Hubbry Logo
Cisco Unified Computing SystemCisco Unified Computing SystemMain
Open search
Cisco Unified Computing System
Community hub
Cisco Unified Computing System
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Cisco Unified Computing System
Cisco Unified Computing System
Cisco Unified Computing System
View on Wikipedia
from Wikipedia

Cisco Unified Computing System (UCS) is a data center server computer product line composed of server hardware, virtualization support, switching fabric, and management software, introduced in 2009 by Cisco Systems.[1][2] The products are marketed for scalability by integrating many components of a data center that can be managed as a single unit.[3]

Computing

[edit]
Cisco UCS blade servers
Cisco UCS C250 servers

The UCS product line was announced in March and first installed in September 2009, at the customer premises of Tutor Perini.[2] The first customer purchase of UCS was by Fiserv, a financial services customer in Milwaukee, Wisconsin. The entire project was code named "California".[4]

The computing component of the UCS is available in two versions: the B-series (a powered chassis and full and/or half slot blade servers), and the C-series for 19-inch racks (that can be used with fabric interconnects). The computer hardware managed by the UCS Manager software on the fabric Interconnects can be any combination of the two. Both form factors use standard components, including Intel x86-64 processors and DIMM memory. The servers are marketed with converged network adapters and port virtualization.

Around 2010, an extended memory technology expanded the number of memory sockets that can be connected to a single memory channel in some models.[5][6]

A fifth generation was announced in July 2018.

Virtualization

[edit]

Cisco UCS supports several hypervisors including VMware ESXi, Microsoft Hyper-V and Citrix Systems' Xen server.[7] VMware provides a special version of ESXi.

In 2017 Cisco announced a partnership with Docker to include its enterprise software in UCS products to directly provide operating-system-level virtualization (containerization).[8]

Networking

[edit]

The Cisco fabric interconnects (FI) provide network connectivity for the chassis, blade servers and rack servers connected to it through different speeds of Ethernet and Fibre Channel over Ethernet (FCoE). The fabric interconnects are derived from the Nexus 5500 and 9300 series switch families and run NX-OS as well as UCS Manager software. FCoE allows connection to storage area network storage, To allow stateless compute in addition to local storage capacity. Cisco has produced the following series for fabric interconnects: 6100 Series (discontinued), 6200 Series (discontinued), 6300 Series, and 6400 Series. The fabric interconnect can further connect to multiple fabric extenders (FEX), into what they call a "unified fabric".[4]

By 2012, analysts noticed that Fibre Channel vendors such as Brocade Communications Systems had already lost market share to Cisco.[9]

Management

[edit]

Management of the system devices is handled by the UCS Manager software integrated with fabric interconnects.[10] Virtual machines can be moved from one physical chassis to another, applications may be moved between virtual machines, and management may even be conducted remotely from an iPhone using SiMU - Simple iPhone Management of UCS. In addition to the embedded software, administrators may also manage the system from VMware's vSphere. The Cisco Integrated Management Controller (CIMC) is used to configure and manage C-Series servers not connected to a manager. The CIMC can also be used to manage B-series blades in addition to the UCS Manager application if configured.

In November 2012, Cisco announced UCS Central which extends management across multiple domains of UCS,[11] and then something called UCS Director. A security flaw in UCS Central was disclosed when it was patched in 2016.[12]

Many configuration details are set up on the UCS manager in a service profile and applied to the servers. Cisco offers a UCSM Platform Emulator, where the full logical configuration of a server can be created from the user interface or the API methods, and later applied to the physical hardware.

In 2017, Cisco announced a cloud-hosted management offering for UCS infrastructure called Cisco Intersight. This platform augmented the UCS management platform.

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Cisco Unified Computing System

View on Grokipedia
from Grokipedia
The Unified Computing System ( UCS) is an integrated designed to unify servers, networking, and storage access within a single system, providing intent-based management for automating application deployments and accelerating business operations. Launched in , it introduced a revolutionary architecture that simplifies by reducing hardware complexity and enabling stateless , where server configurations are abstracted into software policies. At its core, UCS features three primary components: unified fabric, unified , and unified resources. The unified fabric leverages Cisco SingleConnect technology to consolidate LAN, SAN, and traffic over a single set of low-latency, high-bandwidth cables, supporting speeds up to 100 Gbps through fabric interconnects like the UCS 6536, 6664, and 6600 series models. Fabric extenders and programmable Virtual Interface Cards (VICs) further distribute connectivity and I/O resources, allowing up to 512 virtual PCIe devices per server for flexible workload handling. Unified is powered by tools such as UCS Manager for on-premises policy-based automation and Cisco Intersight, a cloud-based SaaS platform that provides role-based lifecycle and global scalability across multiple data centers. The resources encompass diverse server form factors, including blade servers (UCS B-Series), rack servers (UCS C-Series), modular systems (UCS X-Series), and storage servers (UCS S-Series), all supporting processors like EPYC and for workloads ranging from to AI inferencing. This architecture promotes a "wire-once" model, where capacity is provisioned once and dynamically allocated via software, enhancing with active-active and reducing through device consolidation. Key benefits include improved IT productivity via , greater for rapid resource repurposing, and gains that have positioned UCS as a top-tier server vendor, with over 200 world performance records and adoption by 85% of companies as of 2019. Recent innovations, such as the 6th generation UCS components including M8 servers and 6600 Series Fabric Interconnects, along with AI-ready and validated AI PODs, extend its capabilities to modern demands like and hybrid environments as of 2025.

Introduction and History

Overview

The Unified Computing System (UCS) is a next-generation platform that integrates industry-standard x86-architecture servers—encompassing both and rack form factors—with a unified fabric for networking and storage access, alongside centralized management infrastructure to streamline operations. This architecture employs a low-latency, lossless unified network fabric to connect computing resources, enabling scalable, multi-chassis deployments that abstract hardware dependencies for greater flexibility. By fusing these elements, UCS provides end-to-end visibility and control, reducing the silos typically found in traditional environments. At its core, Cisco UCS operates on principles of stateless computing, where server identity, configuration, and connectivity are abstracted into software rather than tied to physical hardware, allowing for rapid reprovisioning without manual reconfiguration. Central to this is the use of service profiles, which encapsulate policies for , enabling policy-based abstraction and consistent deployment across servers. These features, combined with unified management tools like UCS Manager, deliver comprehensive visibility into the entire infrastructure, from compute to fabric. Introduced in 2009 as Cisco's entry into the server market, UCS was positioned to simplify by consolidating networking and computing, thereby lowering operational complexity and (TCO) through reduced cabling, power, and administrative overhead. Key benefits include to manage thousands of servers in a single domain, built-in automation for faster provisioning, and robust support for diverse workloads such as , cloud-native applications, and AI-driven analytics. This approach enhances by enabling intent-based operations that align infrastructure with application needs.

