How Splunk and Cisco Bring Intelligence to Infrastructure

A New Age of Intelligent Infrastructure

AI-driven workloads have fundamentally changed how organizations view infrastructure performance. Modern compute platforms such as Cisco UCS X-Series, C-Series, and the Cisco AI Pod now support GPU-dense environments that power generative AI, visualization, inference, and large language models. These workloads operate under tight performance, cost, and latency constraints, where even small inefficiencies can quickly translate into lost productivity or escalating expenses.

To meet these demands, infrastructure must provide continuous, real-time visibility across every layer of the stack, from hardware and hypervisors to applications and AI frameworks. Whether managed through Cisco Intersight or deployed as independent compute nodes, organizations need a consistent way to collect, analyze, and act on performance data as it happens.

Splunk provides that capability through its complementary platforms:

• Splunk Cloud Platform integrates directly with Cisco Intersight via the Cisco Intersight Add-on for Splunk to ingest logs, events, and API-driven telemetry for lifecycle and health monitoring.
• Splunk Observability Cloud (O11y) uses OpenTelemetry (OTel) to collect metrics, traces, and logs.
• Splunk IT Service Intelligence (ITSI) applies machine learning and predictive analytics to anticipate issues before they occur.

Together, these platforms go beyond monitoring and deliver a complete observability story. Intersight sends operational context and system state to Splunk Cloud. While host-level and AI workload telemetry flow into Splunk Observability Cloud, giving teams a unified view of both infrastructure and application performance for end-to-end visibility and troubleshooting.

Operations teams gain proactive insight into capacity trends, performance bottlenecks, and workload behavior, enabling smarter scaling and faster recovery. Combined with Cisco's powerful compute ecosystem, this creates a new operational model where infrastructure is not only managed but intelligently observed, optimized, and automated.
 
 

From UCS Manager to Intersight Intelligence

Cisco's evolution from UCS Manager to Cisco Intersight represents a major advancement in how compute infrastructure is managed and understood. What began as localized, chassis-level control has matured into a global, cloud-connected platform that delivers API-driven lifecycle orchestration.

Cisco Intersight provides:

• Centralized visibility across UCS and X-Series systems
• Automated firmware and configuration management
• Unified telemetry for health, power, and performance metrics
• Policy-based governance that simplifies lifecycle operations

As data centers expand to support AI and high-performance computing, visibility into hardware alone is no longer sufficient. The latest Cisco AI platforms, to include the Cisco UCS C845A M8, the C880A M8, and the C885A M8, introduce new levels of complexity with GPU acceleration, high-bandwidth fabrics, and advanced software stacks that span from bare metal to inference engines. To maintain peak performance, organizations must connect infrastructure metrics directly to workload behavior, latency, and user experience.

Splunk Cloud brings that visibility to life. By collecting important sources within the Cisco Intersight like Alarms, Audit Logs, Inventory, and Metrics, it turns telemetry and data into actionable intelligence. The Cisco Intersight add-on for Splunk also provides easy-to-use, pre-built dashboards that give you a high-level view of your infrastructure, speeding up analysis and improving operational monitoring. Providing users the ability to correlate hardware state with operational impact, identify configuration drift, and visualize trends that inform long-term capacity planning.

Together, Cisco Intersight and Splunk Cloud enable data-driven operations that combine deep infrastructure awareness with predictive insight.

Unifying Data with OpenTelemetry

Modern hybrid environments require a consistent way to monitor systems and workloads across diverse platforms. (OTel) provides that standard. This vendor-neutral open source observability framework for instrumenting, generating, collecting, and exporting telemetry data such as traces, metrics, and logs. OpenTelemetry is designed to make instrumentation simple and universal—allowing you to easily monitor applications and systems across any programming language, infrastructure, or runtime environment.

At the center of this integration is the Splunk OpenTelemetry Collector, a lightweight, extensible service that gathers telemetry from sources and forwards it to the Splunk Observability Cloud. This enables consistent instrumentation across UCS, X-Series, AI Pod environments and beyond, giving organizations continuous insight into system and application behavior.

