MB-6866 Use different thresholds for kp and kv nodes

For views, since values are mostly raw JSON, compression gains are high, therefore chunk threshold (uncompressed size) can be raised, which decreases the number of kv nodes in the trees. For kp nodes, because they all have a 1024 bits bitmask in their reductions, their branching factor was a bit low, which made the trees deeper. These changes made queries and compaction faster both for generic btree tests and evperf lucky8 views (lucky8.conf and lucky8-3d.conf). Inserts and updates also get a small speedup. Change-Id: I2a94ec6005b881c81fd5295a2c89c7657cc6b19f Reviewed-on: http://review.couchbase.org/21484 Reviewed-by: Damien Katz <damien@couchbase.com> Reviewed-by: Volker Mische <volker.mische@gmail.com> Reviewed-by: Farshid Ghods <farshid@couchbase.com> Tested-by: Farshid Ghods <farshid@couchbase.com>
couchbase · Oct 11, 2012 · c62a4b2 · c62a4b2
1 parent b6454ca
commit c62a4b2
Show file tree

Hide file tree

Showing 7 changed files with 66 additions and 28 deletions.
diff --git a/src/couch_set_view/src/couch_set_view_group.erl b/src/couch_set_view/src/couch_set_view_group.erl
@@ -38,7 +38,8 @@
                             pid()) -> no_return()).
 
 -define(TIMEOUT, 3000).
--define(BTREE_CHUNK_THRESHOLD, 5120).
+-define(BTREE_KV_CHUNK_THRESHOLD, 7168).
+-define(BTREE_KP_CHUNK_THRESHOLD, 6144).
 
 -define(root_dir(State), element(1, State#state.init_args)).
 -define(set_name(State), element(2, State#state.init_args)).
@@ -1619,7 +1620,8 @@ init_group(Fd, Group, IndexHeader) ->
             First, Rest)
     end,
     BtreeOptions = [
-        {chunk_threshold, ?BTREE_CHUNK_THRESHOLD},
+        {kv_chunk_threshold, ?BTREE_KV_CHUNK_THRESHOLD},
+        {kp_chunk_threshold, ?BTREE_KP_CHUNK_THRESHOLD},
         {binary_mode, true}
     ],
     {ok, IdBtree} = couch_btree:open(

diff --git a/src/couchdb/couch_btree.erl b/src/couchdb/couch_btree.erl
@@ -86,8 +86,10 @@ set_options(Bt, [{less, Less}|Rest]) ->
     set_options(Bt#btree{less=Less}, Rest);
 set_options(Bt, [{reduce, Reduce}|Rest]) ->
     set_options(Bt#btree{reduce=Reduce}, Rest);
-set_options(Bt, [{chunk_threshold, Threshold}|Rest]) ->
-    set_options(Bt#btree{chunk_threshold = Threshold}, Rest);
+set_options(Bt, [{kv_chunk_threshold, Threshold}|Rest]) ->
+    set_options(Bt#btree{kv_chunk_threshold = Threshold}, Rest);
+set_options(Bt, [{kp_chunk_threshold, Threshold}|Rest]) ->
+    set_options(Bt#btree{kp_chunk_threshold = Threshold}, Rest);
 set_options(#btree{root = Root} = Bt, [{binary_mode, true}|Rest]) when is_binary(Root) ->
     <<Pointer:?POINTER_BITS, Size:?TREE_SIZE_BITS, Red0/binary>> = Root,
     Red = binary:copy(Red0),
@@ -362,12 +364,12 @@ complete_root(Bt, KPs) ->
     {ok, ResultKeyPointers, Bt2} = write_node(Bt, kp_node, KPs),
     complete_root(Bt2, ResultKeyPointers).
 
-%%%%%%%%%%%%% The chunkify function sucks! %%%%%%%%%%%%%
-% It is inaccurate as it does not account for compression when blocks are
-% written. Plus with the "case byte_size(term_to_binary(InList)) of" code
-% it's probably really inefficient.
+chunkify(#btree{kp_chunk_threshold = T}, kp_node, InList) ->
+    chunkify(T, InList);
+chunkify(#btree{kv_chunk_threshold = T}, kv_node, InList) ->
+    chunkify(T, InList).
 
-chunkify(#btree{chunk_threshold = ChunkThreshold0}, InList) ->
+chunkify(ChunkThreshold0, InList) ->
     case ?term_size(InList) of
     Size when Size > ChunkThreshold0 ->
         NumberOfChunksLikely = ((Size div ChunkThreshold0) + 1),
@@ -499,7 +501,7 @@ encode_node_iolist([{K, V}|RestKvs], Acc) ->
 
 write_node(#btree{fd = Fd, binary_mode = BinMode} = Bt, NodeType, NodeList) ->
     % split up nodes into smaller sizes
-    NodeListList = chunkify(Bt, NodeList),
+    NodeListList = chunkify(Bt, NodeType, NodeList),
     % now write out each chunk and return the KeyPointer pairs for those nodes
     ResultList = [
         begin
@@ -1104,7 +1106,8 @@ stats(#btree{root = Root, fd = Fd} = Bt) ->
         {btree_size, TreeSize},
         {file_size, FileSize},
         {fragmentation, Frag},
-        {chunk_threshold, Bt#btree.chunk_threshold} | Stats
+        {kv_chunk_threshold, Bt#btree.kv_chunk_threshold},
+        {kp_chunk_threshold, Bt#btree.kp_chunk_threshold} | Stats
     ].
 