Development and Milestones

The Unified Computing System (UCS) originated from internal development under the codename "Project California," announced on , , as a revolutionary architecture integrating , networking, storage access, and into a unified platform. The initial products, including the UCS 6100 Series Fabric Interconnects and B-Series servers, began shipping in July , marking 's entry into the server market with a focus on simplifying infrastructure management. The first generation of UCS (2009–2012) emphasized blade servers like the B200 M1 and C200 M1 models, establishing foundational unified fabric capabilities with support. In 2012, the second generation introduced enhanced networking with via the UCS 6200 Series Fabric Interconnects and M3 servers, coinciding with milestone achievements such as world-record performances in VMmark 2.1 virtualization benchmarks and TPC-C tests. Subsequent generations evolved rapidly: the third (2013–2014) with M3/M4 servers and 6300 Series interconnects for broader scalability; the fourth (2014–2016) incorporating E5-2600 v3 processors; and the fifth generation launched in July 2017, featuring M5 servers optimized for data-intensive workloads with support for up to . By the sixth generation in the , UCS incorporated PCIe Gen5 connectivity and AI-specific optimizations in M6 and later models, such as the UCS 6600 Series Fabric Interconnects, enhancing performance for and hybrid environments. Key evolutionary influences included strategic acquisitions that bolstered UCS capabilities. The December 2013 acquisition of Insieme Networks introduced Application Centric Infrastructure (ACI), which integrated seamlessly with UCS Manager to enable policy-based and end-to-end visibility across compute and network domains. Similarly, the March 2017 acquisition of added advanced application performance analytics, allowing real-time monitoring and optimization of UCS-hosted workloads through extensions that track hardware health and service profiles. UCS had achieved widespread adoption, with over 85% of companies utilizing the platform as of 2019 and more than 50,000 customers worldwide as of 2021 benefiting from its streamlined operations. A pivotal milestone in 2025 was the September 2 release of UCS Manager 6.0, which introduced support for the UCS 6664 Fabric Interconnects and enhanced hybrid cloud integration for scalable, policy-driven management across on-premises and multi-cloud environments.

Architecture and Components

Unified Fabric

The Cisco Unified Computing System (UCS) Unified Fabric serves as the foundational networking layer that converges (LAN), (SAN), and management traffic into a single, high-performance Ethernet-based infrastructure. This design eliminates the need for multiple discrete networks by leveraging a unified set of protocols and hardware, enabling efficient operations. At the core of this fabric are the Fabric Interconnects (FIs), which function as top-of-rack switches providing connectivity at speeds of 10, 25, 40, and (GbE), supporting low-latency, line-rate performance for diverse workloads. The pin architecture of the Unified Fabric ensures deterministic traffic paths through a dual redundant FI setup, where two FIs operate in an active-active configuration to provide northbound connectivity to upstream networks. This facilitates automated and , with each FI connecting directly to the parent domain's access-layer switches via unified ports that support both Ethernet and protocols. By integrating connectivity and management at the FI level, the architecture eliminates the requirement for traditional top-of-rack or switches within the UCS domain, reducing latency by a single network hop and simplifying cabling and management overhead. To achieve a lossless Ethernet environment, the Unified Fabric incorporates Data Center Bridging (DCB) enhancements, including IEEE 802.1Qbb Priority Flow Control (PFC) and IEEE 802.3x PAUSE mechanisms, which prevent during congestion for both LAN and storage traffic. (FCoE) is natively supported, allowing SAN traffic to traverse the same Ethernet infrastructure without dedicated switches, thereby consolidating I/O resources and lowering operational costs. Scalability is a key attribute of the Unified Fabric, enabling support for up to 160 servers per domain through the integration of fabric extenders (FEX) and I/O modules (IOM), such as the UCS 6300 series FEX. These components extend the fabric's reach to and rack servers without introducing additional complexity, delivering high aggregate bandwidth per while maintaining the unified . Larger scales beyond a single domain are achievable with Intersight.

Computing Resources

The Cisco Unified Computing System (UCS) employs a stateless computing model, where servers function as undifferentiated pools of central processing units (CPUs), , and (I/O) resources that are abstracted and managed through service profiles. This approach decouples the server's physical hardware from its logical configuration, enabling rapid provisioning and reprovisioning of resources without manual reconfiguration or stateful dependencies on individual hardware components. Service profiles encapsulate the server's identity, , settings, and connectivity policies, allowing identical configurations to be applied dynamically to any compatible server in the pool, which enhances scalability and reduces operational overhead in environments. UCS computing resources support a range of high-performance processors, including up to two 6th-generation Intel Xeon Scalable processors or 5th-generation AMD EPYC processors per server as of 2025, providing robust compute capabilities for demanding workloads. Memory configurations scale up to 8 TB of DDR5 per server across 32 DIMM slots, utilizing registered DIMMs (RDIMMs) or load-reduced DIMMs (LRDIMMs) at speeds up to 6400 MT/s, optimized for memory-intensive applications. Additionally, UCS integrates GPU and accelerator support, such as NVIDIA RTX Pro 6000 Blackwell GPUs and Intel Flex Data Center GPUs, enabling accelerated processing for artificial intelligence and machine learning tasks within the unified infrastructure. Resource pooling in UCS allows servers to be organized into logical pools based on attributes like capacity and performance, facilitating automated allocation without physical intervention. Upon insertion into a or connection to the fabric interconnect, servers are automatically discovered and inventoried by UCS Manager, which applies predefined policies from service profile templates to instantiate the server with the appropriate configuration. This auto-discovery process ensures seamless integration into the resource pool, supporting dynamic workload placement and maintaining through policy-driven association. To optimize energy efficiency, UCS incorporates power capping features that limit maximum power consumption per server or across the domain, preventing overloads while dynamically adjusting based on demands. management capabilities include adjustable processor (TDP) values and thermal failure power capping, which enforces power limits when inlet temperatures exceed specified thresholds, thereby reducing cooling requirements and overall energy use. These mechanisms integrate with the unified fabric for holistic power budgeting, contributing to lower operational costs in large-scale deployments.

Management Infrastructure

The management infrastructure of the Cisco Unified Computing System (UCS) employs a hierarchical model based on organizations, enabling centralized control over the UCS domain. This structure starts with a root organization and allows the creation of sub-organizations to logically partition resources, , and access in multi-tenant environments. Domain-level defined at higher organizational levels propagate downward to and individual servers, ensuring consistent configuration and across the . For instance, when resolving policy names or resource pools, UCS Manager searches hierarchically from the target organization's level upward until a match is found. Central to this infrastructure are service profiles, which serve as XML-based templates that encapsulate a server's complete identity and configuration. Each service profile defines key attributes such as the server's UUID, MAC addresses for virtual network interface cards (vNICs), World Wide Names (WWNs) for host bus adapters (HBAs), versions, settings, and network connectivity policies. These profiles abstract hardware details, allowing stateless where servers can be rapidly provisioned, updated, or migrated by associating or disassociating the profile without manual reconfiguration. Service profiles support two types: those that override inherited hardware identity for flexibility and those that inherit directly from hardware for fixed assignments. Monitoring within the UCS management infrastructure relies on embedded analytics provided by the Data Management Engine (DME), which tracks the health, faults, and of physical and logical components in real time. The DME maintains an of system state, generating faults for critical issues (e.g., hardware failures) and collecting performance metrics such as CPU utilization, usage, and I/O throughput for servers, , and interconnects. Faults are categorized by severity—critical, major, minor, or warning—and can be viewed through integrated logs, including system event logs (SEL) for hardware events and logs for configuration changes. Threshold policies enable proactive alerts, while tools like SNMP and statistics collection support detailed diagnostics without external dependencies. The ecosystem facilitates and third-party integration through the UCS Manager XML , a programmatic interface that uses HTTP/ to interact with the system's Management Information Tree (MIT). This supports operations like authentication, querying managed objects, and configuring modifications via methods such as configConfMo for individual objects or configConfMos for bulk changes. It also includes RESTful capabilities through the UCS Python SDK, enabling scripted workflows for provisioning and monitoring. For example, the integrates with orchestration tools like , allowing automated deployment of service profiles and policy enforcement across UCS domains.

