AliveRCM™

Real-Time Capacity Management for IT Systems

Overview

Capacity management has traditionally been a complex, time-consuming task. To do it effectively, organizations require highly experienced staff that deeply understand their applications and supporting infrastructure. They use that knowledge to build complex mathematical models of system behavior that can then be used to plan for future capacity needs. In the typical environment, capacity requests are issued and models executed to determine capacity requirements for a set period (e.g., six months).

Unfortunately, the supporting infrastructure changes, a new version of the application is rolled out, or the business cycles of the application change. These changes render the existing model unusable, so the capacity planners must revise and re-run the models to get an accurate, current understanding. This is because the models simply cannot adapt to real-time change. Many times, the result is capacity-related performance issues that affect end users and business bottom line. As well, the Operations team has no insight into how real-time change in the environment is affecting capacity until these performance issues manifest.

AliveRCM provides the solution through fully automated, real-time capacity management. AliveRCM employs patented analytics to make capacity planning approachable for all IT environments. Alive understands the normal behavior of your systems through advanced learning analytics and uses that knowledge to extrapolate the trends of your systems. It adapts in real-time to any change and lets you know when you're going to run out of capacity. AliveRCM does this in an automated fashion, with no complex models to build and run.

Problems and Challenges

Capacity Planning cycles are long and can't account for real-time change to business cycles, applications or devices
Building models/profiles for Capacity Management is complex and time-consuming
Forecasting is often based on linear analysis and extrapolation
Lack of early warning before capacity is exceeded
No way for Operations to understand capacity risks for devices, technology tiers or applications

The Solution

Automatically understands normal behavior and business cycles in performance data and forecasts future behavior
Auto-Learning technologies automatically adapt to any change with no models or rules required
Sophisticated, real-time analytics accurately model capacity requirements
Automatically identifies and alerts to capacity issues before users and applications are impacted
Provides a single Capacity Risk Score^TM to the Operations team who can take action to address issues

With AliveRCM, you can select any application/service, technology tier or device. You can then select any metric and set the SLA level (i.e., the value at which capacity level has been breached). AliveRCM then forecasts the future behavior. It's as simple as that.

As part of the analysis, AliveRCM provides the following information:

Min/Max dates and capacity values
Capacity breach dates
Correlated performance metrics
Performance cycles of the metric

Figure 1 - Automated Capacity Forecasting with Performance Cycle Display

AliveRCM also automatically computes a Capacity Risk Score for your applications/services, tiers of technology and individual devices. This mathematical score represents the amount of risk for capacity-related issues. With AliveRCM you won't need highly skilled staff to focus solely on capacity management. You'll have be provided a single measure of capacity directly to the Operations team who can act on it. Capacity Risk Scores are generated automatically because AliveRCM's monitoring adapters are equipped with intelligence that tells it which metrics are capacity-related. Of course, users are provided with the ability to fine tune and customize for their specific environment. Clicking on the Capacity Risk Score will tell you exactly what resources within the container are exhibiting capacity-related constraints and with a single click you can see a view like Figure 1 above for capacity-limited metrics. With AliveRCM, you can even create capacity Supermetrics by arithmetically combining metrics from any resource. Alive treats Supermetrics just like any single metric, learning the normal behavior over time so that capacity issues can be proactively alerted to and accurately forecasted.

Figure 2 - Capacity Risk Score for an Application Shows High Risk for a Capacity Issue

AliveRCM also provides the ability to create "what if" scenarios. You can vary system loads and determine (based on dynamic system behavior rather than static models) when and where your infrastructure will experience capacity-related issues. As business cycles, applications or infrastructure change, you can re-run scenarios without time-consuming, complex model changes because AliveRCM automatically understands these behavioral changes. AliveRCM's "what if" analysis allows you to effectively plan capacity when deploying or migrating applications, major infrastructure changes, mergers, or any other events that result in massive change and uncertainty.

Of course, Alive RCM can do all these things because it is leveraging the existing Alive platform. It uses Alive's Optimized Data Store to store and access virtually unlimited amounts of granular performance data to perform real-time capacity management.

Summary

With AliveRCM, any organization can benefit from completely automated, real-time capacity management without requiring a small army of highly skilled staff to endlessly re-model system performance. AliveRCM's powerful analytics react to any change in real-time and give you proactive alerts to pending capacity issues. It gives you the ability to regularly run "what if" capacity planning exercises without requiring costly model changes as your systems change.