Cisco Blogs


Cisco Blog > Data Center and Cloud

ACI for Splunk Enterprise: Enabling comprehensive application health

On January 13th, 2015, Cisco will celebrate the 1-year anniversary of its launch of Application Centric Infrastructure (ACI), a ground breaking SDN architecture.  It will include a public webcast with the participation of early ACI adopters and our ecosystems partners.  One of these inaugural partners was Splunk,  the Operational Intelligence company for all types of IT organizations.  At the webcast, Splunk and other partners will describe  a range of new solutions with ACI, that dramatically simplify Data Center operations. Here is a preview of Splunk’s solution.

A large portion of the data center operational effort is consumed in managing application health.  This includes:

  • Ensuring the end-user experience for distributed users with different types of performance needs
  • Discovering the physical and virtual resources associated with applications and the user experience
  • Detailed monitoring of resources and events in the infrastructure that affect application performance

These activities have become more complex as applications have become distributed, interconnected or cloud based because they cause applications to move, scale and evolve rapidly.

Splunk Enterprise can monitor and analyze millions of infrastructure events through logs and agents, in real-time. This can provide rapid visibility and isolation of infrastructure that affect application performance. Cisco has been collaborating with Splunk to combine the application visibility of Cisco Application Centric Infrastructure with operational analytics of Splunk Enterprise. The result is “Cisco ACI for Splunk Enterprise” a highly scalable application that is orderable immediately at Splunk.com.

ACI and Splunk have enabled a comprehensive view of application health with the ability to monitor the entire end-to-end environment in real time and proactively prevent issues from impacting end users.

ACI provides visibility to application health from the network perspective by tracking all network dependencies and events that impact application performance and security. Splunk complements Cisco ACI by bringing actionable intelligence across the entire data center infrastructure including storage, compute, virtualization endpoints, as well as application tiers and components provided by ACI. Splunk’s analytical and visualization tools provide real-time insights to data center teams to optimize performance and ensure security policies in a highly dynamic environment.

How does it work

Cisco ACI exposes a wealth of networking data previously inaccessible to Splunk. The Cisco ACI app for Splunk Enterprise gathers data from APIC (Application Policy Infrastructure Controller) including APIC network events, health scores and inventory of logical constructs (e.g. tenants, application profiles, end point groups) and physical constructs ( e.g spines, leafs, VMs).

SplunkACI_CentralizedApplicationHeath

This data is used to:

  1. Reduce resolution time with accelerated root-cause analysis
  • Splunk enables users to reduce the mean time to investigate/resolve problems up to 70%
  • Centralized management of operational health of ACI environment & underlying entities in real-time
  • Detect issues or anomalies in performance or response times and proactively resolve
  • For multiple tenants, quickly navigate to the source of problems using flexible per-role views, including 1) Help Desk view, 2) Tenant View and 3) Fabric view
  1. Provide Central Proactive Monitoring of Cisco ACI
  • Get real-time proactive notification of network traffic and device faults with location, affected objects.
  • Track trends and anticipate application impact
  1. Operational Analytics across the entire virtual and physical infrastructure
  • Optimize network capacity and prevent service deterioration with detailed visibility into fabric path degradation.
  • Meet compliance/security with user analytics, including authentication tracking reports.
  • Correlate data from Cisco ACI with data from storage resources, operating systems, applications, security devices, endpoint and more for enterprise-wide visibility.
  • Trace and monitor transactions through all tiers of a distributed application architecture
  • Gives application managers a perspective on the underlying Cisco ACI infrastructure’s effect on applications without being directly involved in ACI Ops.
  • Monitor key operational metrics such as end-to-end response times to ensure SLAs met.

As an example, a Fortune 100 company is using Splunk with ACI:

  • for operational visibility for their ACI cluster with ability to quickly identify faults and troublesome tenants and determine corrective action.
  • to provide centralized visibility as ACI expands across multiple data centers and for proactive monitoring to establish baselines and triggered alerts when key thresholds exceeded.

This approach to Application Health is part of the broader discipline of Application Performance Management (APM). According to Gartner, “By 2018, 60% of APM deployments will use and integrate data extracted directly from log  files alongside wire data and agent-derived data as a foundation for reporting, prediction, and analysis, up from less than 5% today.”  With our collaboration, ACI for Splunk Enterprise provides important new capabilities for  Application Performance Management.

Learn more about Cisco ACI for Splunk Enterprise here.  And register for Cisco’s webcast on January 13th.

 

 

Tags: , , , , , , , ,

Five Cool Router Tricks with onePK

Network Management is dull.  No excuses.  Monitoring and interacting with the devices that move data from one location to another is a thankless undertaking that most of us building networks leave to an afterthought.  Part of that is the complexity associated with managing networks.  There are at least a dozen common methods for interacting with devices in the network including SNMP, CLI, AAA, Syslog, Netflow, and fancy XML/HTTP interfaces.  So much variety breeds complexity so we tend to set our goals pretty low for interactivity with the network.

What if we had one common mechanism for interacting with the network?  Different devices running different software would all speak a common language to the applications managing and monitoring them.  Now what if that language was something the programmers writing those applications understood implicitly like an API library they could compile directly into their program?  That would make interacting with the network as simple as making a procedure call within the application.  That’s exactly what onePK – or the “one Platform Kit” – accomplishes.

Read More »

Tags: , , , , , , , , , , , , , , , , ,