forked from scala-blitz/scala-blitz
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtasks.TODO
145 lines (132 loc) · 3.92 KB
/
tasks.TODO
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
Reducable interface:
☐ go through all operations in standard collections and filter them for the Reducable interface
☐ foreach
✔ fold @done (13-11-27 16:40)
✔ reduce @done (13-11-27 16:40)
✔ aggregate @done (13-11-27 16:40)
✔ mapReduce @done (13-11-27 16:40)
✔ max @done (13-11-27 16:40)
✔ min @done (13-11-27 16:40)
✔ sum @done (13-11-27 16:40)
✔ product @done (13-11-27 16:40)
✔ map @done (13-11-27 16:40)
✔ filter @done (13-11-27 16:40)
✔ flatMap @done (13-11-27 16:40)
☐ contains\exists\containsSlice
☐ distinct
☐ forall
✔ groupBy @done (13-11-27 16:41)
☐ intersect
☐ sort
☐ to
☐ have a variant of breakOut for parallel collections to allow mapping ranges to arbitrary collections
☐ parallel views
Zippable interface:
☐ copyToArray
☐ zip
☐ zipWithIndex
☐ zipView
☐ corresponds
Parallel ranges:
✔ fold @done (13-05-27 14:24)
✔ reduce @done (13-05-27 14:24)
✔ aggregate @done (13-05-27 14:24)
✔ max @done (13-05-27 14:24)
✔ min @done (13-05-27 14:24)
✔ sum @done (13-05-27 14:24)
✔ mapReduce @done (13-06-21 16:36)
☐ foreach
✔ product @done (13-05-27 14:25)
☐ copyToArray
☐ map
✔ filter @done (13-11-27 16:41)
✔ flatMap @done (13-11-27 16:41)
☐ every operation in the Reducable interface
☐ unit tests
☐ microbenchmarks
☐ to
Reducable:
☐ operations
☐ unit tests
☐ microbenchmarks
Zippable:
☐ operations
☐ unit tests
☐ microbenchmarks
Parallel concs:
✔ reduce @done (13-05-27 14:26)
☐ fold
☐ aggregate
☐ mapReduce
☐ map
☐ filter
☐ flatMap
✔ copyToArray @done (13-05-27 14:26)
Parallel arrays:
✔ reduce @done (13-05-31 10:44)
✔ fold @done (13-05-31 10:44)
✔ aggregate @done (13-05-31 10:44)
✔ sum @done (13-05-31 10:44)
✔ product @done (13-05-31 10:44)
☐ mapReduce
☐ foreach
☐ min
☐ max
✔ map @done (13-05-31 10:44)
✔ filter @done (13-05-31 10:44)
✔ flatMap @done (13-06-07 17:50)
☐ zipView
Parallel hash tables:
☐ reduce
✔ fold @done (13-06-21 16:37)
✔ aggregate @done (13-06-21 16:37)
☐ mapReduce
☐ foreach
☐ map
✔ filter @done (13-06-21 16:37)
☐ unit tests
☐ microbenchmarks
Parallel binary tree operations:
☐ reduce
☐ fold
☐ aggregate
☐ mapReduce
☐ foreach
☐ map
☐ filter
☐ unit tests
☐ microbenchmarks
Splitting package:
☐ same interface for old parallel collections (implicit conversions in the same style as the workstealing package)
To fix:
☐ argument expressions need to be assigned to a local var
☐ parallel array operations need to be specialized
☐ specialize parallel operation instances for the other data-structures where it makes sense
☐ implement all the basic parallel operations
✔ add combiner support @done (13-03-01 17:02)
☐ fix exception handling
✔ replace volatile reads with unsafe calls @done (13-04-26 13:49)
Todo items:
✔ write a comprehensive suite of unit tests @done (13-11-27 16:43)
✔ write a comprehensive suite of microbenchmark regression tests @done (13-11-27 16:43)
✔ implement HashNode and parallel hash tables @done (13-11-27 16:43)
☐ an interface for thread pools and their configuration
✔ value/implicit class conversion to support parallel operations - an API to convert between normal and parallel operations @done (13-11-27 16:44)
☐ verify performance on AMD platforms, UltraSPARC, multi-socket Intels
☐ design an API for scheduler configuration (number of threads, pool, chunking size, schedule, stealing strategy, etc.)
Bigger applications:
☐ PageRank
☐ K-means
✔ Barnes-Hut @done (13-11-27 16:44)
☐ Smith-Waterman
✔ Mandelbrot set computation @done (13-11-27 16:44)
☐ genetic algorithm optimization
☐ flocking
☐ some linear algebra applications
☐ raytracing
☐ adjoint convolution
☐ finding anagrams
☐ brute-force TSP
☐ triangular matrix multiplication
Features to add:
☐ allow asynchronous execution