 collect_stats(nil, _Bt, 0, 0, 0, 0, 0, 0) ->

diff --git a/src/couchdb/couch_btree_copy.erl b/src/couchdb/couch_btree_copy.erl
@@ -24,7 +24,8 @@
     before_kv_write = nil,
     user_acc = [],
     filter = fun(_) -> true end,
-    chunk_threshold,
+    kv_chunk_threshold,
+    kp_chunk_threshold,
     nodes = array:new(),
     cur_level = 1,
     max_level = 1,
@@ -44,7 +45,8 @@ copy(Btree, Fd, Options) ->
     Acc0 = #acc{
         btree = Btree,
         fd = Fd,
-        chunk_threshold = Btree#btree.chunk_threshold
+        kv_chunk_threshold = Btree#btree.kv_chunk_threshold,
+        kp_chunk_threshold = Btree#btree.kp_chunk_threshold
     },
     Acc = apply_options(Options, Acc0),
     {ok, _, #acc{cur_level = 1} = FinalAcc0} = couch_btree:fold(
@@ -58,7 +60,8 @@ from_sorted_file(EmptyBtree, SortedFileName, DestFd, BinToKvFun) ->
     Acc = #acc{
         btree = EmptyBtree,
         fd = DestFd,
-        chunk_threshold = EmptyBtree#btree.chunk_threshold
+        kv_chunk_threshold = EmptyBtree#btree.kv_chunk_threshold,
+        kp_chunk_threshold = EmptyBtree#btree.kp_chunk_threshold
     },
     {ok, SourceFd} = file:open(SortedFileName, [read, raw, binary, read_ahead]),
     {ok, Acc2} = try
@@ -105,7 +108,8 @@ file_sort_output_fun(OrigBtree, Fd, Options) ->
     Acc0 = #acc{
         btree = OrigBtree,
         fd = Fd,
-        chunk_threshold = OrigBtree#btree.chunk_threshold
+        kv_chunk_threshold = OrigBtree#btree.kv_chunk_threshold,
+        kp_chunk_threshold = OrigBtree#btree.kp_chunk_threshold
     },
     Acc = apply_options(Options, Acc0),
     fun(Item) -> file_sort_loop(Item, Acc) end.
@@ -131,8 +135,10 @@ apply_options([{filter, Fun} | Rest], Acc) ->
     apply_options(Rest, Acc#acc{filter = Fun});
 apply_options([override | Rest], Acc) ->
     apply_options(Rest, Acc);
-apply_options([{chunk_threshold, Threshold} | Rest], Acc) ->
-    apply_options(Rest, Acc#acc{chunk_threshold = Threshold}).
+apply_options([{kv_chunk_threshold, Threshold} | Rest], Acc) ->
+    apply_options(Rest, Acc#acc{kv_chunk_threshold = Threshold});
+apply_options([{kp_chunk_threshold, Threshold} | Rest], Acc) ->
+    apply_options(Rest, Acc#acc{kp_chunk_threshold = Threshold}).
 
 
 extract(#acc{btree = #btree{extract_kv = Extract}}, Value) ->
@@ -207,7 +213,7 @@ fold_copy(Item, ItemSize, #acc{cur_level = 1} = Acc) ->
     Kv = extract(Acc, Item),
     LeafSize2 = LeafSize + ItemSize,
     Values2 = [Kv | Values],
-    NextAcc = case LeafSize2 >= Acc#acc.chunk_threshold of
+    NextAcc = case LeafSize2 >= Acc#acc.kv_chunk_threshold of
     true ->
         {LeafState, Acc2} = flush_leaf(Values2, Acc),
         {K, _V} = Kv,
@@ -241,7 +247,7 @@ bubble_up({Key, NodeState}, Level, Acc) ->
         {Size, NextLevelNodes} = array:get(Level + 1, Acc2#acc.nodes),
         NextLevelNodes2 = [Kp | NextLevelNodes],
         Size2 = Size + KpSize,
-        case Size2 >= Acc#acc.chunk_threshold of
+        case Size2 >= Acc#acc.kp_chunk_threshold of
         true ->
             {ok, NewNodeState} = write_kp_node(
                 Acc2, lists:reverse(NextLevelNodes2)),

diff --git a/src/couchdb/couch_db.hrl b/src/couchdb/couch_db.hrl
@@ -286,6 +286,7 @@
     assemble_kv = fun(Key, Value) -> {Key, Value} end,
     less = fun(A, B) -> A < B end,
     reduce = nil,
-    chunk_threshold = 16#4ff,
+    kv_chunk_threshold = 16#4ff,
+    kp_chunk_threshold = 2 * 16#4ff,
     binary_mode = false
 }).
diff --git a/test/etap/020-btree-basics.t b/test/etap/020-btree-basics.t
@@ -24,7 +24,8 @@ rows() -> 250.
     assemble_kv,
     less,
     reduce,
-    chunk_threshold = 16#4ff,
+    kv_chunk_threshold = 16#4ff,
+    kp_chunk_threshold = 2 * 16#4ff,
     binary_mode = false
 }).
 

