Cisco HyperFlex Solution
The Cisco HyperFlex system is built on the Cisco UCS platform, providing a unified network, compute, and storage solution in a single solution for faster application deployment. The system consists of three or more Cisco HyperFlex HX-Series nodes to create a highly available cluster. Each node includes a Cisco HyperFlex HX Data Platform controller that implements the distributed file system using internal flash-based SSD drives and high-capacity HDDs or SSDs to store data. The controllers communicate with each other over 10 or 40 Gigabit Ethernet to present a single pool of storage that spans the nodes in the cluster. Nodes access data through a data layer using file, block, object, and API plug-ins. As nodes are added, the cluster scales linearly to deliver computing, storage capacity, and I/O performance.
What’s New in HyperFlex 3.5
There are many new features with HyperFlex 3.5. Here are a few that stand out to me.
- New to HyperFlex 3.5 is support for All-NVMe with Intel Optane SSDs. The system uses the ultra-fast Intel Optane NVMe drives for caching and Intel 3D NAND NVMe drives for capacity storage.
- Customers creating applications for Artificial intelligence and machine learning (AI/ML) will be pleased with the new integration of NVIDIA Tesla V100 GPUs within the HyperFlex nodes. There is the option of installing 2 NVIDIA P100 or V100 GPUs per Node as well as an option for GPU Only Nodes.
- Simplified 1-Click Disaster Recovery Readiness with VM Replication, Migration, and Test Readiness.
- Real-Time monitoring and predictive intelligence through the Cloud-Managed Intersight solution for HyperFlex and UCS.
HyperFlex Fabric Accelerated AI/ML
Virtualizing GPUs via a high-performance fabric is now possible with the Cisco HyperFlex solution. It supports for low latency and high throughput scenarios.
Cisco's convergence of compute, network and storage with centralized management is what stands out about this new Hyperflex solution. Other solutions out there have compute and storage, but the network is still a component managed separately, which opens competitive and interesting ways to leverage a Hyperflex solution versus some of the competition out there.
Written By: Marshall Hill, LookingPoint Solutions Architect