As organizations race to unearth value within big data, many are finding that traditional server environments are not up to the task.
“Big data requires a fundamentally different architecture,” says Raghunath Nambiar, distinguished engineer and chief architect of big data solutions at Cisco. Recently elected by the Transaction Processing Performance Council (TPC) to lead the development of the industry’s first big data benchmark standard, Nambiar states, “To get the most out of big data, companies need an infrastructure that is tuned for big data workloads, with better performance and scalability than traditional environments.”
To meet these requirements, Cisco designed a comprehensive solution: the Intel® Xeon®processor-based Cisco Unified Computing System™ (Cisco UCS®) Common Platform Architecture (CPA) for Big Data. The architecture takes advantage of the Cisco UCS®, which utilizes Cisco Nexus® switches. The comprehensive stack is designed specifically for big data, and includes compute, storage, connectivity, and unified management.
- Cisco UCS 6200 Series Fabric Interconnects provide high-speed, low-latency connectivity for servers, as well as integrated, unified management for all connected devices
- Cisco UCS 2200 Series Fabric Extenders provide highly scalable and extremely cost-effective connectivity for a large number of nodes
- Cisco UCS C240 M3 Rack Servers are two-rack-unit (2RU) servers designed for a wide range of compute, I/O, and storage capacity demands
- Cisco Nexus switches, the foundation of Cisco Unified Fabric, deliver exceptional availability and outstanding scalability to meet the requirements of mission-critical data centers
“Our core strength in networking technologies, combined with Cisco UCS, enables us to offer big data clusters with hundreds of servers and petabytes of storage,” says Nambiar. “And all of it can be managed from a single pane using UCS Manager and UCS Central, either in a data center or distributed globally.”
The network’s crucial role
Sometimes forgotten in big data discussions and decisions is the network, which plays a crucial role within the cluster and between the cluster and data sources. By combining compute and networking technologies, Cisco UCS CPA for Big Data offers predictable performance, complete data isolation, and network programmability to meet big data requirements. It also seamlessly integrates unified management capabilities with diverse datasets and enterprise applications to minimize the time and cost of deployment.
“If interest and adoption are any indication, Cisco UCS CPA for Big Data is very much needed in the marketplace,” says Nambiar. “Within a few months after it became available, CPA was already being deployed in a range of industries, including finance, retail, service provider, content management, and government.”