Hardware Elements

Blade Servers

The Cisco Unified Computing System (UCS) B-Series blade servers serve as dense, shared-infrastructure compute nodes designed for high-performance data center environments. These servers integrate seamlessly into UCS chassis, enabling efficient resource utilization through stateless computing and unified management. Blade servers in the UCS B-Series adopt half-width or full-width form factors, accommodating up to eight half-width blades or four full-width blades per chassis in models such as the UCS 5108 Blade Server Chassis, which occupies 6 rack units (RU). The UCS 5108 chassis supports flexible partitioning with removable dividers to mix half- and full-width configurations, providing up to 80 Gbps I/O bandwidth per half-width blade and 160 Gbps per full-width blade. Across a UCS domain managed by UCS Manager, this scalability allows for up to 160 blade servers, typically achieved with 20 chassis. Representative models from the 2020s include the UCS B200 M6, a half-width equipped with up to two 3rd-generation Scalable processors offering as many as 40 cores per socket. It supports up to 32 DDR4 slots for a maximum of 16 TB of memory at 3200 MHz, along with optional Optane for up to 12 TB. Storage options feature up to two small form factor (SFF) NVMe or SSD drives (up to 7.7 TB each) or four drives (up to 960 GB each), with controllers for redundancy. Shared resources in the chassis include I/O Modules (IOMs), such as the UCS 2408 or 2304 series, which extend the unified fabric to blades without onboard switches, delivering up to 80 Gbps aggregate I/O via Virtual Interface Cards (VICs). Power and cooling are redundantly provisioned through up to four hot-swappable 2500W power supplies and eight fan modules per , ensuring or N+N for reliability in dense deployments. These servers excel in high-density environments, optimizing space and power for platforms like and (HPC) applications, where their shared infrastructure reduces cabling and operational overhead compared to standalone rack servers.

Rack Servers

The UCS C-Series rack servers provide flexible, standalone options within the Unified Computing System, designed for deployments requiring independent and customization outside of environments. These servers support a range of form factors to accommodate diverse workloads, from high-density general-purpose to storage-intensive or GPU-accelerated applications. Typical configurations include 1U models such as the UCS C220 M8 and C225 M8, which deliver compact performance for edge or tasks with up to two Xeon 6 or processors, respectively. Larger 2U form factors, exemplified by the UCS C240 M8 (Intel-based) and C245 M8 (AMD-based), offer enhanced storage and expansion capabilities, supporting up to 28 small form-factor drives and configurations optimized for or . For GPU-heavy workloads, the C-Series extends to 4U models like the UCS C845A M8 and 8U models such as the UCS C885A M8, which integrate up to eight H100/H200 or MI300X GPUs alongside dual processors, enabling accelerated AI training and inference. These form factors prioritize modularity, with options for up to eight PCIe Gen5 slots in select models to accommodate additional network adapters, storage controllers, or accelerators. Connectivity in C-Series servers emphasizes unified I/O through direct attachment to UCS Fabric Interconnects via Virtual Interface Cards (VICs), which consolidate Ethernet, , and FCoE traffic over a single . The Modular LAN on (mLOM)/OCP 3.0 slot enables installation of VIC 15000 Series —such as the quad-port 100 Gbps VIC 15427—without consuming a PCIe slot, preserving expansion flexibility for other components. This design supports dynamic virtual interfaces, allowing up to 256 PCIe-compliant endpoints per server for . PCIe expansion further enhances modularity, with slots compatible for third-party GPUs, NICs, or HBAs to tailor I/O to specific needs. As of 2025, C-Series updates focus on AI readiness, integrating 4th and 5th Gen processors—offering up to 192 cores per socket in models like the C245 M8 and C885A M8—for superior parallel processing in tasks. These enhancements deliver up to 20% better performance in SPEC benchmarks compared to prior generations, emphasizing energy efficiency with advanced architectures. Liquid cooling options have also emerged for high-density AI configurations, through partnerships enabling direct-to-chip solutions in GPU servers to manage thermal loads exceeding 1 kW per component, reducing power consumption by up to 40% in dense racks.

Fabric Interconnects

The Cisco Unified Computing System (UCS) Fabric Interconnects serve as the foundational switching infrastructure, providing unified connectivity for computing, networking, and storage resources within the UCS domain. These top-of-rack switches integrate Ethernet, , and capabilities, enabling a lossless, low-latency fabric that supports server attachments via direct cabling or through fabric extenders. The Fabric Interconnects have evolved across multiple generations, starting with the 6200 Series introduced in 2010, which offered support with up to 1.92 terabits per second (Tbps) throughput in the UCS 6296UP model featuring 96 ports. The 6300 Series, launched in 2015, advanced to 40 GbE with models like the UCS 6332 providing 32 fixed QSFP+ ports and up to 2.56 Tbps throughput, while also supporting 16-Gbps . Subsequent generations include the 6400 Series from 2019, with the UCS 64108 delivering up to 7.42 Tbps throughput across 108 ports (96x 10/25 GbE SFP28 and 12x 40/100 GbE QSFP28), and the 6500 Series from 2021, exemplified by the UCS 6536 with 36x 100 GbE QSFP28 ports also achieving 7.42 Tbps per unit. The 6600 Series, released in 2025, further advances with the UCS 6664 offering up to 11.65 Tbps throughput across 64 ports (16 unified ports supporting 10/25/50 Gbps Ethernet or 16/32/64 Gbps , and 48 QSFP ports for 40/100 Gbps Ethernet). In clustered configurations, these interconnects scale to aggregate throughputs exceeding 14 Tbps per domain, with port speeds reaching 100 GbE across recent models.
GenerationKey ModelsMax PortsMax Throughput per UnitPort Speeds
6200 SeriesUCS 6248UP, UCS 6296UP961.92 Tbps10 GbE, FCoE, 8-Gbps FC
6300 SeriesUCS 6324, UCS 6332322.56 Tbps40 GbE, FCoE, 16-Gbps FC
6400 SeriesUCS 6454, UCS 641081087.42 Tbps10/25/40/100 GbE, FCoE, 32-Gbps FC
6500 SeriesUCS 6536367.42 Tbps10/25/40/100 GbE, FCoE, 32-Gbps FC
6600 SeriesUCS 66646411.65 Tbps10/25/40/50/100 GbE, FCoE, 16/32/64-Gbps FC
Redundancy is achieved through active/active clustering of paired Fabric Interconnects, where each pair forms a high-availability domain connected via dedicated L1 and L2 Ethernet ports for , supporting up to 108 ports in the highest-density configurations like the UCS 64108. This setup ensures management plane redundancy in an active/passive mode, while data traffic operates in active/active fashion to maintain continuous connectivity. Key features include support for VN-Tag (Virtual Network Tag) technology, which enables by tagging packets with unique identifiers for virtual NICs (vNICs) and virtual HBAs (vHBAs), allowing up to 256 virtual interfaces per physical port without limitations. Integrated FCoE support converges storage and data networking, providing lossless Ethernet for traffic across all generations, with speeds scaling from 8 Gbps in early models to 32 Gbps in recent ones. Physically, the Fabric Interconnects feature a in earlier generations, such as the 6200 Series, which uses expansion modules to increase port density from 40 to 80 ports via hot-swappable SFP+ modules for Ethernet and . Later series like the 6300 and beyond adopt fixed-port architectures with high-density QSFP/QSFP28 transceivers and support for breakout cables to achieve effective densities up to 384x 25 GbE ports on models like the UCS 64108, housed in 1-2 (RU) form factors with redundant power supplies and fans.

