Like batteries, you expect operational simplicity to be included in your time series database, but that’s not always the case. With Riak TS, operational simplicity is designed into the architecture.
Riak TS allows you to easily add and remove nodes in your cluster. Auto-replication eliminates sharding and manual replication. Riak TS also supports rolling upgrades so there is no need for a maintenance window.
Operational simplicity also includes ease of capacity planning. Riak TS is less expensive to operate than other NoSQL databases. It requires less CPU, therefore fewer nodes, to write the same workload. Also, since the resource usage is predictable (near-linear), it is easy to scale up and down.
Riak TS builds in operational simplicity using a “Ring” architecture with automatic replication and hinted handoff, which ensure application resiliency.
Each node in a Riak TS cluster manages one or many virtual nodes, or vnodes. Each vnode is a separate process, which is assigned a partition of the Riak Ring, and is responsible for a number of operations in a Riak TS cluster. The concept of vnodes is important as we look at data replication. No single vnode is responsible for more than one replica of an object. Each object belongs to a primary vnode, and is then automatically replicated to neighboring vnodes located on separate nodes in the cluster.
This automatic replication ensures that there are multiple copies of data across multiple servers. By default, Riak TS makes 3 replicas, each belonging to a unique vnode.
Riak TS automatically distributes data across nodes in a Riak TS cluster and yields a near-linear performance increase as you add capacity. When you add (or remove) machines, data is rebalanced automatically in a non-blocking operation.
Should a node go down, a neighboring node takes over. When the failed node returns, the updates received by the neighboring node are handed back to it. This functionality is referred to as hinted handoff and ensures the availability of writes and updates is maintained automatically, minimizing the operational burden of failure conditions.
Operational simplicity saves time and money, and reduces the burden on your Ops team. With Riak TS your application keeps on running. In the event of a hardware or network failure, you can wait and replace nodes during normal work hours. No need to lose sleep.
Operate IoT applications at peak capacity
IoT applications must support millions of global devices. Riak TS enables you to easily add capacity on demand to scale, ingest, transform, store, and analyze data for these millions of devices.
Reduce operational costs
Operational costs include platform, software, and people. Riak TS scales using commodity hardware and is highly available reducing the need for manual intervention. This allows you to apply your operations team to other tasks. No dedicated operators needed.
Downtime costs money. Riak TS is architected for resiliency, and with rolling upgrades no maintenance window is required. During operations, if a Riak TS node fails, other nodes in the cluster takeover reads and writes ensuring your data is always available.
“Now more than ever, companies rely on data to understand, anticipate, and respond to customer needs. With more than 30,000 clients and 35 million users, Concur needs a database solution that can handle high volumes without latency or downtime. Basho allows us to scale as our client base grows, and helps ensure we keep our customers and users happy across the globe.”
– Cameron Etezadi, Deputy CTO, Concur Technologies