Self-optimizable code for optimizing execution of tasks and allocation of memory in a data processing system

An apparatus for optimizing execution of tasks, comprising:

• a processor;
• and
• a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to:
• for each one of a set of possible code paths, determine a characteristic of performance for the code path by executing a task using the code path and observing performance;
• observe performance of each code path, wherein observing performance further comprises counting a number of cycles needed to execute a code path;
• and
• select the code path determined to have the best performance so that the selected code path is executed in subsequent executions of the task;
• wherein the selecting comprises selecting a memory allocation module for the task, the memory allocation module comprises:
• partitioning a memory into a plurality of sections, wherein each section in the plurality of sections has a size smaller than a size of the memory, wherein each section in the plurality of sections is further divided into granules, and wherein each section in the plurality of sections has a granule size different than the other sections in the plurality of sections;
• comparing a size of an allocation request from the task to the granule size within each section of the plurality of sections in order to identify a section with a granule size that is closest to the size of the allocation request;
• and responsive to identifying the section with the granule size that is closest to the size of the allocation request, allocating a granule from the section with the granule size in response to the allocation request.

A computer-implemented method to improve allocation of memory in a server, the method comprising:

• partitioning a memory into a plurality of sections, wherein each section in the plurality of sections has a size smaller than a size of the memory, wherein each section in the plurality of sections is further divided into granules, and wherein each section in the plurality of sections has a granule size different than the other sections in the plurality of sections;
• comparing a size of an allocation request from a task to the granule size within each section of the plurality of sections in order to identify a section with a granule size that is closest to the size of the allocation request;
• and
• responsive to identifying the section with the granule size that is closest to the size of the allocation request, allocating a granule from the section with the granule size in response to the allocation request.