Andrea Danti - Fotolia
An essential part of using Amazon ElastiCache is monitoring cache utilization and performance. And cost allocation tagging helps organizations monitor ElastiCache and keep cache costs at bay.
Native services such as Amazon CloudWatch can monitor ElastiCache metrics associated with performance, including host-level metrics, Memcached-specific metrics and Redis-specific metrics. Metrics are collected for each cache node every 60 seconds.
Host-level metrics monitor ElastiCache activity at the hardware level and include CPU utilization percentage, available memory on the host, bytes the host has read from or written to the network and the amount of content swapped on the host. There is also an extensive selection of metrics available to monitor ElastiCache -- both Memcached and Redis engines -- including the amount of content written to or read from the cache, the percentage of cache hits and misses, and the number of key commands -- such as Get -- the cache has received.
There is no standard to dictate which metrics IT teams should follow; choose metrics that yield the best overall view for your organization's needs. In general, host-level metrics are a good place to start to examine general performance. However, engine-specific metrics can offer more insights into the Memcached or Redis performance. For example, excessive CPU utilization and swap usage may correspond to lower application performance, while excessive evictions or an increasing number of current connections may suggest cache workload problems.
IT teams can choose to see other metrics for detailed information on cache behavior. ReplicationBytes or ReplicationLag metrics can tell administrators how much replication data Redis is producing and help identify latency in the replication process.
Don't overlook the importance of scheduling a maintenance window in Amazon ElastiCache. Cache software is typically patched during the maintenance window. AWS could drive this by adding a new version of Memcached or Redis to the available repertoire, or an end user could request this when moving from Memcached 1.4.14 to 1.4.24. IT teams choose a maintenance window and Amazon ElastiCache clusters become unavailable during that time. This can affect app performance.
Choose between two popular caching engines in ElastiCache
Choose proper tools for monitoring AWS
Maximize efficiency through performance monitoring
Dig Deeper on AWS database management
Related Q&A from Stephen J. Bigelow
Eliciting performance requirements from business end users necessitates a clearly defined scope and the right set of questions. Expert Mary Gorman ... Continue Reading
Requirements fall into three categories: business, user and software. See examples of each one, as well as what constitutes functional and ... Continue Reading
Navigating data center malfunctions when hardware is off premises can be tricky. Organizations must have strong SLAs with their colo provider to ... Continue Reading