes 23 Feb 2015100% match

HTML Basics
Udacity
16.5K views 75 likes 2 dislikes 16 Jul 2014100% match

The Method Attribute - Web Development
Udacity
9.6K views 14 likes 8 dislikes 22 Apr 2012100% match

Whitespace
Udacity
47.8K views 26 likes 4 dislikes 23 Feb 2015100% match

Course Outline - Web Development
Udacity
46.2K views 169 likes 0 dislikes 23 Feb 2015100% match

Paragraph Tag - Web Development
Udacity
8.6K views 4 likes 0 dislikes 23 Feb 2015100% match

Shopping List Take 1 - Web Development
Udacity
841 views 1 likes 0 dislikes 23 Feb 2015100% match

Cookie Examples - Web Development
Udacity
5.4K views 29 likes 2 dislikes 23 Feb 2015100% match

Overview - Web Development
Udacity
10.4K views 26 likes 0 dislikes 27 May 2012100% match

Types of Databases - Web Development
Udacity
8.5K views 43 likes 4 dislikes 25 Apr 2012100% match

File Types
Udacity
34.4K views 11 likes 3 dislikes 23 Feb 2015100% match

App Server Architecture - Web Development
Udacity
14K views 52 likes 5 dislikes 27 May 2012100% match

Missing End Tag
Udacity
43K views 15 likes 1 dislikes 23 Feb 2015100% match

Missing End Tag Solution - Web Development
Udacity
22.8K views 7 likes 0 dislikes 23 Feb 2015100% match

User Input - Web Development
Udacity
7.7K views 3 likes 0 dislikes 22 Apr 2012100% match

Database Architecture - Web Development
Udacity
8.3K views 29 likes 1 dislikes 27 May 2012100% match

Part 2 - Web Development
Udacity
643 views 2 likes 0 dislikes 07 May 2012100% match

Span and Div
Udacity
50.3K views 27 likes 1 dislikes 23 Feb 2015100% match

Differences Between Get and Post - Web Development
Udacity
78.3K views 516 likes 18 dislikes 22 Apr 2012100% match

Document Structure
Udacity
7.2K views 24 likes 1 dislikes 23 Feb 2015100% match

App Server Scaling - Web Development
Udacity
28.2K views 277 likes 8 dislikes 23 Feb 2015100% match

Cookie Headers - Web Development
Udacity
4.2K views 23 likes 1 dislikes 23 Feb 2015100% match

Datastore Types Solution - Web Development
Udacity
656 views 1 likes 0 dislikes 25 Apr 2012100% match

Scaling - Web Development
Udacity
12.5K views 81 likes 1 dislikes 27 May 201290% match

Introduction - Web Development
Udacity
23.3K views 107 likes 0 dislikes 23 Feb 201573% match

Ad Networks - Web Development
Udacity
6.7K views 27 likes 3 dislikes 23 Feb 201573% match

Code Organization - Web Development
Udacity
7.2K views 25 likes 1 dislikes 27 May 201267% match

Whitespace Solution - Web Development
Udacity
15.7K views 10 likes 1 dislikes 23 Feb 201567% match

Precomputed Caching - Web Development
Udacity
12.2K views 80 likes 0 dislikes 27 May 201267% match


Settings Privacy About Contact Donate