Cisco UCS C4200 multinode rack server deeper dive
KD’s (VP, Product Mgt.) blog discusses why Cisco is now offering a multinode rack server. The short answer to why is the increasing need to support scale out, compute intensive workloads which require increased node density in a small form factor. I’m here to dig a little deeper into the hardware.
The new C-Series multinode rack server is comprised of the chassis and server nodes.
Cisco UCS C4200 Series Rack Server Chassis
The chassis is 2RU and supports four server nodes. There are 24 front loading, hot plug, SFF drive bays. Each of the four nodes is allocated six drive bays – either all SAS/SATA or two NVMe plus four SAS/SATA. It has dual 2400W power supplies providing plenty of power for the beefiest configurations.
Cisco UCS C125 M5 Rack Server Node
The C125 M5 is Cisco’s first server based on AMD EPYC™ processors. Why EPYC CPUs for the C125 M5? For dense environments, performance per watt is a key metric. EPYC processors offers the most cores with lower power consumption than other offerings making it an ideal choice for a multinode server. The key features of the C125 M5 are:
- Up to two AMD EPYC CPUs (up to 32 cores per socket); up to 64 cores per node, up to 256 cores per C4200 Chassis (4 server nodes)
- 2666-MHz DDR4 memory, 16 DIMM slots (8 per CPU socket) for up to 2 TB of capacity using 128-GB DIMMs
- Eight DIMMs per channel providing 33% better memory bandwidth than other M5 servers
- 4th Gen VICs
- Two PCIe 3.0 slots
- Flexible OCP 2.0 mezzanine slot supporting 100Gbps adapters
- Cisco 12G 9460-8i PCIe SAS RAID Controller with 2GB FBWC
- Optional dual SD cards or dual M.2 SATA/NVMe drives for boot or increased storage capacity
If you are going to be at Cisco Live in June, the C4200 and C125 M5 will be on display in the AMD booth (#835) and in Cisco’s Platform for Digital Business area of the World of Solutions.
Scale with enterprise management
While speeds & feeds are nice, they are not the most important thing to consider when purchasing servers for scale out workloads. The real question is how quickly a new server can be deployed to deliver results and how much additional time and resources you will need to manage it. Other server vendors offer multinode servers, only Cisco offers you a system. What do I mean by a system? At its core, UCS is a fabric-centric architecture with a centralized management model. All of Cisco’s servers: B-Series blades, C-Series rack, S-Series dense storage servers, and HyperFlex are manageable with a single tool, Cisco Intersight. Intersight is cloud-based system management platform augmented by analytics and machine learning. It enables organizations to achieve a higher level of automation, simplicity, and operational efficiency. It provides a holistic and unified approach to managing distributed computing environments regardless of the server form factor, workload, or location. Only Cisco can deliver a scale out computing solution delivered as part of an integrated system designed to maximize performance, not complexity. This isn’t just marketing hype, this is what customers are saying – deploying 83% faster and 62% less server management.
Do you have to use Cisco Intersight? Of course not. Like all C-Series rack servers, the C125 M5 has a Cisco IMC (Integrated Management Controller). You can manage the server directly through the CIMC, through its API, or industry-standard management protocols including Redfish and SNMP.
If you want to learn more about the our new multinode system, contact your Cisco sales team or Cisco partner to discuss your scale out, compute intensive workloads and how the UCS C4200 and UCS C125 M5 server node can meet your needs.
You can also find more information on the C4200 & C125 M5 at www.cisco.com/go/ucs4200.
For more information on the entire Cisco UCS portfolio, go to www.cisco.com/go/ucs.