US-based service provider CenturyLink plans to bring cloud-based hyperscale computing and storage to the UK by...
The new service will allow businesses to make use of web-scale big data analytics to provide, for example, near real-time responses to customer activity, such as targeted advertising.
Hyperscale compute/storage architectures are those pioneered by Amazon, Google, Facebook et al and use large numbers of commodity servers with direct-attached disk where compute, throughput and storage are the result of many nodes operating in grid-like fashion with replication between them.
Hyperscale architectures operate with redundancy at the level of entire compute/storage nodes, unlike enterprise IT configurations that swap out individual faulty components.
More on hyperscale storage and compute
They are built around low-cost commodity hardware, including the so-called open source hardware of the Open Compute movement, and are aimed at big data analytics operations, particularly those that need rapid interrogation of large amounts of data, such as in web operations, and using distributed analytics platforms such as Hadoop.
The CenturyLink Cloud platform this week announced the availability of its hyperscale service in the US and is designed for web-scale workloads, big data and cloud-native applications.
Its hyperscale service offers all flash storage and web-scale architectures that can use Couchbase, MongoDB and other NoSQL technologies. CenturyLink claims users of the service will see performance at or above 15,000 input/output operations per second (IOPS) for a range of workloads.
All CenturyLink Cloud services, including Hyperscale, will be available from datacentres in Paris and London area in the second quarter of 2014.