Cisco and Splunk have embraced OTel to close visibility gaps across complex, distributed environments spanning on premises infrastructure, multiple public clouds and various cloud-native platforms. OTel allows Cisco infrastructure and AI workloads to communicate within a shared observability framework. Whether a system is managed through Intersight or operates independently, OTel ensures performance data is captured accurately and streamed in real-time. It turns fragmented metrics into correlated insights, connecting CPU temperature, GPU utilization, token processing, and response times into a single performance story. The result is a seamless observability fabric that scales easily with the growing complexity of today's intelligent data centers.

Cisco Intersight and Splunk: From Insight to Action

Cisco Intersight delivers a deep and extensible data model that exposes telemetry across compute, fabric, and workload domains. Through secure APIs, Intersight provides access to granular metrics such as device health scores, thermal and power readings, inventory states, and policy compliance data. When streamed into Splunk Observability Cloud, this data becomes part of a broader analytics framework that correlates infrastructure telemetry with workload performance and business outcomes.

Splunk (ITSI) extends that context by defining service-level indicators (SLIs) and key performance indicators (KPIs) that map to critical workloads. This allows teams to see beyond individual alerts and understand the health of complete services. Within these service models, metrics from Intersight such as chassis temperature, fan speed, or power consumption can be combined with performance data from Splunk Observability Cloud to provide a complete picture of how infrastructure supports applications.

This hybrid approach enables teams to:

• Correlate events from Cisco Intersight with live telemetry from OTel sources
• Detect anomalies in context using ITSI service models
• Visualize dependencies between hardware, software, and workloads in a single view

Together, Cisco Intersight, Splunk Cloud, and Splunk Observability Cloud create a unified foundation for actionable insight that drives faster troubleshooting and more intelligent automation.

Detecting and Automating with ITSI and Intersight Cloud Orchestrator

Once infrastructure and telemetry data are consolidated, automation becomes the next step. Splunk (ITSI) and Cisco Intersight Cloud Orchestrator (ICO) work together to close the loop between detection and action.

ITSI uses adaptive baselines to identify deviations in performance. When it detects anomalies such as thermal spikes, fan failures, or performance drops, it can forward alerts or events directly to ICO, which executes pre-defined workflows.

For example:

• ITSI detects a UCS node with rising power draw or memory errors.
• It triggers an ICO workflow to rebalance workloads or apply a configuration policy.
• The action and result are reported back to Splunk for verification and trend analysis.

This combination provides an operational feedback loop where Intersight manages infrastructure actions and Splunk provides intelligence and visibility. The outcome is proactive, event-driven automation that improves uptime, responsiveness, and efficiency.

Beyond Intersight: Direct Host-Level Observability

Whether a system is managed through Cisco Intersight or operates independently, both Splunk Cloud Platform and Splunk Observability Cloud (O11y) deliver comprehensive visibility across compute, network, and application layers. The integration between Intersight and Splunk Cloud enhances this experience by adding lifecycle context, health telemetry, and configuration data, while O11y Cloud extends observability through OpenTelemetry Collectors that can be deployed directly on hosts or within Kubernetes clusters.

This unified approach ensures consistent insight across all environments. Metrics such as CPU and GPU utilization, memory consumption, power, and latency can be correlated with traces and application data, providing a holistic view of performance and reliability. Whether workloads are running under Intersight management, in edge deployments, or across containerized clusters, Splunk transforms telemetry into actionable intelligence that supports real-time decision making and automation.

When the Splunk OTel Collector runs on modular UCS or C-Series servers, it continuously gathers and streams data into Splunk Observability Cloud for analysis and visualization. 

It captures:

• System performance metrics such as CPU utilization, GPU activity, memory consumption, and network throughput
• Application telemetry from containers, microservices, and AI workloads
• Traces and logs that connect infrastructure performance with application behavior

This host-level integration provides a transparent view of performance and reliability. In Cisco and Splunk's Keeping LLM Performance in Check demonstration, metrics such as ollama_requests_total, ollama_request_duration_seconds, and system_cpu_usage_percent illustrated how model behavior and system efficiency can be monitored simultaneously.

