---
title: "R Notebook"
output:
  html_document:
    df_print: paged
---

```{r include=FALSE}
library(tidyverse)
```

Using an additional grid with bigger cells for the spatial joins...

[Grid B1](Figures/Grid2/B1.png)
[Grid B2](Figures/Grid2/B2.png)
[Grid B3](Figures/Grid2/B3.png)

```{r echo=FALSE}
fields = c("Timestamp", "Tag", "appId", "Cores", "Executors", "Epsilon", "Mu", "Delta", "Time", "Load")
data_path = "~/Documents/PhD/Research/Scripts/R/R17/nohubGrid2.txt"
data = enframe(read_lines(data_path), name = "n", value = "line") %>% select(line) %>%
  filter(grepl("PFLOCK", line)) %>% separate(line, into = fields, sep = "\\|") %>%
  mutate(Time = as.numeric(Time)) %>%
  group_by(Executors, Epsilon) %>% summarise(Time = mean(Time)) %>%
  arrange(Time)

head(data)
```

```{r echo=FALSE}
data$Executors = factor(data$Executors, levels = c("5","10","15"))
data$Epsilon   = factor(data$Epsilon,   levels = c("100.0","110.0"))
p = ggplot(data = data, aes(x = Epsilon, y = Time, fill = Executors)) +
  geom_bar(stat="identity", position=position_dodge(width = 0.75), width = 0.7) + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  labs(x="Epsilon(m)", y="Time(s)", title="Execution time after varying Epsilon parameter.") 
plot(p)
```

* Dataset: Berlin (x1) with 5 executors, Berlin (x2) with 10 executors and Berlin (x3) with 15 executors.
* 4 cores per executor.  

```{r echo=FALSE}
data = enframe(read_lines(data_path), name = "n", value = "line") %>% select(line) %>%
  filter(grepl("PFLOCK", line)) %>% separate(line, into = fields, sep = "\\|") %>%
  select(appId, Epsilon)

fields2 = c("Timestamp", "Tag", "appId", "Executors", "Cores", "Status", "Duration", "Stage", "Time", "Load", "Interval")
data1 = enframe(read_lines(data_path), name = "n", value = "line") %>% select(line) %>% filter(grepl("\\|FF\\|", line)) %>%
  filter(grepl("END", line)) %>% separate(line, into = fields2, sep = "\\|") %>%
  mutate(Time = as.numeric(Time))
data2 = data %>% inner_join(data1, by = "appId") %>% select(Epsilon, Executors, Stage, Time, Interval) %>%
  filter(grepl("[1-6]\\.", Stage)) %>% filter(Epsilon == "110.0") %>% mutate(Stage = paste0(Interval,"-",str_trim(Stage))) %>%
  group_by(Executors, Stage) %>% summarise(Time = mean(Time))

data2$Executors = factor(data2$Executors, levels = c("5","10","15"))
p = ggplot(data = data2, aes(x = Stage, y = Time, fill = Executors)) +
  geom_bar(stat="identity", position=position_dodge(width = 0.75), width = 0.7) + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  labs(x="Stages", y="Time(s)", title="Execution time during the Flock Finding.") 
plot(p)

```


```{r echo=FALSE}
data = enframe(read_lines(data_path), name = "n", value = "line") %>% select(line) %>%
  filter(grepl("PFLOCK", line)) %>% separate(line, into = fields, sep = "\\|") %>%
  select(appId, Epsilon)

fields3 = c("Timestamp", "Tag", "appId", "Executors", "Cores", "Status", "Duration", "Stage", "Time", "Load", "Interval")
data1 = enframe(read_lines(data_path), name = "n", value = "line") %>% select(line) %>% filter(grepl("\\|MF\\|", line)) %>%
  filter(grepl("END", line)) %>% separate(line, into = fields3, sep = "\\|") %>%
  mutate(Time = as.numeric(Time))
data2 = data %>% inner_join(data1, by = "appId") %>% select(Epsilon, Executors, Stage, Time, Interval) %>%
  filter(grepl("[A-F]\\.", Stage)) %>% filter(Epsilon == "110.0") %>% mutate(Stage = paste0(Interval,"-",str_trim(Stage))) %>%
  group_by(Executors, Stage) %>% summarise(Time = mean(Time))

data2$Executors = factor(data2$Executors, levels = c("5","10","15"))
p = ggplot(data = data2, aes(x = Stage, y = Time, fill = Executors)) +
  geom_bar(stat="identity", position=position_dodge(width = 0.75), width = 0.7) + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  labs(x="Stages", y="Time(s)", title="Execution time during the Maximal Disks Finding.") 
plot(p)

```