jingyugao / aili

Various Concurrent Algorithms for B+ Tree

aili

References

• Palm Tree : Parallel Architecture-Friendly Latch-Free Modifications to B+ Trees on Many-Core Processors
• Mass Tree : Cache Craftiness for Fast Multicore Key-Value Storage
• B^link Tree : Efficient Locking for Concurrent Operations on B-Trees
• Prefetch B+ Tree : Improving Index Performance through Prefetching
• Prefix B Tree : Prefix B-Trees

Version Information

Version	Detail
1.0.0	Palm Tree
1.1.0	Palm Tree with point-to-point synchronization
1.2.0	Simple Prefix Palm Tree
1.2.1	synchronization optimization & remove barrier
1.2.2	adjust Palm Tree descending policy (level descend, not ideal) & sequential insertion optimization
1.2.3	adjust Palm Tree descending policy (lazy descend, great for sequential operation)
1.3.0	a major refactor
1.3.1	Prefetch Palm Tree

Have a Try

1. make third_party
                                       #  key_num   key_len
2. make generate_data && ./generate_data  10000000    10    # test data will be in ./data
                                        #  f_name  n_size  b_size  t_num  q_num  k_num
3. make palm_tree_test && ./palm_tree_test   1      4096    16384     2      8   1000000

# f_name can be 0 (sequential data) or 1 (random data);
# if your machine supports N threads, t_num can be 1, 2, ..., (N-1), not N;
# k_num can be [1, infinity), depending on how many test keys you generate

Other

• checkout example/ to get use example
• follow my 知乎专栏 for blog about this repo