#!/usr/bin/env Rscript

library(ggplot2)
library(dplyr)
require(scales)

args = commandArgs(trailingOnly = TRUE)

file = "job_similarities_5024292.csv"
file = "job_similarities_7488914.csv"

file = args[1]

data = read.csv(file)
# Columns are: jobid alg_id alg_name similarity

data$alg_id = as.factor(data$alg_id)
print(nrow(data))

# FILTER, TODO
data = data %>% filter(similarity <= 1.0)

# empirical cummulative density function (ECDF)
ggplot(data, aes(similarity, color=alg_name, group=alg_name)) + stat_ecdf(geom = "step") + xlab("SIM") + ylab("Fraction of jobs") + theme(legend.position="bottom") + scale_color_brewer(palette = "Set2")
ggsave("ecdf.png")

e = data %>% filter(similarity >= 0.5)
ggplot(e, aes(similarity, color=alg_name, group=alg_name)) + stat_ecdf(geom = "step") + xlab("SIM") + ylab("Fraction of jobs") + theme(legend.position="bottom")  + scale_color_brewer(palette = "Set2")
print(summary(e))
ggsave("ecdf-0.5.png")

# histogram for the jobs
ggplot(data, aes(similarity), group=alg_name) + geom_histogram(color="black", binwidth=0.025) + aes(fill = alg_name) + facet_grid(alg_name ~ ., switch = 'y') + scale_y_continuous(limits=c(0, 100), oob=squish)  +   scale_color_brewer(palette = "Set2") + ylab("Count (cropped at 100)")
ggsave("hist-sim.png")

# +  scale_y_continuous(trans = log2_trans(), breaks = trans_breaks("log2", function(x) 2^x), labels = trans_format("log2", math_format(2^.x)))

#+ ylim(0, 250)  + stat_summary(aes(linetype = alg_id), fun.y=mean, geom="line")   +

exit(0)



########### merged
both = rbind(i[ , (names(i) %in% names(d))], d[ , (names(d) %in% names(i))])
both$tpGiB = both$tpMiBs / 1024
both$PPN = as.factor(both$PPN)

e = both %>% filter(nodes == 500)
ggplot(e, aes(PPN, tpGiB, color=config, group=config)) + geom_boxplot() + facet_grid(op + dim ~ ., switch = 'y') + xlab("PPN") + ylab("Performance in GiB/s") + stat_summary(aes(linetype = config), fun.y=mean, geom="line")   + theme(legend.position="bottom") + ylim(0, 250)
ggsave("500-nodes.png")

ggplot(e, aes(PPN, tpGiB, color=config, group=config)) + geom_boxplot(position=pd) + facet_grid(op + dim ~ ., switch = 'y') + xlab("PPN") + ylab("Performance in GiB/s") + stat_summary(aes(linetype = config), fun.y=mean, geom="line")   + theme(legend.position="bottom")
ggsave("500-nodes-all.png")


e = both %>% filter(nodes == 200)
ggplot(e, aes(PPN, tpGiB, color=config, group=config)) + geom_point(position=pd) + facet_grid(op + dim ~ ., switch = 'y') + xlab("PPN") + ylab("Performance in GiB/s") + stat_summary(aes(linetype = config), fun.y=mean, geom="line")   + theme(legend.position="bottom") + ylim(0, 250)
ggsave("200-nodes.png")

ggplot(e, aes(PPN, tpGiB, color=config, group=config)) + geom_point(position=pd) + facet_grid(op + dim ~ ., switch = 'y') + xlab("PPN") + ylab("Performance in GiB/s") + stat_summary(aes(linetype = config), fun.y=mean, geom="line")   + theme(legend.position="bottom")
ggsave("200-nodes-all.png")


e = both %>% filter(nodes == 100)
ggplot(e, aes(PPN, tpGiB, color=config, group=config)) + geom_point(position=pd) + facet_grid(op + dim ~ ., switch = 'y') + xlab("PPN") + ylab("Performance in GiB/s") + stat_summary(aes(linetype = config), fun.y=mean, geom="line")   + theme(legend.position="bottom") + ylim(0, 250)
ggsave("100-nodes.png")

ggplot(e, aes(PPN, tpGiB, color=config, group=config)) + geom_point(position=pd) + facet_grid(op + dim ~ ., switch = 'y') + xlab("PPN") + ylab("Performance in GiB/s") + stat_summary(aes(linetype = config), fun.y=mean, geom="line")   + theme(legend.position="bottom")
ggsave("100-nodes-all.png")


########### End merged plots

d$tpGiB = d$tpMiBs / 1024
d$nodes = as.factor(d$nodes)
d$PPN = as.factor(d$PPN)
d$volume = as.factor(round(d$sizeMiB/1024,1))

# Compare:

# scale_colour_gradientn(colours=rainbow(3))
# d %>% filter(tpGiB > 100)

e = d %>% filter(nodes==5) # , size>100
ggplot(e, aes(volume, tpGiB, color=PPN, group=PPN)) + geom_point(position=pd) + facet_grid(op + config ~ ., switch = 'y') + xlab("Volume in GiB") + ylab("Performance in GiB/s") + stat_summary(aes(), fun.y=mean, geom="line")   + theme(legend.position="bottom")
ggsave("5nodes-write-size-compare.png", width=4, height=3)




ggplot(e, aes(volume, tpGiB, color=PPN, group=PPN)) + geom_point(position=pd) + facet_grid(type ~ ., switch = 'y') + xlab("Volume in GiB") + ylab("Performance in GiB/s") + stat_summary(aes(), fun.y=mean, geom="line")   + theme(legend.position="bottom")
ggsave("5nodes-read-size-compare.png", width=4, height=3)


e = d %>% filter(size==60000)

ggplot(e, aes(nodes, tpGiB, color=PPN, group=PPN)) + geom_point(position=pd) + facet_grid(type ~ ., switch = 'y') + xlab("Nodes") + ylab("Performance in GiB/s") + stat_summary(aes(), fun.y=mean, geom="line")  + theme(legend.position="bottom")
ggsave("write-60000.png", width=4, height=3)

ggplot(e, aes(nodes, tpGiB, color=PPN, group=PPN)) + geom_point(position=pd) + facet_grid(type ~ ., switch = 'y') + xlab("Nodes") + ylab("Performance in GiB/s") + stat_summary(aes(), fun.y=mean, geom="line")   + theme(legend.position="bottom")
ggsave("read-60000.png", width=4, height=3)

e = d %>% filter(size==60000, PPN==12)
ggplot(e, aes(nodes, write, color=type, group=type)) + geom_point(position=pd) + xlab("Nodes") + ylab("Performance in GiB/s") + theme(legend.position="bottom") + stat_summary(aes(group=type), fun.y=mean, geom="line")
ggsave("compare-write-12.png", width=4, height=3)

e = d %>% filter(size==60000, PPN==4)
ggplot(e, aes(nodes, write, color=type, group=type)) + geom_point(position=pd) + xlab("Nodes") + ylab("Performance in GiB/s") + theme(legend.position="bottom") + stat_summary(aes(group=type), fun.y=mean, geom="line")
ggsave("compare-write-4.png", width=4, height=3)