Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
350 changed files
with
5,506,844 additions
and
1,356 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,22 @@ | ||
1. Make the binder coverings: | ||
- Code listings/Data | ||
- Research Papers | ||
- Work Log | ||
- Graphs and Charts | ||
2. Code Listings | ||
- Figure out how many lines fit on one black and white page | ||
- Decide what parts to print | ||
|
||
3. Work Log | ||
- Write a work log based on the emails with Dusseau | ||
|
||
4. Graphs and Charts | ||
- Add legend to the graphs | ||
- print one out, see how it looks | ||
------------------------------------------------------------------- | ||
Things to print @ OfficeMax | ||
1. Graphs (did that) | ||
2. Graphs on the board (did that) | ||
3. Charts on the board (did that) | ||
4. Binder coverings | ||
5. |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1 @@ | ||
dhaivat@dhaivat-GA-870A-UD3.15841:1392174308 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,6 @@ | ||
dat = read.csv("analysis/latencies.csv", head=TRUE, sep=",") | ||
dat | ||
names(dat) | ||
dat$X.cached.latency.times.in.nanoseconds. | ||
boxplot(data) | ||
boxplot(dat) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,2 @@ | ||
github.com/ | ||
github.com/* |
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,11 @@ | ||
#ROADMAP | ||
|
||
## Issues to solve | ||
1. Implement another system that sits on the namenode and listens around | ||
for calls that change things on the filesystem. Should be independent of the | ||
caching system entirely, i.e. a separate executable. | ||
2. Make the data caching work on multinode (might have to make a Writeable implementation) | ||
3. **Make PUT calls work** | ||
|
||
## New ideas | ||
1. Have to test the system on a popular usecase and see what happens |
Empty file.
Large diffs are not rendered by default.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1 @@ | ||
dhaivat@dhaivat-GA-870A-UD3.15841:1392174308 |
Binary file not shown.
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,264 @@ | ||
source("box-plot.r") | ||
source("box-plot.r") | ||
source("box-plot.r") | ||
dat <- read.csv("latencies.csv") | ||
dat | ||
dat[2] | ||
dat[2].max | ||
max(dat[2]) | ||
source "box-plot.r" | ||
source("box-plot.r") | ||
source("box-plot.r") | ||
source("box-plot.r") | ||
source("box-plot.r") | ||
ls | ||
dat | ||
cached_latency | ||
cached_latencies | ||
cached_latency | ||
cached_latency = unlist(na.omit(da[1])) | ||
cached_latency = unlist(na.omit(dat[1])) | ||
no_cached_latency = unlist(na.omit(dat[2])) | ||
cached_latency | ||
hist(cached_latency, xlim=c(0, 1000000)) | ||
cached_latency | ||
hist(cached_latency, xlim=c(0, 10000000)) | ||
hist(cached_latency, xlim=c(0, 1000000)) | ||
hist(cached_latency, xlim=c(0, 1000000), breaks=12) | ||
hist(cached_latency, xlim=c(0, 10000000), breaks=12) | ||
hist(cached_latency, breaks=12) | ||
hist(cached_latency, xlim=c(0, 10000000)) | ||
cached_latency | ||
dat[1] | ||
plot(cached_latency) | ||
cached_latency | ||
plot(cached_latency, col='red') | ||
plot(cached_latency, col='red') | ||
plot(non_cached_latency, col='blue') | ||
plot(no_cached_latency, col='blue') | ||
plot(cached = cached_latency, non_cached = no_cached_latency) | ||
plot(cached_latency, col='red') | ||
hist(dat) | ||
plot(dat) | ||
plot(dat[1]) | ||
plot(dat[2]) | ||
plot(cached_latency, col='red') | ||
par(new=T) | ||
plot(no_cached_latency, col='blue') | ||
plot(cached_latency, col='red') | ||
lines(no_cached_latency, col='blue') | ||
points(no_cached_latency, col='blue') | ||
plot(cached_latency, col='red') | ||
lines(no_cached_latency, col='blue') | ||
plot(cached_latency, col='red') | ||
points(no_cached_latency, col='blue') | ||
title(ylab="access time (in nanoseconds)") | ||
title(ylab=NULL) | ||
range(cached_latency) | ||
hist(cached_latency, xlim=c(11034, 11034*10)) | ||
hist(cached_latency, xlim=c(11034, 11034*10), breaks=200) | ||
hist(cached_latency, xlim=c(11034, 11034*10), breaks=200) | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
ls | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
ls | ||
source('histogram.r') | ||
hist(cached_latency) | ||
hist(cached_latency, buckets=200) | ||
hist(cached_latency, breaks=200) | ||
mean(cached_latency) | ||
median(cached_latency) | ||
hist(cached_latency, breaks=200, xlim=(0, 1000000)) | ||
hist(cached_latency, breaks=200, xlim=c(0, 1000000)) | ||
hist(cached_latency, breaks=200, xlim=c(0, 10000000)) | ||
hist(cached_latency, breaks=200, xlim=c(0, 5000000)) | ||
hist(cached_latency, breaks=20, xlim=c(0, 5000000)) | ||
hist(cached_latency, breaks=2000, xlim=c(0, 5000000)) | ||
hist(cached_latency, breaks=600, xlim=c(0, 5000000)) | ||
hist(cached_latency, breaks=600, xlim=c(0, 2000000)) | ||
hist(cached_latency, breaks=600, xlim=c(0, 1000000)) | ||
hist(cached_latency, breaks=1000, xlim=c(0, 1000000)) | ||
hist(cached_latency, breaks=1000, xlim=c(0, 1000000)) | ||
hist(cached_latency, breaks=1000, xlim=c(0, 500000)) | ||
hist(cached_latency, breaks=1000, xlim=c(0, 1000000)) | ||
hist(cached_latency, breaks=1000, xlim=c(0, 800000)) | ||
hist(cached_latency, breaks=100, xlim=c(0, 800000)) | ||
hist(cached_latency, breaks=10000, xlim=c(0, 800000)) | ||
hist(cached_latency, breaks=10000, xlim=c(0, 800000), color='red') | ||
hist(cached_latency, breaks=10000, xlim=c(0, 800000), col='red') | ||
hist(no_cached_latency, breaks=10000, col='blue') | ||
hist(cached_latency, breaks=10000, xlim=c(0, 800000), col='red') | ||
hist(no_cached_latency, breaks=10000, col='blue', add=T) | ||
mean(no_cache_latency) | ||
mean(no_cached_latency) | ||
hist(no_cached_latency, breaks=10000, col='red', xlim=c(0,1600000)) | ||
hist(no_cached_latency, breaks=10000, col='red', xlim=c(0,16000000)) | ||
hist(no_cached_latency, breaks=1000, col='red', xlim=c(0,16000000)) | ||
hist(cached_latency, breaks=1000, col='red', add=T) | ||
hist(cached_latency, breaks=100, col='red', add=T) | ||
hist(cached_latency, breaks=100, col='blue', add=T) | ||
hist(no_cached_latency, breaks=10000, col='red', xlim=c(0,16000000)) | ||
hist(cached_latency, breaks=1000, col='blue', add=T) | ||
hist(cached_latency, breaks=10000, col='blue', add=T) | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('histogram.r') | ||
source('box-plot.r | ||
') | ||
source('box-plot.r') | ||
ls | ||
source('point-plot.r') | ||
source('latencies.csv') | ||
source("point-plot.r") | ||
cat | ||
cat("point-plot.r") | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r) | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plox.r') | ||
source("box-plot.r") | ||
dat <- read.csv("latencies.csv") | ||
cached = unlist(na.omit(dat[1])) | ||
noncached_data = unlist(na.omit(dat[2])) | ||
df <- data.frame(metacached = cached_data, non_metacached = noncached_data) | ||
boxplot(df, ylim=c(0, 4000000), col=c('chocolate1', 'cornflowerblue'), xaxt="n", cex.main=1.5) | ||
title("metacached") | ||
par(cex=3.0) | ||
boxplot(df, ylim=c(0, 4000000), col=c('chocolate1', 'cornflowerblue'), xaxt="n", cex.main=1) | ||
help('par') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
source('box-plot.r') | ||
data_latencies <- read.csv("data/latencies.csv") | ||
cached = unlist(na.omit(data_latencies)) | ||
cached = unlist(na.omit(data_latencies[1])) | ||
noncached = unlist(na.omit(data_latencies[2])) | ||
std(cached) | ||
sd(cached) | ||
sd(noncached) | ||
sd(noncached)/sd(cached) | ||
metadata_latencies <- read.csv("data/multinode_meta.csv") | ||
cached = unlist(na.omit(metadata_latencies[1])) | ||
noncached = unlist(na.omit(metadata_latencies[2])) | ||
sd(noncached)/sd(cached) | ||
dat | ||
dat <- read.csv("latencies.csv") | ||
df <- data.frame(dat) | ||
ggplot(df, aes=(x="Panthera (dls | ||
ggplot(df, aes=(x="Panthera (datacached) latency")) + geom_histogram() | ||
library(ggplot2) | ||
ggplot(df, aes=(x="Panthera (datacached) latency")) + geom_histogram() | ||
ggplot(df, aes(x="Panthera (datacached) latency")) + geom_histogram() | ||
ggplot(df, aes(x="Panthera (datacached) latency", data=cached)) + geom_histogram() | ||
ggplot(df, aes(x=cached)) + geom_histogram() + xlab("Panthera (datached) latency") | ||
ggplot(df, aes(x=cached)) + geom_histogram() + xlab("Panthera (datacached) latency") + title("Panthera (Datacached) Latency Distribution") | ||
ggplot(df, aes(x=cached)) + geom_histogram() + xlab("Panthera (datacached) latency") + ggtitle("Panthera (Datacached) Latency Distribution") | ||
pdf("datacached-latency-distribution-blackwhite.pdf") | ||
ggplot(df, aes(x=cached)) + geom_histogram() + xlab("Panthera (datacached) latency") + ggtitle("Panthera (Datacached) Latency Distribution") | ||
dev.off() | ||
dat <- read.csv("data/latencies.csv") | ||
dat[1] | ||
dat | ||
sd(unlist(na.omit(dat[1])) | ||
) | ||
sd(unlist(na.omit(dat[2])) | ||
) | ||
sd(unlist(na.omit(dat[2])) | ||
) | ||
dat <- read.csv("data/old_meta_latencies.csv") | ||
sd(unlist(na.omit(dat[1])) | ||
) | ||
sd(unlist(na.omit(dat[2])) | ||
) | ||
dat <- read.csv("data/multinode_meta.csv") | ||
sd(unlist(na.omit(dat[1])) | ||
) | ||
dat <- read.csv("data/multinode_meta.csv") | ||
sd(unlist(na.omit(dat[2])) | ||
) | ||
ls | ||
source('box-plot.r') | ||
source('box-plot.r') |
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,17 @@ | ||
#read in the data | ||
dat <- read.csv("latencies.csv", head = TRUE, sep=",") | ||
|
||
cached_data = unlist(na.omit(dat[1])) | ||
noncached_data = unlist(na.omit(dat[2])) | ||
|
||
df <- data.frame(metacached = cached_data, non_metacached = noncached_data) | ||
pdf("boxplot.pdf", width=9.5, height=7) | ||
par(mar=c(5, 4.5, 4, 2)) | ||
boxplot(df, ylim=c(0, 40000000), col=c('chocolate1', 'cornflowerblue'), xaxt="n", cex.main=1.5, whisklwd=4, staplelwd=4, outlwd=2, outcex=1.2) | ||
par(cex.main=1.9) | ||
title("Metacached (Panthera) vs Non-metacached (Status Quo)\nMultinode") | ||
par(cex.axis=1.8) | ||
axis(1, at=1:2, labels=c("metacached (Panthera)", "non-metacached")) | ||
par(cex.lab=1.8) | ||
title(ylab = "Access Time (nanoseconds, lower is better)") | ||
dev.off() |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,11 @@ | ||
#read in the data | ||
dat <- read.csv("latencies.csv", head = TRUE, sep=",") | ||
|
||
cached_data = unlist(na.omit(dat[1])) | ||
noncached_data = unlist(na.omit(dat[2])) | ||
|
||
df <- data.frame(cached = cached_data, noncached = noncached_data) | ||
png("boxplot.png") | ||
boxplot(df) | ||
title("Panthera (cached) vs Status Quo (non-cached) - Data") | ||
|
Binary file not shown.
Oops, something went wrong.