13December2017

OpsDataStore Announces Full-Stack Service Level Dashboards

OpsDataStore announces the immediate availability of cross-stack service level dashboards – providing business constituents and application owners of applications monitored by AppDynamics, Dynatrace and ExtraHop a visually appealing and easy to understand view of how their key transactions are performing mapped against the performance of the VMware virtual and physical infrastructure supporting these transactions. 

The Unique Approach and Benefits of OpsDataStore

OpsDataStore collects performance (response time and latency), throughput (calls per second and transaction per second), error, capacity, utilization (CPU and memory utilization) and contention (CPU Ready, Memory Swapping) metrics from APM tools like AppDynamics and Dynatrace, Wire Data tools like ExtraHop, and data center virtualization platforms like VMware vSphere.

OpsDataStore then uniquely calculates the end-to-end and cross-stack relationships between all of these metrics, calculates baselines for each metric and determines anomalies from the baselines. Knowing the metrics across the stack, the anomalies for all of the metrics and the relationships across the stack makes cross-stack service level dashboards possible.

For the first time this gives business constituents and application owners a 360 degree view of the performance of their business critical transactions overlaid against the performance of the virtual and physical infrastructure supporting and running these transactions.

Each of the below dashboards works in the same way. You configure the top graph for the application level KPI (transaction response time) that you want, and then you select the type of infrastructure that you want for the next 3 graphs. The dashboard then automatically finds the virtual and physical infrastructure where the transaction is running using the relationships in OpsDataStore. All of the dashboards shown below are the exact same dashboard, they are just configured differently.

These dashboards provide the following unique benefits:

  • Business constituents and application owners can easily see the performance of business critical transactions and services
  • When problems occur, IT can get out of “guilty until proven innocent mode” as the relationships between transaction performance and infrastructure performance are clearly visible
  • IT can safely “sweat” the infrastructure while maintaining real time visibility into the performance of the transactions that the business cares about.

Full Stack Service Level Dashboard for AppDynamics Monitored Transactions

For this configuration of the Service Level Dashboard, we are using AppDynamics to monitor an instance of our product running in Load Test. The first graph is the response times of the key transactions of that application. The second graph shows the CPU contention (CPU Ready) of the Virtual Server where that transaction is running. The third graph shows the CPU Ready of the Physical Host where that transaction is running. The last graph shows the latency of the datastores where that transaction is running. Yellow dots represent warnings with respect to the baselines. Red dots represent alarms with respect to the baselines.

AppDynamics/VMware Service Level Dashboard (click to expand)

Configuring this dashboard is a simple matter of making the configuration selections in the four columns below. Note that you simply specify the name of the transaction as the Primary Object in the first column The dynamic relationship mapping takes care of picking the correct items for KPI’s 2 -4.

AppDynamics/VMware Service Level Dashboard Configuration (click to expand)

Full Stack Service Level Dashboard for Dynatrace Monitored Transactions

For this configuration of the Service Level Dashboard, we are using Dynatrace AppMon to monitor the Dynatrace EasyTravel demo application. The first graph is the response times of the key transactions of that application. The second graph shows the CPU contention (CPU Ready) of the Virtual Server where that transaction is running. The third graph shows the CPU Ready of the Physical Host where that transaction is running. The last graph shows the latency of the datastores where that transaction is running. Yellow dots represent warnings with respect to the baselines. Red dots represent alarms with respect to the baselines.

Dynatrace/VMware Service Level Dashboard (click to expand)

Configuring this dashboard is a simple matter of making the configuration selections in the four columns below. Note that you simply specify the name of the transaction as the Primary Object in the first column The dynamic relationship mapping takes care of picking the correct items for KPI’s 2 -4.

 Dynatrace/VMware Service Level Dashboard Configuration (click to expand)

Full Stack Service Level Dashboard for ExtraHop Monitored Applications and Services

For this configuration of the Service Level Dashboard, we are using ExtraHop to monitor one of the key Kafka services running in the environment. The first graph is the HTTP Server Processing Time for the VM running this instance of Kafka. The second graph shows the CPU contention (CPU Ready) of the Virtual Server where that transaction is running. The third graph shows the CPU Ready of the Physical Host where that transaction is running. The last graph shows the latency of the datastores where that transaction is running. Yellow dots represent warnings with respect to the baselines. Red dots represent alarms with respect to the baselines.

ExtraHop/VMware Service Level Dashboard (click to expand)

Configuring this dashboard is a simple matter of making the configuration selections in the four columns below. Note that you simply specify the name of the HTTP Service of interest as the Primary Object in the first column. The dynamic relationship mapping takes care of picking the correct items for KPI’s 2 -4.

ExtraHop/VMware Service Level Dashboard Configuration (click to expand)

Summary

OpsDataStore has delivered a breakthrough capability in the form of easy to use and easy to customize service level dashboards developed in Tableau – the market leading BI and data analytics solution. This gives the business owners of applications the ability to easily see how their business critical transactions and services are performing and allows IT and the business to effectively collaborate around service levels and the costs to support these applications and transactions.