Software and Integration

UCS Manager

Cisco Unified Computing System (UCS) Manager is the centralized, on-premises management software that provides unified control over all hardware and software components within a UCS domain, enabling administrators to configure, monitor, and operate the system from a single interface. It runs as an embedded application within the Fabric Interconnects, the core networking and management switches of the UCS architecture, allowing for through a clustered pair of interconnects that form a single logical management entity. This deployment model supports scalability for a UCS domain encompassing up to 20 and hundreds of servers, with management traffic handled internally via the fabric without requiring external servers. UCS Manager offers multiple interfaces for interaction, including a (GUI) for visual configuration and monitoring, a (CLI) accessible via SSH for scripted and detailed operations, and a programmatic XML over HTTP/ for and third-party integrations. The GUI, evolved to an HTML5-based version in recent releases, facilitates intuitive navigation through hierarchical views of the UCS topology, while the CLI supports both interactive sessions and batch scripting for advanced troubleshooting. The enables policy-based , allowing developers to create custom tools for tasks like server discovery and configuration deployment. Key capabilities of UCS Manager include policy creation for defining server configurations, such as settings, network connectivity, and storage mappings, which are applied via service profiles to enable stateless computing and rapid reprovisioning. Server provisioning is streamlined through automated discovery and association of blades or rack servers to , with templates ensuring consistent deployment across the domain. Fault isolation is achieved via integrated monitoring tools that detect hardware issues, generate alerts, and provide diagnostic logs, facilitating proactive resolution without manual intervention. Introduced with the UCS platform in 2009 at version 1.0, UCS Manager has evolved through releases up to 6.0 in 2025, incorporating enhancements like support for advanced fabric interconnects and improved features. Early versions focused on basic unified management, while later iterations, such as 4.x and beyond, added automated provisioning workflows that reduce deployment times by up to 85%. Release 6.0 introduces capabilities like boot support and enhanced on uplinks, building toward more intelligent operations. Additional features encompass firmware orchestration, where UCS Manager coordinates upgrades across all domain components using bundle files to minimize downtime. Backup and restore operations allow exporting configuration snapshots to network locations for disaster recovery, supporting both full-state and logical backups. Integration with LDAP and Active Directory provides centralized authentication, enabling role-based access control synced from enterprise directories. UCS Manager can integrate with cloud-based platforms like Intersight for hybrid management across on-premises and remote domains.

Intersight Platform

The Cisco Intersight platform is a software-as-a-service (SaaS) solution that extends the capabilities of the Cisco Unified Computing System (UCS) to hybrid and multi-cloud environments, providing centralized and control over distributed . It enables IT teams to monitor and manage UCS domains, HyperFlex systems, and other infrastructure from a single, cloud-based , eliminating the need for on-premises appliances. This approach supports global operations across data centers, edge locations, and public clouds, facilitating scalable IT delivery without hardware dependencies. Intersight's SaaS model delivers comprehensive device visibility through real-time inventory and data collection from connected UCS instances, allowing administrators to assess hardware health, configurations, and performance metrics across multiple sites. , powered by algorithms, analyze this data to forecast potential failures, identify vulnerabilities, and recommend proactive remediations, enhancing operational reliability. workflows, including a drag-and-drop designer, streamline tasks such as policy deployment and resource provisioning, integrating with tools like and Terraform for orchestrated IT operations. As of 2025, Intersight has incorporated enhanced AI and features, particularly in Intersight Workload Optimizer, which dynamically balances for optimal performance and efficiency in hybrid cloud setups while assessing security posture through threat detection and compliance insights. Recent updates include support for advanced operating systems like Windows Server 2025 and improved recommendation engines that leverage ML for resource self-optimization and vulnerability prioritization. These capabilities integrate seamlessly with UCS Manager for hybrid management of on-premises resources, HyperFlex for hyperconverged deployments, and third-party platforms such as vCenter via dedicated plugins that enable inventory monitoring and proactive hardware alerts. The platform's benefits include simplified global management of distributed UCS environments, reduced operational overhead through cloud-native scalability, and enhanced security via encrypted data transmission and role-based access controls, all without requiring local hardware installations. By unifying insights and , Intersight helps organizations achieve faster issue resolution and cost savings in multi-site deployments.

Firmware Management

Firmware management in the Cisco Unified Computing System (UCS) encompasses the processes for updating and maintaining across hardware components such as fabric interconnects (FIs), I/O modules (IOMs), and servers to ensure compatibility, , and . These updates are orchestrated to minimize operational disruptions in environments, leveraging centralized tools for validation and deployment. Auto-upgrade policies enable orchestrated firmware updates through service profiles, which associate host firmware packages to UCS endpoints, allowing deferred or scheduled deployments to reduce . In clustered environments, these policies facilitate staged upgrades—first updating infrastructure firmware (FIs and UCS Manager), followed by chassis and servers—ensuring by evacuating traffic from the subordinate FI before proceeding. This approach supports seamless integration with management platforms like UCS Manager and Intersight for policy enforcement. Bundle management involves downloading and applying HCL-verified bundles specific to UCS components, including bundles (A-bundles) for FIs, IOMs, and UCS Manager; server bundles (B-bundles for blades, C-bundles for racks); and capability catalogs (T-bundles) for hardware descriptors. These bundles ensure , with providing compatibility matrices to validate cross-version support before upgrades. As of 2025, UCS bundles incorporate support for secure boot and TPM 2.0, enhancing server through protection and hardware-based root of trust in secured-core configurations. Tools such as UCS Manager pools and the Intersight platform provide capabilities for scheduling upgrades, performing rollbacks via backup slots, and conducting compliance checks against and pre-upgrade validation tools. In UCS Manager, the Auto Install feature sequences updates while Intersight supports managed-mode upgrades for fabric and server components, including direct firmware image selection and reboot . Rollbacks are automated if a startup version fails, restoring from secondary bootflash partitions. Best practices for firmware management emphasize pre-upgrade preparations, including backing up configurations (XML and full-state files), resolving critical faults, syncing NTP across FIs, and ensuring at least 20% free space in bootflash and workspace partitions. In clustered setups, staged upgrades minimize by upgrading one FI at a time, with validation using the UCS Manager and Pre-Upgrade Check Tool to confirm endpoint functionality and compatibility.

