Free memory manager scheme and cache

A method of managing free memory in a memory system having a plurality of memory banks in a plurality of memory channels, said method comprising:

• creating a plurality of free lists having entries with addresses of free memory locations, wherein a free list is created for each memory bank in each memory channel;
• and
• writing entries from said free lists to a global cache, wherein said entries are written to said global cache are distributed between memory channels and memory banks,further comprising:
• writing entries from a first free list to said global cache, wherein said first free list is associated with a first memory bank in a first memory channel;
• writing entries from a second free list to said global cache, wherein said second free lint is associated with a second memory bank in a second memory channel;
• writing entries from a third free lint to said global cache, wherein said third free list is associated with a third memory bank in said first memory channel;
• and
• writing entries from a fourth free list to said global cache, wherein said fourth free list is associated with a fourth memory bank in said second memory channel.

A method of managing free memory, said method comprising:

• creating a free list having entries with addresses of free memory locations;
• caching a portion of said free list in a cache having an upper threshold and a lower threshold;
• reading a block of entries from said free to said cache when the number of valid entries in said cache reaches said lower threshold;
• and
• writing a block of entries from said cache to said free list when the number of modified entries in said cache reaches said upper threshold.

A method of managing free memory, said method comprising:

• creating a plurality of free lists having entries with addresses of free memory locations, wherein said free lists are associated with a plurality of memory banks in a plurality of memory channels;
• writing entries from said free lists to a plurality of caches;
• and
• writing entries from said caches to a global cache, wherein said entries written to said global cache are distributed between said memory channels and said memory banks, further comprising:
• writing entries from a first free list to said global cache, wherein said first free list is associated with a first memory bank in a first memory channel;
• and
• writing entries from a second free list to said global cache, wherein said second free list is associated with a second memory bank in a second memory channel.