1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
|
%%
%% %CopyrightBegin%
%%
%% Copyright Ericsson AB 1996-2016. All Rights Reserved.
%%
%% Licensed under the Apache License, Version 2.0 (the "License");
%% you may not use this file except in compliance with the License.
%% You may obtain a copy of the License at
%%
%% http://www.apache.org/licenses/LICENSE-2.0
%%
%% Unless required by applicable law or agreed to in writing, software
%% distributed under the License is distributed on an "AS IS" BASIS,
%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
%% See the License for the specific language governing permissions and
%% limitations under the License.
%%
%% %CopyrightEnd%
%%
-module(pool).
%% Supplies a computational pool of processors.
%% The chief user interface function here is get_node()
%% Which returns the name of the nodes in the pool
%% with the least load !!!!
%% This function is callable from any node including the master
%% That is part of the pool
%% nodes are scheduled on a per usgae basis and per load basis,
%% Whenever we use a node, we put at the end of the queue, and whenever
%% a node report a change in load, we insert it accordingly
% User interface Exports ...
-export([start/1,
start/2,
stop/0,
get_nodes/0,
get_nodes_and_load/0,
get_node/0,
pspawn/3,
attach/1,
pspawn_link/3]).
%% Internal Exports
-export([statistic_collector/0,
do_spawn/4,
init/1,
handle_call/3,
handle_cast/2,
handle_info/2,
terminate/2]).
%% User interface
%% Start up using the .hosts.erlang file
-spec start(Name) -> Nodes when
Name :: atom(),
Nodes :: [node()].
start(Name) ->
start(Name,[]).
-spec start(Name, Args) -> Nodes when
Name :: atom(),
Args :: string(),
Nodes :: [node()].
start(Name, Args) when is_atom(Name) ->
_ = gen_server:start({global, pool_master}, pool, [], []),
Hosts = net_adm:host_file(),
Nodes = start_nodes(Hosts, Name, Args),
lists:foreach(fun attach/1, Nodes),
Nodes.
%%
%% Interface functions ...
%%
-spec get_nodes() -> [node()].
get_nodes() ->
get_elements(2, get_nodes_and_load()).
-spec attach(Node) -> 'already_attached' | 'attached' when
Node :: node().
attach(Node) ->
gen_server:call({global, pool_master}, {attach, Node}).
get_nodes_and_load() ->
gen_server:call({global, pool_master}, get_nodes).
-spec get_node() -> node().
get_node() ->
gen_server:call({global, pool_master}, get_node).
-spec pspawn(Mod, Fun, Args) -> pid() when
Mod :: module(),
Fun :: atom(),
Args :: [term()].
pspawn(M, F, A) ->
gen_server:call({global, pool_master}, {spawn, group_leader(), M, F, A}).
-spec pspawn_link(Mod, Fun, Args) -> pid() when
Mod :: module(),
Fun :: atom(),
Args :: [term()].
pspawn_link(M, F, A) ->
P = pspawn(M, F, A),
link(P),
P.
start_nodes([], _, _) -> [];
start_nodes([Host|Tail], Name, Args) ->
case slave:start(Host, Name, Args) of
{error, {already_running, Node}} ->
io:format("Can't start node on host ~w due to ~w~n",[Host, {already_running, Node}]),
[Node | start_nodes(Tail, Name, Args)];
{error, R} ->
io:format("Can't start node on host ~w due to ~w~n",[Host, R]),
start_nodes(Tail, Name, Args);
{ok, Node} ->
[Node | start_nodes(Tail, Name, Args)]
end.
-spec stop() -> 'stopped'.
stop() ->
gen_server:call({global, pool_master}, stop).
get_elements(_Pos,[]) -> [];
get_elements(Pos,[E|T]) -> [element(Pos,E) | get_elements(Pos,T)].
stop_em([]) -> stopped;
stop_em([N|Tail]) ->
rpc:cast(N, erlang, halt, []),
stop_em(Tail).
init([]) ->
process_flag(trap_exit, true),
spawn_link(pool, statistic_collector, []),
{ok,[{0,node()}]}.
handle_call(get_nodes, _From, Nodes)->
{reply, Nodes, Nodes};
handle_call(get_node, _From, [{Load,N}|Tail]) ->
{reply, N, Tail++[{Load+1, N}]};
handle_call({attach, Node}, _From, Nodes) ->
case lists:keymember(Node, 2, Nodes) of
true ->
{reply, already_attached, Nodes};
false ->
erlang:monitor_node(Node, true),
spawn_link(Node, pool, statistic_collector, []),
{reply, attached, Nodes++[{999999,Node}]}
end;
handle_call({spawn, Gl, M, F, A}, _From, Nodes) ->
[{Load,N}|Tail] = Nodes,
Pid = spawn(N, pool, do_spawn, [Gl, M, F, A]),
{reply, Pid, Tail++[{Load+1, N}]};
handle_call(stop, _From, Nodes) ->
%% clean up in terminate/2
{stop, normal, stopped, Nodes}.
handle_cast(_, Nodes) ->
{noreply, Nodes}.
handle_info({Node,load,Load}, Nodes) ->
Nodes2 = insert_node({Load,Node}, Nodes),
{noreply, Nodes2};
handle_info({nodedown, Node}, Nodes) ->
{noreply, lists:keydelete(Node, 2, Nodes)};
handle_info(_, Nodes) -> %% The EXIT signals etc.etc
{noreply, Nodes}.
terminate(_Reason, Nodes) ->
N = lists:delete(node(), get_elements(2, Nodes)),
stop_em(N),
ok.
-spec do_spawn(pid(), module(), atom(), [term()]) -> term().
do_spawn(Gl, M, F, A) ->
group_leader(Gl, self()),
apply(M, F, A).
insert_node({Load,Node},[{L,Node}|Tail]) when Load > L ->
%% We have a raised load here
pure_insert({Load,Node},Tail);
insert_node({Load,Node},[{L,N}|Tail]) when Load =< L ->
%% Move forward in the list
T = lists:keydelete(Node,2,[{L,N}|Tail]),
[{Load,Node} | T];
insert_node(Ln,[H|T]) ->
[H | insert_node(Ln,T)];
insert_node(X,[]) -> % Can't happen
error_logger:error_msg("Pool_master: Bad node list X=~w\n", [X]),
exit(crash).
pure_insert({Load,Node},[]) ->
[{Load,Node}];
pure_insert({Load,Node},[{L,N}|Tail]) when Load < L ->
[{Load,Node}, {L,N} | Tail];
pure_insert(L,[H|T]) -> [H|pure_insert(L,T)].
%% Really should not measure the contributions from
%% the back ground processes here .... which we do :-(
%% We don't have to monitor the master, since we're slaves anyway
statistic_collector() ->
statistic_collector(5).
statistic_collector(0) -> exit(normal);
statistic_collector(I) ->
sleep(300),
case global:whereis_name(pool_master) of
undefined ->
statistic_collector(I-1);
M ->
stat_loop(M, 999999)
end.
%% Do not tell the master about our load if it has not changed
stat_loop(M, Old) ->
sleep(2000),
case statistics(run_queue) of
Old ->
stat_loop(M, Old);
NewLoad ->
M ! {node(), load, NewLoad}, %% async
stat_loop(M, NewLoad)
end.
sleep(I) -> receive after I -> ok end.
|