Networking and Connectivity

I/O Fabric Design

The I/O fabric design in the Unified System (UCS) unifies network and storage traffic through a converged that eliminates the need for dedicated switches, enabling efficient data flow across servers and interconnects. Central to this design are the Virtual Interface Cards (VICs), which serve as high-performance adapters integrated into UCS and rack servers. These adapters, such as the UCS VIC 15000 series, support up to 512 PCIe virtual devices, including virtual network interface cards (vNICs) and virtual host bus adapters (vHBAs), allowing a single physical card to present hundreds of logical interfaces to the operating system for flexible workload provisioning. The UCS I/O fabric supports a range of protocols to handle diverse traffic types, including Ethernet at speeds from 10 to 200 Gbps, (FCoE) for converged storage networking, for IP-based block storage, and RDMA over Converged Ethernet version 2 (RoCEv2) for low-latency, high-throughput applications such as AI and workloads. RoCEv2, in particular, enables between servers with minimal CPU overhead, supporting protocols like NVMe over Fabrics and SMB Direct on operating systems including Windows, , and VMware ESXi. This protocol integration occurs over a single lossless Ethernet fabric, with physical connectivity provided via Fabric Interconnects that aggregate traffic from VICs. Quality of service (QoS) and traffic management in the UCS I/O fabric ensure reliable performance under load through mechanisms like Priority Flow Control (PFC) and Enhanced Transmission Selection (ETS). PFC, defined in IEEE 802.1Qbb, provides per-priority pausing to create a lossless Ethernet environment, preventing frame drops for sensitive traffic such as FCoE and RoCEv2 by sending pause frames when buffers approach congestion thresholds. ETS, per IEEE 802.1Qaz, allocates minimum bandwidth guarantees to traffic classes (e.g., 20% for storage and 80% for Ethernet), dynamically adjusting during low utilization to maximize throughput while avoiding over-subscription. These features are applied via UCS Manager policies to vNICs and vHBAs, with all internal links supporting PFC for end-to-end consistency. Scalability in the UCS I/O fabric reaches up to 200 Gbps of bandwidth per server in the 5th generation architecture, leveraging PCIe Gen4 interfaces in UCS 15000 Series VICs for M6/M7 servers across B-, C-, and X-Series platforms. This design supports non-oversubscribed chassis bandwidth of 1.6 Tbps in systems like the UCS X9508, facilitating dense deployments for AI inferencing and data analytics without bottlenecks. In UCS Manager Release 6.0 (September 2025), enhancements include support for the UCS 6664 Fabric Interconnect, offering up to 11.65 Tbps throughput and advanced features like boot over , MACsec encryption on uplinks, and ERSPAN for traffic monitoring, further improving networking and .

Storage Networking

The Unified Computing System (UCS) integrates storage networking through its unified fabric architecture, enabling that supports block storage protocols over Ethernet without requiring dedicated (FC) infrastructure. This approach leverages (FCoE) to encapsulate FC frames within Ethernet packets, allowing servers to access (SAN) resources via a single network fabric. UCS Fabric Interconnects natively support FCoE on Ethernet ports, facilitating seamless connectivity to upstream FCoE-enabled switches like the Nexus series, which reduces cabling complexity by consolidating LAN and SAN traffic on 10/25/40/ links. UCS also provides native support for NVMe over Fabrics (NVMe-oF), extending the high-performance NVMe protocol across the network to enable low-latency block storage access. This includes NVMe/FC for direct FC integration and Ethernet-based transports such as RoCEv2, with UCS C-Series and B-Series servers supporting NVMe-oF initiators on VIC adapters starting from the M5 generation. By embedding NVMe-oF capabilities in the fabric interconnects, UCS allows for disaggregated storage architectures where compute nodes can directly attach to remote NVMe storage arrays, improving scalability for data-intensive workloads. Boot policies in UCS Manager enable stateless SAN booting, where virtual Host Bus Adapters (vHBAs) are configured within service profiles to initiate boot from LUNs on external SAN storage. Administrators define boot order in the policy, specifying primary and secondary SAN targets via WWNs or IQNs, and UCS automatically handles and connectivity through the UCS domain, ensuring consistent server provisioning without local boot devices. This integration supports both FC and FCoE vHBAs, with managed at the fabric level to secure LUN access across the UCS domain. The converged design eliminates the need for separate FC switches, as UCS Fabric Interconnects emulate FC ports on unified I/O modules, supporting up to 64 Gbps FC speeds on UCS 6600 Series models like the UCS 6664, and up to 32 Gbps on UCS 6400 Series. This consolidation simplifies topology, reduces operational costs, and enhances efficiency by sharing infrastructure for Ethernet and storage traffic while maintaining lossless delivery through Data Center Bridging (DCB) standards. In 2025 enhancements, UCS introduced support for NVMe/TCP as a transport option within NVMe-oF, enabling Ethernet-based NVMe access using standard TCP/IP without specialized RDMA hardware, as demonstrated in integrations like the UCS X215c M8 with Hitachi VSP One Block Storage over 100 Gbps NVMe/TCP networks. Additionally, HyperFlex, UCS's solution, incorporates distributed storage via its HX Platform, which aggregates node-local NVMe drives into a shared, policy-driven object store accessible over the UCS fabric, supporting enhanced scalability for all-NVMe clusters.

Virtualization and Workloads

Hypervisor Support

The Cisco Unified Computing System (UCS) supports a range of certified hypervisors to enable robust deployments, including ESXi, Microsoft , and KVM-based platforms such as KVM. These hypervisors are integrated with UCS hardware through Cisco's pre-qualified Hardware Compatibility List (HCL), which verifies compatibility for specific UCS servers, fabric interconnects, and virtual interface cards (VICs) to ensure optimal and stability in virtual environments. The HCL includes tested configurations for various versions, such as ESXi 8.0 and later, on , and KVM on RHEL 9.x, allowing administrators to deploy virtual machines without compatibility issues. A key feature enhancing integration in UCS is the Virtual Machine Fabric Extender (VM-FEX), which provides direct I/O passthrough by extending the UCS fabric to , thereby reducing overhead through hardware-based switching at the fabric interconnect level. This technology minimizes CPU utilization for network processing and enables consistent policy application across physical and virtual interfaces. However, since UCS Manager release 4.1(1), VM-FEX support has been limited to Linux KVM environments, with VMware ESXi and versions deprecated in favor of alternative virtualization optimizations. UCS further optimizes virtualized network performance via Single Root I/O Virtualization (SR-IOV) and (DPDK) acceleration, supported on VIC 1400 and 15000 series adapters. SR-IOV partitions the physical NIC into virtual functions assignable directly to VMs, bypassing hypervisor mediation to achieve near-native I/O throughput and lower latency for high-bandwidth applications. DPDK complements this by enabling user-space packet processing, which accelerates data flows in virtualized setups and supports up to 100 Gbps Ethernet speeds without kernel interruptions. As of 2025, UCS has introduced enhancements for container integration, including native support for Kubernetes through validated designs with Red Hat OpenShift Container Platform on UCS X-Series and C-Series servers, bridging traditional hypervisors with containerized workloads for hybrid deployments. Additionally, UCS incorporates confidential computing features, utilizing Intel Software Guard Extensions (SGX) on compatible processors like those in the UCS X210c M7 node, to isolate and encrypt sensitive data and code execution within virtual machines, safeguarding against unauthorized access even from privileged software.

