Having the right replication functionality is key to ensuring that your data is always available. Intelligent replication is a cornerstone of Riak KV. By default, Riak KV makes three replicas of your data on different nodes. Should one node be unavailable due to server failures or network outages, your application can still read and write data.
Riak KV is designed to ensure your data is always available.
Intelligent replication is a core feature of the Riak KV architecture. Riak KV was designed to operate as a clustered system containing multiple nodes (commodity servers or cloud instances), and intelligent replication allows data to live on multiple machines at once with a single write request.
Riak KV intelligent replication ensures that you can still read, write, and update data if nodes are unavailable. Riak KV allows you to set a replication variable, N (also known as the n_val), that specifies the number of nodes on which a value will be replicated.
An n_val value of 3 (the default) means that each object is replicated 3 times. When an object’s key is mapped onto a given node, Riak KV will automatically replicate the data onto two additional nodes. The n_val parameter enables you to define your replication requirements.
Automatically replicating data inside a cluster ensures data redundancy, and provides the high availability and fault tolerance necessary for today’s Big Data applications.
Your customers won’t wait. They expect your application to be available. In Riak KV, intelligent replication distributes data across the cluster. If a node is unavailable due to a server failure or a network partition, your data remains available.
Scale your application
Riak KV keeps your application scalable by ensuring your data is always available. Riak KV intelligent replication automatically redistributes the load across the cluster as you add and remove nodes.
Improve customer experiences
When millions of people around the world are counting on your application to be up and running, Riak KV intelligent replication protects against outages to ensure your customers have uninterrupted access to your application.
“Monitoring is really a Big Data problem. The challenge for our service is being able to collect, process, store, and visualize billions of metrics every minute from thousands of customers. As a start-up offering these monitoring services, it is vitally important Dataloop.IO works with a platform able to cope with the huge demand being placed upon it. Any losses in data or disruptions to the service would be hugely damaging to our growing reputation. Riak has given us exactly what we need.”
– Colin Hemmings, Co-founder and Chief Architect of Dataloop.IO