Which Metrics Should I Monitor? - Amazon MemoryDB for Redis

Which Metrics Should I Monitor?

The following CloudWatch metrics offer good insight into MemoryDB performance. In most cases, we recommend that you set CloudWatch alarms for these metrics so that you can take corrective action before performance issues occur.

CPUUtilization

This is a host-level metric reported as a percentage. For more information, see Host-Level Metrics.

For smaller node types with 2vCPUs or less, use the CPUUtilization metric to monitor your workload.

Generally speaking, we suggest you set your threshold at 90% of your available CPU. Because Redis is single-threaded, the actual threshold value should be calculated as a fraction of the node's total capacity. For example, suppose you are using a node type that has two cores. In this case, the threshold for CPUUtilization would be 90/2, or 45%. To find the number of cores (vCPUs) your node type has, see MemoryDB Pricing.

You will need to determine your own threshold, based on the number of cores in the node that you are using. If you exceed this threshold, and your main workload is from read requests, scale your cluster out by adding read replicas. If the main workload is from write requests, we recommend that you add more shards to distribute the write workload across more primary nodes.

Tip

Instead of using the Host-Level metric CPUUtilization, you might be able to use the Redis metric EngineCPUUtilization, which reports the percentage of usage on the Redis engine core. To see if this metric is available on your nodes and for more information, see Metrics for MemoryDB.

For larger node types with 4vCPUs or more, you may want to use the EngineCPUUtilization metric, which reports the percentage of usage on the Redis engine core. To see if this metric is available on your nodes and for more information, see Metrics for MemoryDB.

EngineCPUUtilization

For larger node types with 4vCPUs or more, you may want to use the EngineCPUUtilization metric, which reports the percentage of usage on the Redis engine core. To see if this metric is available on your nodes and for more information, see Metrics for MemoryDB.

SwapUsage

This is a host-level metric reported in bytes. For more information, see Host-Level Metrics.

This metric should not exceed 50 MB.

Evictions

This is a engine metric. We recommend that you determine your own alarm threshold for this metric based on your application needs.

CurrConnections

This is a engine metric. We recommend that you determine your own alarm threshold for this metric based on your application needs.

An increasing number of CurrConnections might indicate a problem with your application; you will need to investigate the application behavior to address this issue.

Memory

Memory is a core aspect of Redis. Understanding the memory utilization of your cluster is necessary to avoid data loss and accommodate future growth of your dataset. Statistics about the memory utilization of a node are available in the memory section of the Redis INFO command.

Network

One of the determining factors for the network bandwidth capacity of your cluster is the node type you have selected. For more information about the network capacity of your node, see Amazon MemoryDB pricing.

Replication

The volume of data being replicated is visible via the ReplicationBytes metric. You can monitor MaxReplicationThroughput against the replication capacity throughput. It is recommended to add more shards when reaching the maximum replication capacity throughput.

ReplicationDelayedWriteCommands can also indicate if the workload is exceeding the maximum replication capacity throughput. For more information about replication in MemoryDB, see Understanding MemoryDB replication