Workload Orchestration

Service profile templates in the Unified Computing System (UCS) enable administrators to define reusable configurations that encapsulate server identity, settings, boot policies, and network/storage connectivity parameters, allowing for rapid provisioning of multiple servers with consistent setups. These templates support the creation of service profiles through instantiation, where a single template can generate numerous profiles for cloning virtualized environments, reducing deployment times from hours to minutes by automating hardware associations and policy inheritance. For instance, templates can specify vNIC and vHBA configurations, ensuring scalability in operations without manual reconfiguration for each server. UCS integrates with orchestration platforms like Cisco Application Centric Infrastructure (ACI) and UCS Director to facilitate end-to-end automation of workloads, combining compute provisioning with network and storage policies. Through ACI integration, UCS Manager (UCSM) allows the Application Policy Infrastructure Controller (APIC) to dynamically apply networking configurations to UCS domains, enabling automated endpoint group assignments and policy enforcement for virtualized workloads. UCS Director extends this by providing a multitenant orchestration layer that automates service profile deployment across UCS and third-party infrastructure, supporting workflows for resource pooling and lifecycle management in private cloud environments. A key feature of UCS workload orchestration is stateless mobility, achieved via service profiles that abstract server hardware details, permitting seamless migration of workloads between physical and virtual hosts without reconfiguration. This mobility relies on pools of MAC addresses, WWNs, and UUIDs, allowing a service profile to be disassociated from one server and reassociated with another, preserving network identity and boot configurations for uninterrupted operation. SAN booting enhances this capability by minimizing dependencies on local storage, enabling profile migrations across UCS domains in under a minute for non-disruptive maintenance or load balancing. As of 2025, UCS leverages AI-assisted through Cisco Intersight, incorporating for proactive workload scaling in hybrid setups. Intersight's AI-driven insights forecast potential failures and resource demands, automating self-healing actions and policy adjustments to optimize scaling across on-premises UCS and extensions. This includes predictive forecasting for AI workloads in hybrid environments, such as those using UCS X-Series with AI Pods, where models anticipate capacity needs and trigger workflows for efficient resource allocation without manual intervention.

Applications and Advancements

Data Center Use Cases

The Unified Computing System (UCS) facilitates consolidation in enterprise data centers by enabling the reduction of server sprawl through efficient resource pooling and stateless computing models. In virtual desktop infrastructure (VDI) deployments, UCS servers such as the B250 M2 demonstrated, as of 2010 tests, the ability to consolidate up to 29 legacy physical servers into a single , significantly decreasing the physical footprint and management overhead while supporting dense environments. For private cloud implementations, UCS's extended memory technology allows for higher density, optimizing compute utilization and addressing sprawl in dynamic, multi-tenant setups. In (HPC) and database workloads, UCS provides scalable configurations tailored for demanding and SQL Server environments. For instance, as demonstrated in 2011 tests, the UCS B250 supported consolidation of up to 11 virtual machines running 44 instances of 10 GB SQL Server databases under (OLTP) loads, achieving linear scaling in operations per minute (OPM) with 88% utilization and near-full CPU engagement. Similarly, the UCS B440 handled seven virtual machines with 100 GB databases, maintaining 86% utilization and 74% CPU while delivering consistent transaction performance per minute (TPM), making it suitable for large-scale and Microsoft SQL workloads in consolidated data centers. For applications, UCS E-Series servers offer compact, integrated solutions for branch offices, with current M6 models providing high-density blade servers with single-socket processors for distributed environments. Earlier integrations, such as with Cisco ISR G2 routers (end-of-support 2022), consolidated branch applications like , DHCP, and virtualized network services, reducing infrastructure costs and enabling local hosting of services. In banking scenarios, these provided for teller applications through isolated virtual partitions, minimizing downtime in remote locations. Case studies illustrate the economic impact of UCS in these use cases, with organizations reporting up to 50% reductions in (TCO) through lower power, cooling, and infrastructure expenses, as per a 2014 Forrester study. Deployment times have also accelerated dramatically; for example, provisioning shifted from 6-8 weeks to under one hour via UCS service profiles, representing over 3x faster implementation in virtualized environments. In Teradata's deployment, UCS achieved 50% reduced provisioning time alongside 35% overall savings, enhancing agility for data-intensive operations.

Recent Innovations

In recent years, Cisco has advanced the Unified Computing System (UCS) to support artificial intelligence and machine learning workloads through the introduction of the UCS X-Series modular system in 2022. This platform integrates NVIDIA GPUs, such as the A100 Tensor Core, L4, T4, A40, and A16 models, enabling high-density configurations with up to 24 GPUs per chassis for tasks like model training, inferencing, and generative AI applications. The design accommodates PCIe expansion nodes for GPU acceleration, supporting comprehensive NVIDIA AI Enterprise tools and frameworks to optimize performance across diverse AI pipelines. Additionally, the UCS X9508 chassis provides enhanced airflow and space for advanced thermal solutions, including liquid cooling, to handle high-power components in dense AI clusters capable of exceeding 1 petaFLOPS of compute performance when scaled. UCS has evolved to facilitate hybrid cloud environments by leveraging the Intersight cloud-based management platform for seamless integration with services like and AWS Outposts. Through Intersight, UCS deployments enable dynamic , such as turning processor cores on or off to align with demands, while providing fault-tolerant reassignment across clusters of 1 to 16 servers equipped with NVMe storage. This integration extends enterprise-grade networking policies to AWS Outposts via Nexus fabrics, supporting consistent segmentation, low-latency I/O, and data locality for multicloud orchestration without on-premises management overhead. Intersight's analytics further simplify hybrid operations, allowing UCS to bridge on-premises infrastructure with cloud resources for scalable, policy-driven deployments. Sustainability enhancements in UCS focus on energy-efficient designs, particularly with the UCS X-Series, which achieves 54% greater energy efficiency compared to prior generations through optimized packaging and support for advanced silicon like 5th-generation processors. These improvements reduce power consumption in operations, aligning with Cisco's broader goals of implementing energy efficiency projects that avoided 1.3 gigawatt-hours in fiscal 2024. The modular architecture minimizes waste by enabling scalable upgrades, while features like liquid cooling for high-density AI workloads further lower thermal overhead and overall power draw. Security innovations in UCS incorporate zero-trust principles through integration with XDR and the Security Cloud, which provide orchestration for threat detection and response across the infrastructure. UCS Manager provides built-in AES encryption (128- or 256-bit) for management sessions and sensitive data, including Type-6 keys for MACsec, ensuring without exporting master keys in backups. This encrypted management extends to Intersight for cloud-delivered zero-trust access, combining identity verification, device posture checks, and adaptive policy enforcement to protect UCS environments in hybrid setups. In 2025, introduced UCS M8 solutions featuring processors and the X215c compute node for high-performance applications, enhancing scalability and efficiency. UCS now supports AI PODs built on GPUs integrated with UCS infrastructure, as announced at Cisco Live 2025. Additionally, partnerships such as with G42 enable secure end-to-end AI infrastructure using MI350X GPUs, powering large-scale AI clusters.

