Skip to content
This repository
branch: release-096li4
Fetching contributors…

Octocat-spinner-32-eaf2f5

Cannot retrieve contributors at this time

file 117 lines (89 sloc) 6.492 kb
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116
Why it is time for a new database

Management of shared mutable state is the crux of scalability. An application without shared state can be scaled trivially
by buying a load balancer and more servers.

[why we need to share state]

Shared state is what makes the application interesting. Modern web applications are getting more shared state--all
the personalization and user data that is associated with web 2.0 is a whole lot of state.

Relational databases are the gold standard of state management and have been for the last 30 years. The relational
data model is one of the better contributions of computer science. Relational databases are here to stay, but they are not
appropriate for every storage situation (we still have files and hashmaps, after all). In particular relational databases,
and in particular the current SQL products available, present problems for large scale software.

But first lets review what is good about the relational model:
1. It separates data from look-up strategy, queries are declarative
2. It is incredibly flexible--most desirable data operations can be expressed in relations

The problems with current RDBMSes are the following:
1. Building a website on a centralized database combines the two worst enemies, disk io and network io, for every
server in the system, and centralizes them onto a single machine. This is clearly not feasible.
2. The relational model does not parallelize well, due to the difficulties of sharing state. Most databases favor
consistency over performance (2PC). You can't have both.
3. SQL sucks for application development. Most web apps are programs that construct little sql statement mini-programs
by concatenating together strings. This completely ungainly. SQL is vender specific so you are tied to the vendors
database. This means development suffers immensely. Unit testing any piece of code that involves database access is quite
problematic because the unit test will be slow and will depend on a server running on another machine which can't be
bootstrapped as part of the test code. This could easily be solved if vendors provided a light-weight drop in
replacement that could be used for testing. Approximately 10000 products attempt to work around this problem
by automatically mapping your application to sql. This has been called the Vietnam of Computer Science (referring to the
American war, not the country...and of course computer scientists are not particularly interested in this problem as
it is too practical).

To begin to understand the problems with relational databases, let's first review some basic facts

For context let's review some basic facts about a modern computer:
                  roundtrip throughput
  Java HashMap
  BDB Btree
  MySQL local
  MySQL remote
  
  1000 iterations of a trivial loop
  allocating and initializing 1000 10 byte objects
  1000 200 byte local http requests
  1000 200 byte remote http requests (100 mbit ethernet)
  
1. Disk IO is stagnant
2. Moore's law is going gangbusters for parallel applications (e.g. any application that has shared state)
  
So from this you can get some feel for the cost of various operations. It is clear that the disk operations are
our greatest enemy followed by the network. We can iterate over XXX items of an array in the time required to
look-up a single item via a log(n) b-tree index. We can iterate over YYY items in the time required to complete
a mysql request.

The relational model is great, when accessed programmatically, but it is best for little applications (of which there are many).

In a high scalability scenario with a shared db, the advantages of the relational model disappear. No longer do you have
flexibility, what you have is a system in which most operations will bring the database to its knees along with
everyone depending on it. Think about the average table, of the set of possible queries, only a few can be issued without
difficulty. With this being the case, the abstraction of the lookup structure becomes a major problem--it is impossible to
tell by looking at code whether it will run quickly or bring down the system, that is totally dependent on data sizes and
indexing structure.

There are two basic strategies for scaling up databases
1. Partitioning and replication
2. Caching

Caching is the most popular because it is the simplest to implement. Basically you replace all your fancy queries with
simple puts and gets, and each operation goes first to the cache and then to the db. The cache may not actually be much
faster than the db, but it can be distributed whereas the db cannot.

There are two basic caching strategies
1. Local read/write with background replication
2. Remote access

In the first strategy we try to keep each object on each server. We read and write to the local cache, and in the background
the cache replicates our changes out to the other servers. This is the ehcache strategy.

The problem with this is that we must store the whole cache local which means either our entire dataset must fit in memory
or we have to use the local disk. Of course the local disk may well be slower than the network, so our cache may become slower
then the database itself (though, importantly, it is distributed). In addition each write now must be written to the db as well
as every application server.

How it works?

Most storage systems look up your data in a table. This is difficult in a distributed system because the lookup table is large and
always changing. Keeping this in sync and uncorrupted is a scalability problem--metadata updates are a big bottleneck.

A simpler approach is much faster--use the key to calculate the location. This means coming up with a function
f(key)->(n1,n2,n3)
where n1, n2, n3 are the nodes containing the data.

In this model only the key and the cluster topology (which servers are where) are needed to locate and retrieve data.


- Relational data model is excellent
- SQL is a pain
- relational very hard to distribute
- SQL breaks testability (without great pain)
- Don't care about POSIX filesystem type stuff

Needs For A High-Scalability Distributed Storage System
- Commodity hardware -- cheap hardware, commodity network, shared nothing
- Incremental scalability to hundreds of nodes
- High throughput
- Low latency
- No single point of failure
- Easy enough for OPS team or software engineer to manage (No DBA)

References
- Google BigTable
- Amazon Dynamo
- One-size fits all

Something went wrong with that request. Please try again.