-
Notifications
You must be signed in to change notification settings - Fork 15
/
bench.pl
120 lines (98 loc) · 2.58 KB
/
bench.pl
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
:- module(bench,
[ bench/0,
bench_first/1,
create/1, % +Triples
remove_p/1, % +Fraction
stress/2 % SizeM, Threads
]).
:- include(local_test).
:- use_module(library(semweb/rdf_db)).
:- use_module(library(apply_macros)).
:- use_module(library(statistics)).
:- use_module(library(random)).
:- if(\+current_predicate(rdf_resource/1)).
rdf_resource(S) :-
rdf_subject(S).
rdf_resource(S) :-
rdf_current_predicate(S).
:- endif.
:- initialization set_random(seed(111)).
bench :-
time(create(100000)),
bench_first(100).
%! bench_first
%
% Query the first triple of all resources
bench_first(N) :-
findall(Q, rdf_resource(Q), Resources),
length(Resources, Len),
format('~D resources~n', [Len]),
time(forall(between(1, N, _),
forall(member(Q,Resources),
first_triple(Q)))).
first_triple(Q) :-
rdf(Q,_,_),
!.
first_triple(_).
%! create(N)
%
% Create an RDF DB with N triples
create(N) :-
flag(offset, Off, Off+N),
create(N,Off).
create(N,Off) :-
forall(between(1, N, I),
add_random(N, I, Off)).
add_random(N, I, Off) :-
random_triple(N,I,Off, S,P,O),
rdf_assert(S, P, O).
random_triple(N,I,Off, S,P,O) :-
random_subject(N, I, Off, S),
random_predicate(N, I, Off, P),
random_object(N, I, Off, O).
random_subject(N,_I,Off,S) :-
I is random(N//10)+Off,
atom_concat(s,I,S).
random_predicate(_N,_I,_Off,S) :-
I is random(10),
atom_concat(p,I,S).
random_object(N,I,Off,O) :-
( maybe(0.3),fail
-> R is random(N//30),
atom_concat(l,R,L),
O = literal(L)
; random_subject(N,I,Off,O)
).
%! remove_p(+P)
%
% Remove a fraction of the DB. E.g., remove_p(0.1) removes
% 10% of the DB.
remove_p(Prob) :-
( rdf(S,P,O),
maybe(Prob),
rdf_retractall(S,P,O),
fail
; true
).
%! stress(SizeM, FillerThreads)
%
% Start threads that fill, remove and GC the DB.
stress(SizeM, Fillers) :-
MaxCount = round(SizeM*1000000),
forall(between(1, Fillers, FI),
( atom_concat(filler_, FI, Alias),
thread_create(filler(MaxCount), _, [alias(Alias)])
)),
thread_create(emptier, _, [alias(emptier)]).
filler(MaxCount) :-
rdf_statistics(triples(Count)),
( Count > MaxCount
-> sleep(1)
; Bunch = 100000,
Offset is random(MaxCount-Bunch),
time(create(Bunch,Offset))
),
filler(MaxCount).
emptier :-
remove_p(0.1),
emptier.