Cisco HyperFlex Data Platform
Hyperconverged infrastructure (HCI) combines compute, virtualization, storage, and networking in a single cluster. Starting with as few as three nodes, users can easily scale out to match computing and storage resource needs. Hyperconvergence brings cloudlike simplicity on-premises and within a single, easily managed platform.
Nowadays, the organizations must meet the challenges of growing data center requirements because of a wider set of application deployment models:
- Traditional enterprise applications: Monolithic components are transitioning to a more scalable, cloud-like model (for example, components such as presentation layers).
- Big data and analytics: Require scale-out architectures with large amounts of high-performance storage. Data scientists need massive amounts of GPU acceleration for artificial intelligence and machine learning processes.
- Cloud-native architectures: Based on microservices, cloud-native architectures require both virtualized and container-based deployment. Developers are creating these applications so they can run across multiple clouds with automatic deployment and scaling.
Building and integrating data center infrastructures that meet these versatile requirements is complex, challenging, and takes a lot of resources. Reducing this complexity is the only way to support existing and new applications, ensure service delivery, maintain control over data, and attain necessary performance. This makes the hyperconverged solutions the right answer because they deliver the compute, networking, storage, virtualization, and container support as one integrated and tested solution.
The Cisco HyperFlex System, shown in Figure 16-16, is an adaptive, hyperconverged system built on the foundation of the Cisco UCS. It meets the requirements for the different application deployments by supporting multicloud environments.
Figure 16-16 Cisco HyperFlex System
Cisco HyperFlex HX-Series servers combine software-defined networking and computing with the next-generation Cisco HyperFlex HX data platform.
With hybrid or all-flash memory storage configurations and a choice of management tools, Cisco HyperFlex systems are deployed as a pre-integrated cluster with a unified pool of resources that you can quickly provision, adapt, scale, and manage to efficiently power your applications and your business.
The Cisco HyperFlex platform is faster to deploy, simpler to manage, easier to scale, and ready to provide a unified pool of resources to power your business applications. You harness these resources with your choice of centralized management tools: Cisco Intersight cloud-based Management as a Service (covered in Chapter 21, “Cloud Computing”), Cisco HyperFlex Connect, Microsoft System Center Virtual Machine Manager, Hyper-V Manager, or a VMware vSphere plug-in. Cisco HyperFlex systems integrate into the data center you have today without creating an island of IT resources. You can deploy Cisco HyperFlex systems wherever you need them—from central data center environments to remote locations and edge-computing environments.
Cisco HyperFlex systems are based on Cisco UCS M5 rack servers. Based on Intel Xeon Scalable processors, these fifth-generation servers have faster processors, more cores, and faster and larger-capacity memory than previous-generation servers. In addition, they are ready for Intel 3D XPoint nonvolatile memory, which can be used as both storage and system memory, increasing your virtual server configuration options and flexibility for applications.
- Cisco HyperFlex HX240c M5/M6
- Cisco HyperFlex HX240c M5/M6 All Flash
- Cisco HyperFlex HX240c M5 HX240c M5 LFF
- Cisco HyperFlex HX220c M5/M6
- Cisco HyperFlex HX220c M5/M6 All Flash
- Cisco HyperFlex HX220c M5/M6 All NVMe
Physically, the system is delivered as a cluster of three or more of Cisco HyperFlex HX240c M5, HX240c M5 All Flash, HX240c M5 LFF, HX220c M5, HX220c M5 All Flash, or HX220c M5 All NVMe nodes that are integrated into a single system by a pair of Cisco UCS 6300 or 6400 Series Fabric Interconnects. The HX240c M5 node is excellent for high-capacity clusters, and the HX240c M5 All Flash node is excellent for high-performance, high-capacity clusters.
The Cisco HX240c M5 node is excellent for balanced-capacity clusters. The HX240c M5 LFF node delivers high-capacity clusters, and the HX240c M5 All Flash node is excellent for balanced-performance and capacity clusters. Each node configuration includes the following:
- Memory (up to 3TB of memory supported)
- Storage:
- All-flash memory, all-NVMe, or hybrid storage configurations (combination of HDDs and SSDs)
- Cisco 12Gbps modular SAS host bus adapter (HBA) with internal SAS connectivity
M.2 SATA SSD drive for boot
Data platform logging drive
- M.2 boot drive for VMware vSphere
- One Cisco UCS VIC
- Cisco UCS service profile templates for automated cluster configuration
- Software:
- Cisco HyperFlex HX Data Platform software (software subscription)
- VMware vSphere ESXi 6.0 software preinstalled (ESXi 6.5 supported but not preinstalled)
Cisco HyperFlex nodes can be deployed with various Cisco UCS B-series blade servers and C-series rack servers to create a hybrid cluster. With a single point of connectivity and management, you can easily scale your cluster to support more workloads and deliver the performance, bandwidth, and low latency your users and applications need.