Solution Package for TIBCO ActiveSpaces
The Solution Package for TIBCO ActiveSpaces is a plug-in application to RTView® Enterprise Edition that allows you to monitor the health and performance of TIBCO ActiveSpaces® instances and services in real-time.
TIBCO ActiveSpaces is a distributed in-memory data grid for building highly scalable, fault-tolerant applications. The distributed, highly-scalable nature of TIBCO ActiveSpaces often include 10s if not 100s of individual nodes which can be clustered locally or geographically across multiple datacenters which makes monitoring a distributed cluster quite a challenge. Often, performance issues on one node are negligible and are handled via fault tolerance but cluster wide problems can be catastrophic.
RTView® Enterprise Edition provides the visibility necessary to monitor such a highly distributed environment in the context of managing TIBCO Infrastructure components on up to the application layers that rely on TIBCO ActiveSpaces as a crucial caching mechanism.
View Global Displays Showing Each Space within One or More Metaspaces
A global, consolidated view is especially important when it comes to in-memory data grids. Typically, data and activity (gets / puts) are not evenly distributed across a grid that contains multiple, independently typed spaces. Data spaces of disparate sizes can be instantiated on different subsets of members, some overlapping, some not. It is essential to be able to quickly determine and remedy those situations where individual spaces may be hogging resources and causing a performance bottleneck affecting users.
View Performance and Utilization across All Members of a Cluster Hosting Each Space
When content size and activity metrics are not evenly distributed across all members within a single space, an individual member may become overloaded and cause a performance bottleneck affecting end users. TIBCO ActiveSpaces Monitor can visually identify where these “hotspots” are in each cluster. By knowing how space members interact with the space via their role and metrics, like data volume and gets/puts/takes per second, system administrators are enabled to dynamically manage the system resources in order to maintain high performance of a cluster.
Monitor TIBCO ActiveSpaces at the Application Level
No matter whether you are responsible for just ActiveSpaces, or all the TIBCO middleware components within a given application or even the entire application platform including JVMs, virtual machines, physical machines and databases, TIBCO ActiveSpaces Monitor provides the visual context that allows you to view the overall health and stability of an application as an aggregate of the health of its individual infrastructure components. RTView allows the user to drill-down on alerts at the application level to see the status of all the infrastructure components, and down to the individual alerts within any given component to quickly identify the source of the problem without having to resort to multiple different native monitoring tools.
Alerting in a Dynamic Data Grid
Though it is possible to use the TIBCO Hawk Console to define alerts on each member in the grid, one at a time, it becomes impractical when the grid gets larger (e.g. > 20 members), or when new members are being added dynamically. In this common scenario, an alert configuration mechanism that dynamically assigns alerts to each member and space as they come and go makes managing the system much easier as it undergoes change.
TIBCO ActiveSpaces Monitor adds Historical Alerts to Aid in Problem Resolutions
Naturally, you need real-time metrics to ensure that you spot alerts that might help you to head off problems before they affect the end user but often we need to diagnose the root cause of failure AFTER we get our systems back up and running. RTView collects and displays those time-stamped snapshots of metrics to help you understand what went wrong as well as provide the context on whether a given spike is normal or whether it is an outlier that needs immediate attention.
Perform Scalability and Capacity Analysis to Ensure Optimal Response Times
RTView® Enterprise Edition gives you visual confirmation of cluster-wide workloads so that you can ensure you have the optimum number of nodes required to support current and peak activity levels. Capacity analysis is also useful in determining whether you have enough compute capacity to add new applications to the ActiveSpaces cluster or whether you need to add nodes to support the additional workload. Historical data analysis also allows you to view capacity from day-to-day or minute-to-minute and is particularly useful in determining the impact of application changes and updates to system resources.
Identify Potentially Abusive Usage Patterns that affect System Performance
What users are using my resources? Who is consuming the most capacity? Am I using my system resources efficiently? Am I dedicating enough capacity to my most valuable applications or am I wasting it on lower value services? These are all common questions when we run into unexpected capacity issues. RTView Enterprise Monitor helps you to answer these questions to ensure that your resources are used efficiently for the most mission critical applications.
Cross-Correlation of ActiveSpaces Metrics with Hardware, Database and Network Metrics
So you see a backup within your ActiveSpaces environment. Is the problem in the ActiveSpaces cluster? Or is it in fact caused by excess latency within the network layer. You could take hours to track down the answer to these types of questions. Fortunately, RTView® Enterprise Edition is able to correlate events and metrics from different systems and aggregate them in a visually intuitive way so that you can instantly see the impact across multiple layers all at the same time.