> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://akamai.ferndocs.com/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://akamai.ferndocs.com/_mcp/server. # Resource tier limitations These limitations are unique for each resource tier. There are currently three resource tiers available, **Dynamic Compute**, **Basic Compute** and **Enterprise Compute**. > 👍 The limits for the EdgeWorker IDs created without a resource tier will not change. By default, these EdgeWorkers functions will execute using **Dynamic Compute**. The Dynamic Compute limits are the same as the limits applied prior to the introduction of resource tiers.
Event handler Description Basic Compute Dynamic Compute Enterprise Compute
all event handlers Maximum CPU time during initialization 60 milliseconds 60 milliseconds 300 milliseconds
Maximum wall time during initialization 200 milliseconds 500 milliseconds 500 milliseconds
onClientRequest onOriginRequest onOriginResponse onClientResponse Maximum memory usage per event handler 1.5 MB 2.5 MB 4 MB
Maximum CPU time per event handler 10 milliseconds 20 milliseconds 70 milliseconds
Maximum wall time per event handler 4 seconds 5.5 seconds 10 seconds
Maximum number of HTTP sub-requests allowed from a parent request, per event handler
**Note:** You can use content fetched using the onClientRequest event handler to modify the original end-user request's cache key. For cache keys modified in this way, you need to use purge by CP Code and not by URL. For instructions on how to purge a CP code see [Purge Content by CP code](https://techdocs.akamai.com/purge-cache/docs/purge-content-cp-code). For instructions on how to create a CP code see [Create a CP code](https://techdocs.akamai.com/cp-codes/docs/welcome-cp-codes-rpt-grps). 		2 4 10
Maximum number of HTTP sub-requests allowed in parallel per request 2 4 10
Maximum wall time per HTTP sub-request 4 seconds 5.5 seconds 10 seconds
Maximum response size per HTTP sub-request 5 MB 5 MB 8 MB
responseProvider Maximum memory usage for responseProvider 2 MB 4 MB 6 MB
Maximum CPU time for responseProvider 100 milliseconds 200 milliseconds 300 milliseconds
Maximum wall time for responseProvider 4 seconds 8 seconds 10 seconds
Maximum number of HTTP sub-requests allowed for responseProvider 50 50 50
Maximum number of HTTP sub-requests allowed in parallel for responseProvider 5 5 10
Maximum wall time per HTTP sub-request during the execution of the responseProvider event handler 4 seconds 8 seconds 10 seconds
Maximum response size per HTTP sub-request during the execution of the responseProvider event handler 5 MB 5 MB 8 MB