diff --git a/test/etap/022-btree-copy.t b/test/etap/022-btree-copy.t
@@ -21,7 +21,8 @@
     assemble_kv,
     less,
     reduce,
-    chunk_threshold = 16#4ff,
+    kv_chunk_threshold = 16#4ff,
+    kp_chunk_threshold = 2 * 16#4ff,
     binary_mode = false
 }).
 

diff --git a/test/etap/023-btree-guided-purge.t b/test/etap/023-btree-guided-purge.t
@@ -48,7 +48,11 @@ no_purged_items_test() ->
     end,
 
     {ok, Fd} = couch_file:open(filename(), [create, overwrite]),
-    {ok, Btree} = couch_btree:open(nil, Fd, [{reduce, ReduceFun}, {chunk_threshold, 6 * 1024}]),
+    {ok, Btree} = couch_btree:open(nil, Fd, [
+        {reduce, ReduceFun},
+        {kv_chunk_threshold, 6 * 1024},
+        {kp_chunk_threshold, 6 * 1024}
+    ]),
 
     N = 211341,
     KVs = [{I, I} || I <- lists:seq(1, N)],
@@ -86,7 +90,11 @@ all_purged_items_test() ->
     end,
 
     {ok, Fd} = couch_file:open(filename(), [create, overwrite]),
-    {ok, Btree} = couch_btree:open(nil, Fd, [{reduce, ReduceFun}, {chunk_threshold, 6 * 1024}]),
+    {ok, Btree} = couch_btree:open(nil, Fd, [
+        {reduce, ReduceFun},
+        {kv_chunk_threshold, 6 * 1024},
+        {kp_chunk_threshold, 6 * 1024}
+    ]),
 
     N = 211341,
     KVs = [{I, I} || I <- lists:seq(1, N)],
@@ -128,7 +136,11 @@ partial_purges_test() ->
     end,
 
     {ok, Fd} = couch_file:open(filename(), [create, overwrite]),
-    {ok, Btree} = couch_btree:open(nil, Fd, [{reduce, ReduceFun}, {chunk_threshold, 6 * 1024}]),
+    {ok, Btree} = couch_btree:open(nil, Fd, [
+        {reduce, ReduceFun},
+        {kv_chunk_threshold, 6 * 1024},
+        {kp_chunk_threshold, 6 * 1024}
+    ]),
 
     N = 211341,
     KVs = [{I, I} || I <- lists:seq(1, N)],
@@ -190,7 +202,11 @@ partial_purges_test_2() ->
     end,
 
     {ok, Fd} = couch_file:open(filename(), [create, overwrite]),
-    {ok, Btree} = couch_btree:open(nil, Fd, [{reduce, ReduceFun}, {chunk_threshold, 6 * 1024}]),
+    {ok, Btree} = couch_btree:open(nil, Fd, [
+        {reduce, ReduceFun},
+        {kv_chunk_threshold, 6 * 1024},
+        {kp_chunk_threshold, 6 * 1024}
+    ]),
 
     N = 320000,
     KVs = [{iolist_to_binary(io_lib:format("doc_~6..0b", [I])), {I rem 64, I}} || I <- lists:seq(1, N)],
@@ -257,7 +273,11 @@ partial_purges_test_with_stop() ->
     end,
 
     {ok, Fd} = couch_file:open(filename(), [create, overwrite]),
-    {ok, Btree} = couch_btree:open(nil, Fd, [{reduce, ReduceFun}, {chunk_threshold, 6 * 1024}]),
+    {ok, Btree} = couch_btree:open(nil, Fd, [
+        {reduce, ReduceFun},
+        {kv_chunk_threshold, 6 * 1024},
+        {kp_chunk_threshold, 6 * 1024}
+    ]),
 
     N = 211341,
     KVs = [{I, I} || I <- lists:seq(1, N)],
@@ -323,7 +343,11 @@ add_remove_and_purge_test() ->
     end,
 
     {ok, Fd} = couch_file:open(filename(), [create, overwrite]),
-    {ok, Btree} = couch_btree:open(nil, Fd, [{reduce, ReduceFun}, {chunk_threshold, 6 * 1024}]),
+    {ok, Btree} = couch_btree:open(nil, Fd, [
+        {reduce, ReduceFun},
+        {kv_chunk_threshold, 6 * 1024},
+        {kp_chunk_threshold, 6 * 1024}
+    ]),
 
     N = 211341,
     KVs = [{I, I} || I <- lists:seq(1, N)],