Get API access to the information needed to effectively manage and review the health of your cluster.
Cassandra Monitoring API
Our Cassandra Monitoring API enables access to Apache Cassandra performance metrics including CPU utilization, disk utilization, reads and writes, latency tasks, pending compactions, live cells and tombstones per read, SSTable, thread pool task statistics, and more. The Instaclustr Console also delivers a visual view of these metrics as part of our monitoring dashboard.
Get Started in Minutes
For all Apache Cassandra clusters and nodes that are managed through the Instaclustr Managed Platform, you can get started monitoring your Cassandra database in just a few minutes by creating your cluster from a centralized console. All data metrics data can be viewed either on a per-node basis or for all nodes in a cluster, helping you diagnose any issues and enabling better capacity planning. All available metrics are updated every 20 seconds.
The metric information viewable on our monitoring dashboard is available as part of our console once you authenticate the monitoring API.
Advantages of the Cassandra Monitoring Tool
View a summary of key metrics to understand the health of your cluster at a glance.
Get a tabular view of key metrics—ideal for understanding relatively static information (for example, tombstones-per-read over the last 5 minutes).
Graph View Key Metrics
A graphical view of metric values over time—ideal for pinpointing both the times of sudden metric changes and identifying broader trends over time.
Cluster Health Check
Instaclustr’s Cluster Health Check exposes a number of indicators to help you understand your cluster’s long-term performance. There are three potential states for each indicator:
Green represents a healthy state;
Amber represents a warning state;
Red represents a failed state.
Instaclustr’s monitoring API allows you to integrate all monitoring information from your Instaclustr managed cluster with external monitoring tools such as Datadog or Prometheus. This makes it easy to construct a single monitoring view of your application stack.
Our Support Portal has details of all the monitoring tools that we provide and how you can get the most of our console monitoring feature.
The Instaclustr Managed Platform provides a Monitoring Dashboard through our console. This enables important and quick access to key metrics that help you check the health of your cluster and better manage your overall Cassandra performance.
Sustained high CPU usage is an indicator that your cluster is reaching processing capacity, and you may need to consider adding capacity to cope with any increase in load.
Keeping an eye on disk usage should be a key part of your capacity planning. Maintaining disk usage at less than 70% during normal operations is recommended.
This metric helps you identify changing levels of load on your cluster. A significantly uneven distribution could indicate driver misconfiguration or data model issues.
Compactions are a continuous, background process in Cassandra. A high or increasing number of pending compactions indicates that your cluster does not have sufficient capacity to process the level of operations it requires.
Repair is a Cassandra operation that ensures data consistency is eventually attained across the ring. Repairs are a scheduled operation and represent an additional load on the cluster.
This metric checks size of the largest partition in each table. We recommended limiting the maximum partition size to 10MB for optimal performance, with 100MB as a upper limit for ongoing stability. Large partitions may significantly impact the performance of Cassandra operation.
Checks the average ratio of the number of tombstones and live cells per read in each table. High ratios of tombstones to live cells (greater than 5x as a starting guide) can cause substantially reduced performance in reads from a table.
Monitor the latest mean and maximum metrics for each column family, averaged across the cluster.
High numbers of SSTables per read (more than 3 or 4, as a starting guide) can reduce read performance, and if read performance is below desired levels you may need to change compaction strategy for the affected column family.
This indicator checks the replication class used for each keyspace. NetworkTopologyStrategy is highly recommended to ensure data is replicated in order to minimize the impact of infrastructure failure.
These metrics are associated with each stage of Cassandra’s Staged Event-Driven Architecture (SEDA).