References

  1. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/solution-overview-c22-744677.html
  2. https://www.cisco.com/c/m/en_us/products/servers-unified-computing/ucs-timeline.html
  3. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-6600-series-fabric-interconnect-ds.html
  4. https://www.cisco.com/c/en/us/products/collateral/interfaces-modules/unified-computing-system-adapters/ucs-vic-15000-series-ds.html
  5. https://www.cisco.com/site/us/en/products/computing/servers-unified-computing-systems/index.html
  6. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/hw/site-prep-guide/ucs_site_prep/ucs_siteprep_chapter_01.pdf
  7. https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2009/m03/cisco-unleashes-the-power-of-virtualization-with-industry-s-first-unified-computing-system.html
  8. https://blog.scottlowe.org/2009/03/16/cisco-ucs-announcements/
  9. https://newswire.telecomramblings.com/2012/01/cisco-tops-10000-unified-computing-system-customers-captures-53-industry-benchmark-world-records/
  10. https://www.cisco.com/c/dam/en/us/products/collateral/switches/nexus-7000-series-switches/vmm_infrastructure.pdf
  11. https://www.cisco.com/c/dam/en/us/products/collateral/servers-unified-computing/le_32801_pb_ucs_worldrecords.pdf
  12. https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2017/m07/cisco-delivers-powerful-5th-generation-servers-for-its-unified-computing-system.html
  13. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/x/release/notes/6600_series_fi_ordering-guide.html
  14. https://www.ciscolive.com/c/dam/r/ciscolive/global-event/docs/2025/pdf/BRKCOM-2658.pdf
  15. https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2013/m12/cisco-completes-acquisition-of-insieme-networks.html
  16. https://www.cisco.com/c/en/us/td/docs/switches/datacenter/aci/apic/sw/4-x/virtualization/Cisco-ACI-Virtualization-Guide-42x/aci-with-ucsm-integration-42x.pdf
  17. https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2017/m03/cisco-completes-acquisition-of-appdynamics.html
  18. https://appdynamics.github.io/ucs-monitoring-extension-ps1/ucsdashboard.html
  19. https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2021/m06/cisco-flexes-its-muscle-in-cloud-helping-customers-be-cloud-smart-to-deliver-exceptional-digital-experiences.html
  20. https://www.cisco.com/site/us/en/products/computing/servers-unified-computing-systems/ucs-manager/index.html
  21. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/release/notes/b_release-notes-ucsm-6_0.html
  22. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs6536-fabric-interconnect-ds.html
  23. https://www.cisco.com/c/en/us/products/collateral/cloud-systems-management/intersight/stateless-computing-ucs-intersight.html
  24. https://community.cisco.com/t5/data-center-and-cloud-knowledge-base/what-is-stateless-computing-in-ucs/ta-p/3116244
  25. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-x-series-modular-system/ucs-x210c-m8-compute-node-ds.html
  26. https://www.cisco.com/c/dam/en/us/products/collateral/servers-unified-computing/ucs-x-series-modular-system/ucs-x215c-m8-compute-node.pdf
  27. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/intel-xeon-6-cpu-faq.html
  28. https://www.cisco.com/c/dam/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/ucs-intel-m8-memory-guide.pdf
  29. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/ucs-c845a-m8-rack-server-ds.html
  30. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-x-series-modular-system/ucs-x-series-intel-flex-dc-gpu-vdi-wp.html
  31. https://www.cisco.com/c/dam/en/us/solutions/collateral/borderless-networks/advanced-services/ucs_ibm.pdf
  32. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/UCS_CVDs/ucs_s3260_cohesity_smartfiles.html
  33. https://www.cisco.com/c/dam/global/sv_se/assets/pdfs/Cisco_UCS_Manager_overview.pdf
  34. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Server-Mgmt/4-3/b_cisco_ucs_manager_server_mgmt_guide_4_3/m_power_capping_in_cisco_ucs.html
  35. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/c/sw/cli/config/guide/4_2/b_cisco_ucs_c-series_cli_configuration_guide_for_s3260_servers_42/b_Cisco_UCS_C-Series_CLI_Configuration_Guide_for_S3260_Servers_41_chapter_0100.html
  36. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Server-Mgmt/4-3/b_cisco_ucs_manager_server_mgmt_guide_4_3.pdf
  37. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Admin-Management/4-3/b_cisco_ucs_admin_mgmt_guide_4-3/m_organizations_in_ucs_manager.html
  38. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Server-Mgmt/6-0/b_cisco_ucs_manager_server_mgmt_guide_6_0/m-service-profiles.html
  39. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/System-Monitoring/4-3/b-ucsm-gui-system-monitoring-guide-4-3/m_system_monitoring_overview.html
  40. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/sw/api/b_ucs_api_book/b_ucs_api_book_chapter_00.html
  41. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/hw/chassis-install-guide/ucs5108_install/ucs5108_install_chapter_01.html
  42. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-b-series-blade-servers/data_sheet_c78-520522.html
  43. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-b-series-blade-servers/datasheet-c78-2368888.html
  44. https://www.cisco.com/c/en/us/solutions/data-center/data-center-computing/model-comparison.html
  45. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/ucs-c220-m8-rack-server-ds.html
  46. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/ucs-c225-m8-rack-server-ds.html
  47. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/ucs-c240-m8-rack-server-ds.html
  48. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/ucs-c245-m8-rack-server-ds.html
  49. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/ucs-c885a-m8-ds.html
  50. https://www.cisco.com/c/en/us/products/collateral/interfaces-modules/unified-computing-system-adapters/datasheet-c78-732515.html
  51. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/ucs-amd-epyc-4-gen-20-world-rec-so.html
  52. https://www.ciscolive.com/c/dam/r/ciscolive/global-event/docs/2025/pdf/BRKCOM-1047.pdf
  53. https://www.wjbf.com/business/press-releases/ein-presswire/787712695/deepcoolai-and-cisco-partner-to-engineer-the-future-of-ai-data-centers
  54. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-6200-series-fabric-interconnects/data_sheet_c78-675245.html
  55. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-6300-series-fabric-interconnects/datasheet-c78-736682.html
  56. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/datasheet-c78-741116.html
  57. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Fabric-Interconnect-Migration/6-0/b_migration_guide_for_cisco_ucs_fabric_interconnects_6_0/m-overview-of-cisco-ucs-fabric-interconnects.html
  58. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Infrastructure-Mgmt/4-1/b_UCSM_GUI_Infrastructure_Management_Guide_4_1/b_UCSM_GUI_Infrastructure_Management_Guide_4_1_chapter_01011.html
  59. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/sw/gui/config/guide/2-0/b_UCSM_GUI_Configuration_Guide_2_0/b_UCSM_GUI_Configuration_Guide_2_0_chapter_010.pdf
  60. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Getting-Started/4-3/b_ucsm_getting_started_guide_4_3/m_overview.html