Whether deployed in a data center, an edge site, or an AI development environment, Splunk converts isolated telemetry into actionable intelligence. Even without Intersight, organizations gain the same depth of analytics, real-time visibility, and predictive insight, making Splunk a consistent observability layer for every Cisco compute environment.

Observability for the Cisco AI Pod

The Cisco AI Pod represents the next stage in high-performance computing. It is a modular, GPU-optimized architecture designed for scalable AI training, inference, and data-intensive workloads. Built on Cisco UCS X-Series and C-Series servers and interconnected through Cisco Fabric Interconnects, the AI Pod delivers exceptional compute density and performance with unified management through Cisco Intersight.
 

As AI workloads become more complex, monitoring infrastructure health alone is not enough. Performance issues can emerge from multiple layers of the stack, including resource contention, network latency, storage throughput, or inefficiencies within the workload itself. Splunk Observability Cloud addresses these challenges by providing a unified, multi-layered view that connects system metrics, application traces, and model-level telemetry in one platform.

With OTel integrated throughout the AI Pod, Splunk continuously collects and analyzes data from every layer of the environment, including:

• GPU and interconnect utilization for hardware efficiency
• Container and application metrics for workload visibility
• Inference latency and response times for model and service performance

When combined with Splunk ITSI and Cisco Intersight Cloud Orchestrator, the AI Pod can respond intelligently to performance changes. ITSI detects the anomaly, while ICO executes remediation workflows that rebalance workloads or apply configuration policies automatically.

Together, Cisco and Splunk enable organizations to operate AI workloads with greater predictability and confidence. The result is a smarter, data-driven infrastructure that performs consistently whether deployed in a data center, at the edge, or within the WWT AI Proving Ground (AIPG).

Business Value and Continuous Improvement

Integrating Cisco Intersight, Splunk Cloud, and Splunk Observability Cloud delivers a foundation for intelligent, data-driven operations. By unifying telemetry from compute, GPU, and application layers, organizations gain complete visibility into how infrastructure performance affects user experience and business outcomes.

With the addition of Splunk ITSI providing predictive analytics and Intersight Cloud Orchestrator enabling automated action, infrastructure becomes self-aware, capable of detecting anomalies, understanding their cause, and initiating corrective measures. This proactive approach ensures systems remain stable, efficient, and responsive to evolving demands.

Organizations realize measurable value through:

• Reduced downtime and faster incident response
• Optimized resource utilization and performance efficiency
• Early detection of drift and bottlenecks
• Automated workflows that improve consistency and speed
• Predictive insights that guide capacity and investment planning

Beyond these results, enterprises gain confidence that their infrastructure is not only monitored but fully understood. Observability becomes a strategic advantage that drives reliability, innovation, and operational excellence.

Bringing It to Life in the ATC and AI Proving Ground

At World Wide Technology (WWT), this integration is already in action. Inside the Advanced Technology Center (ATC) and AI Proving Ground (AIPG), customers can explore the full ecosystem of Cisco Intersight, Splunk Cloud Platform, Splunk Observability Cloud, ITSI, and Intersight Cloud Orchestrator working together in live environments.

Within the ATC and AIPG, customers can:

• Simulate production-scale workloads across Cisco UCS X-Series, C-Series, and AI Pod platforms
• Test OpenTelemetry pipelines that stream real-time metrics into Splunk Observability Cloud
• Visualize GPU utilization, inference latency, and health status through integrated dashboards
• Validate automation workflows that execute through Intersight Cloud Orchestrator

These labs provide a hands-on environment to test, benchmark, and refine observability architectures before deployment.

The partnership between Cisco, Splunk, and WWT demonstrates what becomes possible when operational data and intelligent automation come together. Together, we are helping organizations modernize infrastructure, accelerate AI adoption, and build systems that think, adapt, and perform with confidence.

Try the Interactive Demo

Experience how Cisco and Splunk technologies bring intelligence to infrastructure.
This 10- to 15-minute browser-based demo walks you through real-world monitoring, analytics, and automation scenarios that showcase how observability drives smarter operations.