-
Notifications
You must be signed in to change notification settings - Fork 66
/
gradeonly.r
139 lines (129 loc) · 5.81 KB
/
gradeonly.r
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
grades <- function(base, scale, grade,fct) {
gscale <- matrix(c(.05,"A+", .1,"A", .2,"A-", .3,"B", .4,"B-", .6,"C", .9,"C-" ), nrow=2)
gscale <- t(gscale)
# =========================== lat grades ======================
base$grade95[ base$p95_00 < as.numeric(grade[1]) ] <- grade[1,2]
base$grade99[ base$p99_00 < as.numeric(grade[1]) ] <- grade[1,2]
base$gradeavg[ base$lat < as.numeric(grade[1]) ] <- grade[1,2]
for ( i in 2:nrow(grade) ) {
base$grade95[ base$p95_00 < as.numeric(grade[i,1]) &
base$p95_00 >= as.numeric(grade[i-1,1]) ] <- grade[i,2]
base$grade99[ base$p99_00 < as.numeric(grade[i,1]) &
base$p99_00 >= as.numeric(grade[i-1,1]) ] <- grade[i,2]
base$gradeavg[ base$lat < as.numeric(grade[i,1]) &
base$lat >= as.numeric(grade[i-1,1]) ] <- grade[i,2]
}
base$grade95[ base$p95_00 > as.numeric(grade[nrow(grade),1]) ] <- "D"
base$grade99[ base$p99_00 > as.numeric(grade[nrow(grade),1]) ] <- "D"
base$gradeavg[ base$lat > as.numeric(grade[nrow(grade),1]) ] <- "D"
# =========================== scale grades ======================
base$scalelat <- ((scale['lat'] /base['lat'] )-1)/fct
base$scale95 <- ((scale['p95_00']/base['p95_00'])-1)/fct
base$scale99 <- ((scale['p99_00']/base['p99_00'])-1)/fct
base$gscale95[ base$scale95 <= as.numeric(gscale[1]) ] <- gscale[1,2]
base$gscale99[ base$scale99 <= as.numeric(gscale[1]) ] <- gscale[1,2]
base$gscaleavg[ base$scalelat <= as.numeric(gscale[1]) ] <- gscale[1,2]
for ( i in 2:nrow(gscale) ) {
base$gscale95[ base$scale95 <= as.numeric(gscale[i,1]) &
base$scale95 > as.numeric(gscale[i-1,1]) ] <- gscale[i,2]
base$gscale99[ base$scale99 <= as.numeric(gscale[i,1]) &
base$scale99 > as.numeric(gscale[i-1,1]) ] <- gscale[i,2]
base$gscaleavg[ base$scalelat <= as.numeric(gscale[i,1]) &
base$scalelat > as.numeric(gscale[i-1,1]) ] <- gscale[i,2]
}
base$gscale95[ base$scale95 > as.numeric(gscale[nrow(gscale),1]) ] <- "D"
base$gscale99[ base$scale99 > as.numeric(gscale[nrow(gscale),1]) ] <- "D"
base$gscaleavg[ base$scalelat > as.numeric(gscale[nrow(gscale),1]) ] <- "D"
print( cbind(base['gradeavg'],
base['grade95'],
base['grade99'],
base['lat'],
base['p95_00'],
base['p99_00'],
base$scalelat ,
base$scale95 ,
base$scale99 ,
base$gscaleavg ,
base$gscale95 ,
base$gscale99
))
for ( i in 1:nrow(base) ) {
cat(sprintf(" %-2s/%-2s %4.1f/%2.1f %-2s/%-2s %4.1f/%2.1f %-2s/%-2s %4.1f/%2.1f \n"
, base[i,'gradeavg'],base[i,'gscaleavg'], base[i,'lat'] ,as.numeric(base[i,'scalelat'])
, base[i,'grade95'] ,base[i,'gscale95'] , base[i,'p95_00'],as.numeric(base[i,'scale95'])
