Splunk for Virtualization

The old way: Virtualization adds new complexity and uncertainty.

Many applications behave unpredictably after a move to a virtual environment. Virtual applications fight for resources on the same physical host. And the data needed to diagnose these problems is scattered and often lost as VMs are redeployed. And the move to virtualization adds far more management complexity. Existing tools built for relatively stable dependencies can't keep up. New tools to monitor the hypervisor and virtual machines cannot see the resident guest operating systems or applications. This makes it impossible to correlate activity and performance at an application level with resource utilization and performance down to the bare metal.

The new way: IT Search puts you back in control.

Only Splunk easily crosses tiers throughout the entire virtual stack - both inside and outside of the VM - to give you a complete picture of availability.

Splunk indexes all IT data across every tier - the physical servers, hypervisor, VMs, and deployed applications, capturing and persisting 100% of your data in real-time. Powerful search and navigation lets you trace performance problems and errors across components. Visibility across VMs highlights resource competition issues. Flexible alerting and reporting give you continuous visibility and monitoring of changing virtual environments. Whether you're testing a new virtualization rollout or managing an existing infrastructure, Splunk puts you back in control.

Why Splunk for Server Virtualization Management?

  • Get a comprehensive view of 100% of your IT data - server, hypervisor, VM, guest OS, applications, network, and more
  • Increase visibility even with the greater complexity of virtualized environments
  • Optimize your initial planning and ongoing use of virtualization
  • Retain data from transient guests for root cause analysis and compliance

Use Splunk for:

Virtualization Planning
Splunk helps with planning even before the first hypervisor is loaded. Use Splunk reports for historical app and OS activity and utilization to identify good targets for virtualization. Splunk shows you a historical view of the apps with spiky workloads and the hosts with underutilized available resources., helping you marry the tasks and resources faster, instead of guessing through your first deployments.

Workload Optimization
Once your virtualization environment is running, Splunk helps you optimize it. Use reports in Splunk across all tiers to track actual utilization and application performance to identify both application contention issues and lingering resource inefficiencies.

Performance Monitoring
Splunk acts as a great monitoring tool since it indexes 100% of your IT data - inside and outside of your virtualization environment. You can schedule searches and alerts in Splunk to generate alarms on performance thresholds based on data gathered from the VMWare CIM and Citrix Xen Management API about the VM, guests, physical hosts, and virtual and physical network interfaces. Splunk also includes pre-built searches and reports that monitor key virtualization metrics.

Splunk can alert you when your VMs or guest OSs are short on free memory for too long. You can extend monitoring based on the outcome of root cause analysis: schedule alerts via email, warnings via RSS, or send events to consoles and ticketing systems.

Root Cause Analysis
Splunk is the answer when IT staff asks "where did that instance go?" Use Splunk to index IT data historically from all tiers as instances come and go to do root cause analysis. Then tie real application errors and perf problems to information about the state of the underlying VM and other guests. Even if the environment changed between the problem occurring and the investigation beginning, Splunk still indexed it, and can help you solve it.

Take a common scenario: users complain about intermittent CRM app performance issues. Splunk can pinpoint the exact times and application server instances where performance fell below a threshold, then correlate it with configuration history captured from the virtualization platform APIs. Now you know which other guests shared the same physical hosts, can identify the I/O utilization hog, and even trend the I/O utilization of all of the guests over time.

Log Management
Splunk closes the gap in meeting log management requirements in virtualized environments. Unlike traditional log management solutions, Splunk securely and remotely captures all IT data in real time, even application logfiles, so you can meet log centralization and monitoring requirements even for applications deployed on transient virtual hosts.

Talk to an Operations Expert

Eric Garner Expertise: J2EE monitoring and troubleshooting and messaging infrastructures

Dan Goldburt Expertise: Monitoring and service level management for mission critical applications and infrastructure

Michael Wilde Expertise: Monitoring and troubleshooting Windows applications and infrastructures

Vi Ly Expertise: Monitoring and troubleshooting large-scale LAMP and J2EE environments on Linux and Windows

Robert Ide Expertise: Managing large-scale virtualization and grid computing environments

Mark Bagley Expertise: ITIL, IT Governance, SOX and large-scale Splunk deployments

Harper Mann Expertise: Network management, change management and virtualization

Steve Hudson Expertise: Troubleshooting failures and monitoring large-scale, N-Tier J2EE and .NET applications

Jeff Blake Expertise: Managing high availability databases infrastructures

close

Flash required to play this video.

Click here to download the free Flash Player.

Description:

Permalink: