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
|
library("tidyverse")
library("latex2exp")
## |--------------------------------+---------------------+-------------+------------------------------+--------------------------------------------+------------------------------------------------------------------|
## | Monitored Hardware | Alim | I2C Address | Notes | OS | sha256sum (.img file) |
## |--------------------------------+---------------------+-------------+------------------------------+--------------------------------------------+------------------------------------------------------------------|
## | 1Kohm resistor | some random 5V alim | 0x40 | | | |
## | Raspberry Pi 3 model B (2016) | 5.1V rpi alim | 0x41 | Powered through GPIO 2 and 6 | 2023-05-03-raspios-bullseye-arm64-lite.img | 9cd68ff450bc0fe0ac5b60e32723bc403ebca89d1214714e8e74361370d25204 |
## | Raspberry Pi 3 model B+ (2017) | 5.1V rpi alim | 0x42 | Powered through GPIO 2 and 6 | 2023-05-03-raspios-bullseye-arm64-lite.img | 9cd68ff450bc0fe0ac5b60e32723bc403ebca89d1214714e8e74361370d25204 |
## | Raspberry Pi 4 model B (2018) | 5.1V rpi alim | 0x43 | Powered through GPIO 2 and 6 | 2023-05-03-raspios-bullseye-arm64-lite.img | 9cd68ff450bc0fe0ac5b60e32723bc403ebca89d1214714e8e74361370d25204 |
## |--------------------------------+---------------------+-------------+------------------------------+--------------------------------------------+------------------------------------------------------------------|
pure_read=read_csv("results/pure_read.csv")
zmq=read_csv("results/zmq.csv")
GG_DELAY_READ=function(exp,for_inmem=0){
delays=exp%>%filter(inmem==for_inmem)%>%group_by(usen,deviceid)%>%mutate(tsnsec=(nsecs*10e-10)+timestamp)
delays=delays%>%mutate(delay=tsnsec-lag(tsnsec,1,default=NA))%>%filter(!is.na(delay))%>%mutate(time=timestamp-startat)
delays=delays%>%group_by(usen)%>%mutate(duration=max(time))
delays=delays%>%group_by(usen,deviceid,time)%>%summarize(delay_sd=sd(delay),delay_mean=mean(delay),duration=mean(duration),delay_max=max(delay),delay_min=min(delay))
##### Store in file
f=paste0("figures/delay_inmem",for_inmem,"_stddev.txt")
window=0
ina1<<-delays%>%filter(usen==1,time>window,time<duration-window)
ina2<<-delays%>%filter(usen==2,time>window,time<duration-window)
ina3<<-delays%>%filter(usen==3,time>window,time<duration-window)
ina4<<-delays%>%filter(usen==4,time>window,time<duration-window)
write(paste0("With 1 INA260 sd=",mean(ina1$delay_sd)," min=",mean(ina1$delay_min)," max=",mean(ina1$delay_max)),f)
write(paste0("With 2 INA260 sd=",mean(ina2$delay_sd)," min=",mean(ina2$delay_min)," max=",mean(ina2$delay_max)),f,append=TRUE)
write(paste0("With 3 INA260 sd=",mean(ina3$delay_sd)," min=",mean(ina3$delay_min)," max=",mean(ina3$delay_max)),f,append=TRUE)
write(paste0("With 4 INA260 sd=",mean(ina4$delay_sd)," min=",mean(ina4$delay_min)," max=",mean(ina4$delay_max)),f,append=TRUE)
##### Compute additional metrics
durations=delays%>%group_by(usen,deviceid)%>%summarise(duration=max(time))
med=delays%>%group_by(usen)%>%summarize(median=round(median(delay_mean),digits=9),duration=mean(duration))
#### End
label_at <- function(n) function(x) ifelse(x %% n == 0, x, ifelse (x==0,x,""))
ggplot(delays,aes(time,delay_mean,color=deviceid))+geom_point()+
facet_wrap(~usen,labeller = labeller(usen = c("1" = "With 1 ina260 in use",
"2" = "With 2 ina260 in use",
"3" = "With 3 ina260 in use",
"4" = "With 4 ina260 in use")))+
ylab(TeX(r'(Average delay per seconds $\bar{d}$)'))+
xlab("Time (s)")+
labs(color="Device id")+
