FlashStack Technical overview
FlashStack is a converged infrastructure offering that combines technology from two leading enterprise vendors — Cisco and Pure Storage — in a single integrated solution. Featuring server and networking technology from Cisco and all-flash storage technology from Pure Storage.
FlashStack provides an IT infrastructure solution that can be used for multi-hypervisor, bare metal, or container deployments. This solution supports the ability to independently scale both compute and storage, and it features proven, validated interoperability between components for quick, easy deployment. FlashStack is virtual machine (VM) aware and hybrid cloud-ready, featuring the performance, scalability, and functionality enterprises look for when modernizing their IT infrastructure. FlashStack includes Cisco UCS server platforms, Pure Storage FlashArray storage, and Cisco Unified Fabric networking technology.
Cisco UCS is a modular system that combines computing, storage, and fabric interconnects to run bare metal and virtualized workloads. Cisco UCS ships as a stateless device that is programmable through the use of UCS service profiles. Service profiles are provisioning templates (which can be customer-defined and/or modified) that include server, storage, and networking configuration parameters that simplify and speed system deployment. A service profile is a software definition of a server and its LAN and SAN network connectivity — in other words, a service profile defines a server and its storage and networking characteristics. The service profiles reside in Cisco UCS — tools that automate the configuration delivery of compute, fabric extenders, and interconnects to match the selected service profile.
Cisco SingleConnect technology
SingleConnect technology provides an exceptionally easy, intelligent, and efficient way to connect and manage computing in the data centre. An exclusive Cisco innovation, SingleConnect technology dramatically simplifies the way that data centres connect to rack and blade servers; physical servers and virtual machines; and LAN, SAN, and management networks.
Cisco UCS fabric interconnects
Cisco UCS fabric interconnects provide a single point of connectivity and management for an entire Cisco UCS system. Deployed as an active-active pair, the system’s fabric interconnects integrate all components into a single, highly available management domain controlled by Cisco UCS Manager. Cisco fabric interconnects support low-latency, line-rate, lossless Ethernet and Fibre Channel over Ethernet (FCoE) connectivity.
Cisco UCS B-Series Blades
Cisco UCS B-Series Blade based on Intel Xeon processors and are available as either 2-socket servers or 4-socket servers. Systems can include GPUs to provide increased performance for big data and analytics and/or artificial intelligence/machine learning workloads. Cisco servers feature high-performance density as well as ENERGY STAR certification, integrated security (Cisco Secure Development Lifecycle), and fully programmable Virtual Interface Cards for maximum I/O infrastructure flexibility.
Cisco’s technology contribution offers three key differentiators for FlashStack customers. First, the integrated management makes the integrated configuration (which includes server, storage, and networking products) much easier to manage than infrastructure that is built from similar components from different vendors and enforces a siloed management approach. Second, compatibility of the separate components is ensured because all components have been pre-validated by Cisco and Pure Storage working together to create what is called a CVD.
Pure Storage FlashArray //X
FlashStack runs Pure Storage’s Purity operating environment, featuring a comprehensive set of enterprise-class data services including inline storage efficiency technologies (compression, deduplication, and thin provisioning), dynamic multipathing, dual-parity RAID, space-efficient snapshots, quality of service, encryption, and various replication (synchronous and asynchronous) options including support for stretch cluster configurations (which Pure Storage calls ActiveCluster). ActiveCluster configurations that follow Pure Storage best practices deliver “six-nines” (99.9999%) availability — that’s roughly five seconds of downtime per year.
