Question posted in 
I am little confused when reading the docs about ElastiCache network limits. Each node type has it’s own bandwidth defined in Gbps. There are two metrics counting network traffic for Redis, these metrics are NetworkBytesIn and NetworkBytesOut. In the docs I’ve found the following statement:
The AWS/ElastiCache namespace includes the following host-level metrics for individual cache nodes. These metrics are measured and published for each Cache node in 60-second intervals
Does it mean the numbers I see on the charts are telling me how much do we transmit in one minute? For example baseline bandwidth for cache.t2.medium node type is 0.256 Gbps. That is 32 MBps. Here comes the question – what is my limit in NetworkBytesIn/Out metrics? Would it be 32 MB in there or 1920 MB assuming they’re published for 60 seconds intervals? Thanks in advance for any clarification on this! Regards!
2
Answers
ElastiCache Host-Level Metrics describe NetworkBytesOut as
In other words, each data point should represent the number of bytes used in a specific minute. Hence, 32MBps network usage will appear as 1920MB per minute in the metrics.
Yes. In CloudWatch use the
SUMstatistic. Each datapoint will then represent the total number of bytes sent out in the given Period selected.If a node offers 5 Gbps network bandwidth then it means:
of data can be transferred in a given minute. Traffic exceeding this limit will be queued. If this exceeds for a longer duration then the packets will be dropped.
NOTE: The network bandwidth limit is separate for Inbound and Outbound traffic meaning 5 Gbps for inbound and 5 Gbps for outbound.