Extend the benefits of Cisco UCS architecture and policy-based management to parallelized workloads with Cisco UCS M-Series servers. Our M-Series servers disaggregate components and use policy-based management to compose system resources for a wide variety of agile, Development and Operations (DevOps), cloud-scale, and new-style IT applications.

Today's new breed of cloud applications demand a different, better architecture: a composable infrastructure. Computing needs to be separate from infrastructure to increase density, deliver easier scale, and provide the best design for management, efficiency, and performance.

The unique design of the Cisco UCS M-Series eliminates the complexity of traditional servers by disaggregating the underlying component parts. Underutilized and overprovisioned local resources in traditional servers are now shared across multiple computing nodes.

Infrastructure resources are pooled and managed at the ID and logical-unit level so they can be composed as required through Cisco UCS management software and APIs to support dynamic application workloads.

These subcomponents include:

This separation of components by the M-Series server platform decouples the lifecycles of the component subsystems. You no longer need to replace an entire system just to update the processors, hard drives, or I/O. With Cisco UCS management software, resources can be pooled and dynamically managed across systems, racks, and even data centers.

With the M-Series you can optimize application performance by composing and scaling subsystems to achieve the best ratio of computing of I/O and storage in small, discreet increments. The design also allows for smaller increments of scale, so your applications have the number of computing nodes they need to achieve desired performance and availability.

No more overprovisioning. No more wasted resources. The M-Series’ groundbreaking design provides a highly dense, modular, and power-efficient platform that can meet the needs of parallelized workloads.

The Cisco UCS M-Series offers a variety of processors, including Intel® Xeon® E3-1200 v3 and v4 and Intel Xeon E5-2600 v3 and v4 series. Now it's much easier to choose the right processor for your workload, without compromising manageability or changing your processes.

The M-Series server system features a new, 2-rack-unit (RU) chassis built on Cisco Virtual Interface Card (VIC) technology called System Link. All Cisco UCS M4308 Modular Chassis connect to a pair of Cisco UCS Fabric Interconnects with Cisco UCS Manager, providing easy, fast scalability with industry-leading Cisco UCS management.

Cisco UCS Management provides management for the M-Series with consistency and simplicity in device management. The award-winning Cisco UCS Manager is a model-based, automated tool for element management, which allows for easy integration with higher-level tools using our open XML API. This streamlines operations and facilitates the transition to DevOps.

Cisco M-Series Modular Servers provide all the cost savings, ease of management, and scalability of a composable infrastructure solution.

• Improve total cost of ownership (TCO) and increase business agility
• Balance simplicity, performance, and server density for production-level virtualization, web infrastructure, and data center workloads
• Extend the benefits of Cisco Unified Computing and virtual interface card technologies to rack server environments

• Highly scalable computing with high-capacity local storage
• Designed from the ground up to offer flexibility in configuration and modularity
• Accommodates up to 60 high-capacity drives with dual server nodes in a 4RU form factor
• Ideal for deployment of big data applications, software-defined storage environments, backup and archiving, and content distribution

• Deliver up to 80 Gbps aggregate bandwidth
• Simplify with Cisco SingleConnect Technology and get one connection for LAN, SAN, and systems management
• Help ensure policy adherence and consistent network visibility down to the individual vNIC
• Reduce TCO by consolidating NICs, HBAs, cables, and switches
• Collapse networking infrastructure using Virtual Machine Fabric Extender (VM FEX)