PEF::CacheLRU - a very simple and very fast implementation of LRU cache in pure perl

Comparing it with Cache::LRU there are following speed differences (measured by Devel::NYTProf):

  spent 9.61s (8.97+637ms) within PEF::CacheLRU::get which was called 7500000 times, avg 1µs/call: 
  5000000 times (5.36s+0s) by main::cache_hit at line 18 of simple_bench.pl, avg 1µs/call 
  2500000 times (3.61s+637ms) by main::cache_set_hit at line 31 of simple_bench.pl, avg 2µs/call

  spent 25.3s (17.8+7.48) within Cache::LRU::get which was called 7500000 times, avg 3µs/call: 
  5000000 times (11.7s+4.91s) by main::cache_hit at line 18 of simple_bench.pl, avg 3µs/call 
  2500000 times (6.16s+2.57s) by main::cache_set_hit at line 31 of simple_bench.pl, avg 3µs/call

  spent 4.23s (4.23+2.36ms) within PEF::CacheLRU::set which was called 1320720 times, avg 3µs/call: 
  1310720 times (4.21s+2.36ms) by main::cache_set at line 56 of simple_bench.pl, avg 3µs/call 
     5000 times (13.0ms+0s) by main::cache_hit at line 16 of simple_bench.pl, avg 3µs/call 
     5000 times (10.6ms+0s) by main::cache_set_hit at line 25 of simple_bench.pl, avg 2µs/call

	
  spent 9.46s (7.38+2.08) within Cache::LRU::set which was called 1320720 times, avg 7µs/call:
  1310720 times (7.32s+2.06s) by main::cache_set at line 56 of simple_bench.pl, avg 7µs/call
     5000 times (28.5ms+11.2ms) by main::cache_hit at line 16 of simple_bench.pl, avg 8µs/call
     5000 times (24.9ms+9.10ms) by main::cache_set_hit at line 25 of simple_bench.pl, avg 7µs/call



Github: https://github.com/pef-secure/pef-cache-lru