scale_y_continuous(breaks = seq(0, 0.1,by=0.0001),labels=seq(0, 0.1,by=0.0001))+
geom_text(data=med,aes(x=max(med$duration)/2,y=median,label=median),inherit.aes=FALSE,nudge_y=0.00004)+
geom_hline(data=med,aes(yintercept=median),size=0.9,show.legend = FALSE,size=0.3)+
theme(legend.position="top",legend.margin=margin(b = -0.2, unit='cm'))
ggsave(paste0("figures/pure_read_inmem",for_inmem!=0,"_delay.pdf"),width=6,height=5.5)
}
GG_READ_PERF=function(exp,for_inmem=0){
##### Analyze
exp=exp%>%filter(inmem==for_inmem)%>%mutate(time=timestamp-startat)
data=exp%>%group_by(usen,deviceid)%>%group_by(usen,deviceid,timestamp,startat,endat)%>%summarize(nperdev=n())%>%mutate(time=timestamp-startat,duration=endat-startat)
##### Now complete missing measurements (say no value on at time 5 then a row with a zero at that time must be added)
data=data%>%group_by(usen,deviceid)%>%complete(timestamp=seq(min(startat),min(endat),by=1),fill=list(nperdev=0))%>%mutate(startat=min(startat,na.rm=TRUE),endat=max(startat,na.rm=TRUE),time=timestamp-startat,duration=max(duration,na.rm=TRUE))
##### Compute additional metrics
durations=exp%>%group_by(usen,deviceid)%>%summarise(duration=max(endat)-max(startat))
med=data%>%group_by(usen)%>%summarize(median=round(median(nperdev)),duration=mean(duration))
#### End
# Plot
msg=paste0("(inmem=",for_inmem,")")
nread=max(exp$nread)
label_at <- function(n) function(x) ifelse(x %% n == 0, x, ifelse (x==0,x,""))
ggplot(data,aes(x=time,y=nperdev,color=deviceid)) + geom_point(size=0.3) +geom_line() + xlab("Time (s)") + ylab(TeX(r'(Sample read frequency $f_r$)')) +
facet_wrap(~usen,labeller = labeller(usen = c("1" = "With 1 ina260 in use",
"2" = "With 2 ina260 in use",
"3" = "With 3 ina260 in use",
"4" = "With 4 ina260 in use")))+
scale_x_continuous(breaks = seq(0,3600,by=60),labels=label_at(60))+
scale_y_continuous(breaks = seq(0, max(data$nperdev), by = 1000),expand = expansion(mult = 0.1))+
#ggtitle(paste("Stop until",nread,"measurements are done", msg))+
labs(color="Device id")+
geom_vline(data=durations,aes(xintercept=duration,color=deviceid),linetype="dashed",show.legend = FALSE)+
geom_hline(data=med,aes(yintercept=median),show.legend = FALSE,size=0.3)+
geom_text(data=med,aes(x=duration/2,y=median,label=median),inherit.aes=FALSE,vjust=-0.55)+
theme(legend.position="top",legend.margin=margin(b = -0.2, unit='cm'))
ggsave(paste0("figures/pure_read_inmem",for_inmem!=0,".pdf"),width=6,height=5.5)
}
TABLE_DELAY=function(exp,for_inmem=0){
delays=exp%>%filter(inmem==for_inmem)%>%group_by(usen,deviceid)%>%mutate(tsnsec=(nsecs*10e-10)+timestamp)
delays=delays%>%group_by(usen,deviceid)%>%mutate(delay=tsnsec-lag(tsnsec,1,default=NA))%>%filter(!is.na(delay))%>%mutate(time=timestamp-startat)
f=paste0("figures/fp_inmem",for_inmem,".txt")
write("", f)
delays%>%group_by(usen)%>%group_walk(function(data,grp){
cur_usen=grp$usen
cdf=ecdf(data$delay) # Get CDF
#### For fd
fdefault=454.54
percent=1-cdf(1/fdefault)
write(paste0("inmem=",for_inmem," usen=",cur_usen,": f=",fdefault,"Hz percent=",round(percent*100,digits=2),"%", " missings=",percent*length(data$delay)),f, append=TRUE)
#### For fmax
fmax=3571.429
percent=1-cdf(1/fmax)
write(paste0("inmem=",for_inmem," usen=",cur_usen,": f=",fmax,"Hz percent=",round(percent*100,digits=2),"%", " missings=",percent*length(data$delay),"\n----"),f, append=TRUE)
# plot(cdf,xlim=c(0,0.001))
})
}
GG_READ_POWER=function(exp,for_inmem=0){
# Analyze
exp=exp%>%filter(inmem==for_inmem)%>%mutate(time=timestamp-startat+nsecs*1e-9)
nread=max(exp$nread)
msg=paste0("(inmem=",for_inmem,")")
