Partitioning is the process of splitting your data into multiple KeyDB instances, so that every instance will only contain a subset of your keys. The first part of this document will introduce you to the concept of partitioning, the second part will show you the alternatives for KeyDB partitioning.
Why partitioning is useful
Partitioning in KeyDB serves two main goals:
- It allows for much larger databases, using the sum of the memory of many computers. Without partitioning you are limited to the amount of memory a single computer can support.
- It allows scaling the computational power to multiple cores and multiple computers, and the network bandwidth to multiple computers and network adapters.
There are different partitioning criteria. Imagine we have four KeyDB instances R0, R1, R2, R3, and many keys representing users like
user:2, ... and so forth, we can find different ways to select in which instance we store a given key. In other words there are different systems to map a given key to a given KeyDB server.
One of the simplest ways to perform partitioning is with range partitioning, and is accomplished by mapping ranges of objects into specific KeyDB instances. For example, I could say users from ID 0 to ID 10000 will go into instance R0, while users form ID 10001 to ID 20000 will go into instance R1 and so forth.
This system works and is actually used in practice, however, it has the disadvantage of requiring a table that maps ranges to instances. This table needs to be managed and a table is needed for every kind of object, so therefore range partitioning in KeyDB is often undesirable because it is much more inefficient than other alternative partitioning approaches.
An alternative to range partitioning is hash partitioning. This scheme works with any key, without requiring a key in the form
object_name:<id>, and is as simple as:
- Take the key name and use a hash function (e.g., the
crc32hash function) to turn it into a number. For example, if the key is
crc32(foobar)will output something like
- Use a modulo operation with this number in order to turn it into a number between 0 and 3, so that this number can be mapped to one of my four KeyDB instances.
93024922 modulo 4equals
2, so I know my key
foobarshould be stored into the R2 instance. Note: the modulo operation returns the remainder from a division operation, and is implemented with the
%operator in many programming languages.
There are many other ways to perform partitioning, but with these two examples you should get the idea. One advanced form of hash partitioning is called consistent hashing and is implemented by a few KeyDB clients and proxies.
Different implementations of partitioning
Partitioning can be the responsibility of different parts of a software stack.
- Client side partitioning means that the clients directly select the right node where to write or read a given key. Many KeyDB clients implement client side partitioning.
- Proxy assisted partitioning means that our clients send requests to a proxy that is able to speak the KeyDB protocol, instead of sending requests directly to the right KeyDB instance. The proxy will make sure to forward our request to the right KeyDB instance accordingly to the configured partitioning schema, and will send the replies back to the client. The KeyDB and Memcached proxy Twemproxy implements proxy assisted partitioning.
- Query routing means that you can send your query to a random instance, and the instance will make sure to forward your query to the right node. KeyDB Cluster implements an hybrid form of query routing, with the help of the client (the request is not directly forwarded from a KeyDB instance to another, but the client gets redirected to the right node).
Disadvantages of partitioning
Some features of KeyDB don't play very well with partitioning:
- Operations involving multiple keys are usually not supported. For instance you can't perform the intersection between two sets if they are stored in keys that are mapped to different KeyDB instances (actually there are ways to do this, but not directly).
- KeyDB transactions involving multiple keys can not be used.
- The partitioning granularity is the key, so it is not possible to shard a dataset with a single huge key like a very big sorted set.
- When partitioning is used, data handling is more complex, for instance you have to handle multiple RDB / AOF files, and to make a backup of your data you need to aggregate the persistence files from multiple instances and hosts.
- Adding and removing capacity can be complex. For instance KeyDB Cluster supports mostly transparent rebalancing of data with the ability to add and remove nodes at runtime, but other systems like client side partitioning and proxies don't support this feature. However a technique called Pre-sharding helps in this regard.
Data store or cache?
Although partitioning in KeyDB is conceptually the same whether using KeyDB as a data store or as a cache, there is a significant limitation when using it as a data store. When KeyDB is used as a data store, a given key must always map to the same KeyDB instance. When KeyDB is used as a cache, if a given node is unavailable it is not a big problem if a different node is used, altering the key-instance map as we wish to improve the availability of the system (that is, the ability of the system to reply to our queries).
Consistent hashing implementations are often able to switch to other nodes if the preferred node for a given key is not available. Similarly if you add a new node, part of the new keys will start to be stored on the new node.
The main concept here is the following:
- If KeyDB is used as a cache scaling up and down using consistent hashing is easy.
- If KeyDB is used as a store, a fixed keys-to-nodes map is used, so the number of nodes must be fixed and cannot vary. Otherwise, a system is needed that is able to rebalance keys between nodes when nodes are added or removed, and currently only KeyDB Cluster is able to do this - KeyDB Cluster is generally available and production-ready.
We learned that a problem with partitioning is that, unless we are using KeyDB as a cache, to add and remove nodes can be tricky, and it is much simpler to use a fixed keys-instances map.
However the data storage needs may vary over the time. Today I can live with 10 KeyDB nodes (instances), but tomorrow I may need 50 nodes.
Since KeyDB is extremely small footprint and lightweight (a spare instance uses 1 MB of memory), a simple approach to this problem is to start with a lot of instances since the start. Even if you start with just one server, you can decide to live in a distributed world since your first day, and run multiple KeyDB instances in your single server, using partitioning.
And you can select this number of instances to be quite big since the start. For example, 32 or 64 instances could do the trick for most users, and will provide enough room for growth.
In this way as your data storage needs increase and you need more KeyDB servers, what to do is to simply move instances from one server to another. Once you add the first additional server, you will need to move half of the KeyDB instances from the first server to the second, and so forth.
Using KeyDB replication you will likely be able to do the move with minimal or no downtime for your users:
- Start empty instances in your new server.
- Move data configuring these new instances as slaves for your source instances.
- Stop your clients.
- Update the configuration of the moved instances with the new server IP address.
- Send the
SLAVEOF NO ONEcommand to the slaves in the new server.
- Restart your clients with the new updated configuration.
- Finally shut down the no longer used instances in the old server.
Implementations of KeyDB partitioning
So far we covered KeyDB partitioning in theory, but what about practice? What system should you use?
KeyDB Cluster is the preferred way to get automatic sharding and high availability. It is generally available and production-ready as of April 1st, 2015. You can get more information about KeyDB Cluster in the Cluster tutorial.
Once KeyDB Cluster will be available, and if a KeyDB Cluster compliant client is available for your language, KeyDB Cluster will be the de facto standard for KeyDB partitioning.
KeyDB Cluster is a mix between query routing and client side partitioning.
Twemproxy is a proxy developed at Twitter for the Memcached ASCII and the KeyDB protocol. It is single threaded, it is written in C, and is extremely fast. It is open source software released under the terms of the Apache 2.0 license.
Twemproxy supports automatic partitioning among multiple KeyDB instances, with optional node ejection if a node is not available (this will change the keys-instances map, so you should use this feature only if you are using KeyDB as a cache).
It is not a single point of failure since you can start multiple proxies and instruct your clients to connect to the first that accepts the connection.
Basically Twemproxy is an intermediate layer between clients and KeyDB instances, that will reliably handle partitioning for us with minimal additional complexities.
Clients supporting consistent hashing
An alternative to Twemproxy is to use a client that implements client side partitioning via consistent hashing or other similar algorithms. There are multiple KeyDB clients with support for consistent hashing, notably redis-rb and Predis.
Please check the full list of redis clients(KeyDB compatible) to check if there is a mature client with consistent hashing implementation for your language.