Time to live for load commands

A method, comprising:

• receiving, from a memory bus connected to a host interface of a memory module having a plurality of memory devices with different latencies in accessing data stored in the memory devices, a load command from a central processing unit, the load command specifying a memory address and a requested time duration for responding to the load command;
• identifying a first memory device, among the plurality of memory devices, that currently stores an item at the memory address;
• and
• determining whether a latency of the first memory device is sufficient to retrieve the item from the first memory device as a response to the load command within the requested time duration.

A non-transitory computer storage medium storing instructions which, when executed in an apparatus having a plurality of memory devices with different latencies in accessing data stored in the memory devices, cause the apparatus to perform a method, the method comprising:

• receiving, from a memory bus connected to a host interface of a memory module, a load command from a central processing unit, the load command specifying a memory address and a requested time duration for responding to the load command;
• identifying a first memory device, among the plurality of memory devices, that currently stores an item at the memory address;
• and
• determining whether a latency of the first memory device is sufficient to retrieve the item from the first memory device as a response to the load command within the requested time duration.