data=exp%>%group_by(usen,deviceid)%>%mutate(duration=endat-startat)
# Plot
label_at <- function(n) function(x) ifelse(x %% n == 0, x, ifelse (x==0,x,""))
ggplot(data,aes(x=time,y=power,color=deviceid)) + geom_point(size=0.3) +geom_line() + xlab("Time (s)") + ylab("Power consumption (W)") +
facet_wrap(~usen,labeller = labeller(usen = c("1" = "With 1 ina260 in use",
"2" = "With 2 ina260 in use",
"3" = "With 3 ina260 in use",
"4" = "With 4 ina260 in use")))+
scale_x_continuous(breaks = seq(0,3600,by=60),labels=label_at(60)) + ggtitle(paste("Stop until",nread,"measurements are done", msg))
ggsave(paste0("figures/pure_read_inmem",for_inmem!=0,"_power.pdf"),width=10)
}
TABLE_READ_POWER=function(exp){
power_0x40<<-median((exp%>%filter(addr=="0x40"))$power)
power_0x41<<-median((exp%>%filter(addr=="0x41"))$power)
power_0x42<<-median((exp%>%filter(addr=="0x42"))$power)
power_0x43<<-median((exp%>%filter(addr=="0x43"))$power)
power_0x40_std<<-round(sd((exp%>%filter(addr=="0x40"))$power),digits=2)
power_0x41_std<<-round(sd((exp%>%filter(addr=="0x41"))$power),digits=2)
power_0x42_std<<-round(sd((exp%>%filter(addr=="0x42"))$power),digits=2)
power_0x43_std<<-round(sd((exp%>%filter(addr=="0x43"))$power),digits=2)
f="figures/power_median.txt"
write(paste0("1Kohm resistor 0x40: ",power_0x40,"W(",power_0x40_std,")"), f)
write(paste0("Raspberri Pi 3 model B (2016) 0x41: ",power_0x41,"W(",power_0x41_std,")"), f, append=TRUE)
write(paste0("Raspberri Pi 3 model B+ (2017) 0x42: ",power_0x42,"W(",power_0x42_std,")"), f, append=TRUE)
write(paste0("Raspberry Pi 4 model B (2018) 0x43: ",power_0x43,"W(",power_0x43_std,")"), f, append=TRUE)
}
GG_ZMQ_PERF=function(exp){
##### Analyze
data=exp%>%group_by(usen,deviceid,timestamp)%>%summarize(nperdev=n(),duration=max(duration))%>%mutate(time=timestamp-min(timestamp))
data=data%>%group_by(usen,deviceid)%>%mutate(startat=min(timestamp),endat=max(timestamp))
jean <<- data
##### Now complete missing measurements (say no value on at time 5 then a row with a zero at that time must be added)
data=data%>%group_by(usen,deviceid)%>%complete(timestamp=seq(min(startat),min(startat)+max(duration),by=1),fill=list(nperdev=0))%>%mutate(startat=min(startat,na.rm=TRUE),endat=max(startat,na.rm=TRUE),time=timestamp-startat)
#### Now compute additional metrics
med=data%>%group_by(usen)%>%summarize(median=round(median(nperdev)),duration_real=max(timestamp)-min(timestamp))
#### Plot
label_at <- function(n) function(x) ifelse(x %% n == 0, x, ifelse (x==0,x,""))
ggplot(data,aes(x=time,y=nperdev,color=deviceid))+geom_point()+geom_line()+
facet_wrap(~usen,labeller = labeller(usen = c("1" = "With 1 ina260 in use",
"2" = "With 2 ina260 in use",
"3" = "With 3 ina260 in use",
"4" = "With 4 ina260 in use")))+
scale_x_continuous(breaks = seq(0,3600,by=60),labels=label_at(60))+
scale_y_continuous(breaks = seq(0, max(data$nperdev), by = 1000),expand = expansion(mult = 0.1))+
geom_hline(data=med,aes(yintercept=median),show.legend = FALSE,size=0.3)+
geom_text(data=med,aes(x=duration_real/2,y=median,label=median),inherit.aes=FALSE,vjust=-0.55)+
#ggtitle(paste("Running zmq experiment for",unique(exp$duration),"s"))+
labs(color="Device id")+
xlab("Time (s)") + ylab(TeX(r'(Sample read frequency $f_r$)'))+
theme(legend.position="top",legend.margin=margin(b = -0.2, unit='cm'))
ggsave("figures/zmq.pdf")
}
TABLE_DELAY(pure_read,0)
TABLE_DELAY(pure_read,1)
GG_DELAY_READ(pure_read,0)
GG_DELAY_READ(pure_read,1)
GG_READ_POWER(pure_read,0)
GG_READ_POWER(pure_read,1)
GG_READ_PERF(pure_read,0)
GG_READ_PERF(pure_read,1)
GG_ZMQ_PERF(zmq)
TABLE_READ_POWER(pure_read)
|