  61. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/sw/api/UCSM_API_Guide/b_UCSM_API_Guide_3_1_chapter_01.html
  62. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/sw/gui/config/guide/1-2-1/b_UCSM_GUI_Configuration_Guide_1_2_1/UCSM_GUI_Configuration_Guide_1_2_1_chapter2.html
  63. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Server-Mgmt/4-3/b_cisco_ucs_manager_server_mgmt_guide_4_3/m_server_management_overview-4_3.html
  64. https://www.oit.va.gov/services/trm/ToolPage.aspx?tid=8828
  65. https://www.cisco.com/c/en/us/support/servers-unified-computing/ucs-manager/products-release-notes-list.html
  66. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Admin-Management/4-3/b_cisco_ucs_admin_mgmt_guide_4-3/m_backup_and_restore_configuration.html
  67. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/CLI-User-Guides/Admin-Management/6-0/b_cisco_ucs_manager_cli_administration_mgmt_guide_6-0/m_cli_backing_up_and_restoring_the_configuration.html
  68. https://www.cisco.com/c/en/us/support/docs/ip/lightweight-directory-access-protocol-ldap/218189-configure-ldap-in-ucs-manager.html
  69. https://www.cisco.com/c/en/us/products/collateral/cloud-systems-management/intersight/intersight-ds.html
  70. https://www.cisco.com/c/en/us/products/collateral/cloud-systems-management/intersight/at-a-glance-c45-739419.html
  71. https://www.cisco.com/site/us/en/products/computing/hybrid-cloud-operations/intersight-platform/index.html
  72. https://video.cisco.com/detail/video/6384262008112
  73. https://www.cisco.com/site/us/en/products/computing/hybrid-cloud-operations/intersight-workload-optimizer/index.html
  74. https://www.intersight.com/help/whats_new/
  75. https://intersight.com/help/saas/resources/plugin_vsphere
  76. https://www.cisco.com/c/en/us/products/collateral/cloud-systems-management/intersight/compute-intersight-hardening-guide-wp.html
  77. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Firmware-Mgmt/4-3/b_UCSM_GUI_Firmware_Management_Guide_4-3/b_UCSM_GUI_Firmware_Management_Guide_chapter_01.html
  78. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Firmware-Mgmt/6-0/b_UCSM_GUI_Firmware_Management_Guide_6_0/m-guidelines-and-prerequisites.html
  79. https://www.cisco.com/c/en/us/support/docs/servers-unified-computing/unified-computing-system/217433-upgrade-infrastructure-and-server-firmwa.html
  80. https://www.cisco.com/c/dam/en/us/td/docs/unified_computing/ucs/ucs-manager/cross_version_firmware_matrix_6_0_onwards/index.html
  81. https://www.cisco.com/c/en/us/solutions/collateral/data-center-virtualization/microsoft-applications-on-cisco-ucs/secured-core-servers-enabling-guide.html
  82. https://intersight.com/help/saas/resources/Upgrading_Server_Firmware
  83. https://www.cisco.com/c/en/us/support/docs/servers-unified-computing/ucs-infrastructure-ucs-manager-software/217601-ucsm-health-and-pre-upgrade-check-tool.html
  84. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-5th-gen-unified-fabric-aag.html
  85. https://www.ciscolive.com/c/dam/r/ciscolive/global-event/docs/2025/pdf/BRKCOM-2657.pdf
  86. https://www.cisco.com/c/en/us/support/docs/servers-unified-computing/ucs-manager/116188-configure-fcoe-00.html
  87. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-c-series-rack-servers/solution-overview-c22-740196.html
  88. https://www.cisco.com/c/en/us/support/docs/servers-unified-computing/unified-computing-system/222553-configure-boot-from-san-in-ucs-manager.html
  89. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Server-Mgmt/4-3/b_cisco_ucs_manager_server_mgmt_guide_4_3/m_server_boot_configuration.html
  90. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/hw/6454-install-guide/6454.pdf
  91. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/UCS_CVDs/ucs_hitachi_adaptive_solutions_nvme_tcp_design.html
  92. https://www.cisco.com/c/en/us/products/collateral/hyperconverged-infrastructure/hyperflex-hx-series/solution-overview-c22-744674.html
  93. https://ucshcltool.cloudapps.cisco.com/public/
  94. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-central/GUI-User-Guides/Server-Mgmt/1-5/b_CiscoUCSCentral_Server_Management_Guide_1-5/b_CiscoUCSCentral_Server_Management_Guide_1-5_chapter_0100.pdf
  95. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-director/datasheet-c78-730830.html
  96. https://www.cisco.com/c/dam/en/us/td/docs/unified_computing/ucs/reference/command_reference.pdf
  97. https://www.ciscolive.com/c/dam/r/ciscolive/global-event/docs/2025/pdf/PSODCN-1012.pdf
  98. https://www.cisco.com/c/dam/en/us/solutions/collateral/data-center-virtualization/unified-computing/server_consolidation_white_paper.pdf
  99. https://www.cisco.com/c/dam/en/us/solutions/collateral/borderless-networks/advanced-services/cisco_ucs_sql_consolidation.pdf
  100. https://www.cisco.com/c/dam/en/us/products/collateral/servers-unified-computing/ucs-e-series-servers/extending_the_data_center.pdf
  101. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-e-series-servers/ucs-e-series-m6-server-ds.html
  102. https://bsgl.co.uk/wp-content/uploads/2018/03/The-Total-Economic-Impact-of-Cisco-UCS-Blue-Sky.pdf
  103. https://www.cisco.com/c/dam/en_us/about/ciscoitatwork/data_center/docs/Case_Study_How_Cisco_is_Progressing_Toward_IT_as_a_Service.pdf
  104. https://www.cisco.com/site/us/en/about/case-studies-customer-stories/teradata.html
  105. https://www.cisco.com/c/en/us/products/servers-unified-computing/ucs-x-series-modular-system/index.html
  106. https://www.ciscolive.com/c/dam/r/ciscolive/apjc/docs/2024/pdf/BRKCOM-3618.pdf
  107. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/UCS_CVDs/vsan_ucsx_vmware_vsphere.html
  108. https://www.cisco.com/c/en/us/products/collateral/servers-unified-computing/ucs-x-series-modular-system/ai-ucs-m7-blades-5th-gen-xeon-cpu-wp.html
  109. https://www.cisco.com/c/en/us/solutions/data-center-virtualization/microsoft-applications-on-cisco-ucs/microsoft-azure-stack-hci.html
  110. https://www.cisco.com/c/en/us/solutions/collateral/data-center-virtualization/application-centric-infrastructure/solution-overview-c22-743174.html
  111. https://www.cisco.com/c/dam/en/us/products/collateral/servers-unified-computing/ucs-x-series-modular-system/idc-bets-big-hybrid-cloud-ucs-intersight.pdf
  112. https://www.cxoinsightme.com/news/cisco-unveils-five-ways-it-can-support-in-driving-sustainability-for-businesses/
  113. https://www.cisco.com/c/dam/m/en_us/about/csr/esg-hub/_pdf/purpose-report-2024.pdf
  114. https://blogs.cisco.com/datacenter/improving-data-center-energy-efficiency-to-power-the-ai-revolution
  115. https://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/ucs-manager/GUI-User-Guides/Admin-Management/6-0/b_cisco_ucs_admin_mgmt_guide_6_0/m-security-management.html
  116. https://www.cisco.com/c/en/us/solutions/collateral/enterprise/design-zone-security/zt-ag.html
  117. https://www.cisco.com/c/en/us/products/security/xdr/index.html
  118. https://www.wwt.com/video/cisco-ucs-m8-solutions-with-amd-epyc-redefining-data-center-performance
  119. https://www.forrester.com/blogs/key-takeaways-from-cisco-live-2025-ciscos-big-bets-for-unified-security-and-ai/
  120. https://investor.cisco.com/news/news-details/2025/Cisco-and-G42-Deepen-US-UAE-Technology-Partnership-to-Build-Secure-End-to-End-AI-Infrastructure-in-the-UAE/default.aspx
Add your contribution
Related Hubs
User Avatar
No comments yet.