, base[i,'grade99'] ,base[i,'gscale99'] , base[i,'p99_00'],as.numeric(base[i,'scale99'])))
}
return(base)
}
grade_only <- function(m) {
#
# setup GRADING SCALES
#
colnames <- c("lat", "grade" )
gscale <- matrix(c(.05,"A+", .1,"A", .2,"A-", .3,"B", .4,"B-", .6,"C", .9,"C-" ), nrow=2)
gscale <- t(gscale)
colnames(gscale)=colnames
grr <- matrix(c(2,"A+", 4,"A", 6,"A-", 8,"B", 10,"B-", 12,"C", 14,"C-"), nrow=2)
grr <- t(grr)
colnames(grr)=colnames
gsr <- matrix(c( 12 , "A+" , 14 , "A", 16 , "A-", 18 , "B", 20 , "B-", 22 , "C", 24 , "C-" ), nrow=2)
gsr <- t(gsr)
colnames(gsr)=colnames
gw1 <- matrix(c( 0.2,"A+" , 0.5,"A", 1,"A-", 1.5,"B", 2,"B-", 2.5,"C", 3,"C-" ), nrow=2)
gw1 <- t(gw1)
colnames(gw1)=colnames
gw128 <- matrix(c( 2 , "A+" , 4 , "A", 6 , "A-", 8 , "B", 10 , "B-", 12 , "C", 14 , "C-" ), nrow=2)
gw128 <- t(gw128)
colnames(gw128)=colnames
#gm <- cbind(gw128,gw1,gsr,grr)
#gm <- rbind(gw128,gw1,gsr,grr)
# get all the latencies
gm <- cbind(gw128[,1],gw1[,1],gsr[,1],grr[,1])
#
# EXTRACT data sets: RANDOM READ, SEQUENTIAL READ, WRITE 1K, WRITE 128K
#
rr8 <- subset(m,m['name'] == "randread" & m['users'] == 8 )
rr16 <- subset(m,m['name'] == "randread" & m['users'] == 16 )
rr32 <- subset(m,m['name'] == "randread" & m['users'] == 32 )
#
sr1 <- subset(m,m['name'] == "read" & m['users'] == 1 & m['bs'] == "1M" )
sr8 <- subset(m,m['name'] == "read" & m['users'] == 8 & m['bs'] == "1M" )
sr16 <- subset(m,m['name'] == "read" & m['users'] == 16 & m['bs'] == "1M" )
#
w1k_1 <- subset(m,m['name'] == "write" & m['users'] == 1 & m['bs'] == "1K" )
w1k_4 <- subset(m,m['name'] == "write" & m['users'] == 4 & m['bs'] == "1K" )
w1k_16 <- subset(m,m['name'] == "write" & m['users'] == 16 & m['bs'] == "1K" )
#
w128k_1 <- subset(m,m['name'] == "write" & m['users'] == 4 & m['bs'] == "128K" )
w128k_4 <- subset(m,m['name'] == "write" & m['users'] == 4 & m['bs'] == "128K" )
w128k_16 <- subset(m,m['name'] == "write" & m['users'] == 16 & m['bs'] == "128K" )
#
# CALCULATE GRADES based on latency data
#
rr <- grades(rr16,rr32,grr,2)
rr$system = "randread 16 users"
sr <- grades(sr1,sr8,gsr,8)
sr$system = "seq read 1 users"
w1 <-grades(w1k_4,w1k_16,gw1,4)
w1$system = "write 1k 4 users"
w128 <- grades(w128k_4,w128k_16,gw128,4)
w128$system = "write 128k 4 users"
#
# CREATE dataset with GRADES and DATA from
# from RANDOM READ, SEQUENTIAL READ, WRITE 1K, WRITE 128K
#
mold <- m
m <- rbind(w128,w1,sr,rr)
#
# EXTRACT HISTOGRAMS
#
hist <- cbind(m['us50'],m['us100'], m['us250'],m['us500'],m['ms1'],
m['ms2'],m['ms4'],m['ms10'],m['ms20'],m['ms50'],
m['ms100'],m['ms250'],m['ms500'],m['s1'],m['s2'],m['s5'])
thist <- t(hist)
print(m)
#cat(m,"\n");
#cat(thist,"\n");
}
m <- read.csv("data_colorado.csv")
grade_only(m)