"<small><i>This notebook was prepared by [Donne Martin](http://donnemartin.com). Source and license info is on [GitHub](https://github.com/donnemartin/data-science-ipython-notebooks).</i></small>"
"Before I discovered [S3cmd](http://s3tools.org/s3cmd), I had been using the [S3 console](http://aws.amazon.com/console/) to do basic operations and [boto](https://boto.readthedocs.org/en/latest/) to do more of the heavy lifting. However, sometimes I just want to hack away at a command line to do my work.\n",
"\n",
"I've found S3cmd to be a great command line tool for interacting with S3 on AWS. S3cmd is written in Python, is open source, and is free even for commercial use. It offers more advanced features than those found in the [AWS CLI](http://aws.amazon.com/cli/)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Install s3cmd:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"!sudo apt-get install s3cmd"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Running the following command will prompt you to enter your AWS access and AWS secret keys. To follow security best practices, make sure you are using an IAM account as opposed to using the root account.\n",
"\n",
"I also suggest enabling GPG encryption which will encrypt your data at rest, and enabling HTTPS to encrypt your data in transit. Note this might impact performance."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"!s3cmd --configure"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Frequently used S3cmds:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"# List all buckets\n",
"!s3cmd ls\n",
"\n",
"# List the contents of the bucket\n",
"!s3cmd ls s3://my-bucket-name\n",
"\n",
"# Upload a file into the bucket (private)\n",
"!s3cmd put myfile.txt s3://my-bucket-name/myfile.txt\n",
"\n",
"# Upload a file into the bucket (public)\n",
"!s3cmd put --acl-public --guess-mime-type myfile.txt s3://my-bucket-name/myfile.txt\n",
"\n",
"# Recursively upload a directory to s3\n",
"!s3cmd put --recursive my-local-folder-path/ s3://my-bucket-name/mydir/\n",
"\n",
"# Download a file\n",
"!s3cmd get s3://my-bucket-name/myfile.txt myfile.txt\n",
"\n",
"# Recursively download files that start with myfile\n",
"!s3cmd --recursive get s3://my-bucket-name/myfile\n",
"\n",
"# Delete a file\n",
"!s3cmd del s3://my-bucket-name/myfile.txt\n",
"\n",
"# Delete a bucket\n",
"!s3cmd del --recursive s3://my-bucket-name/\n",
"\n",
"# Create a bucket\n",
"!s3cmd mb s3://my-bucket-name\n",
"\n",
"# List bucket disk usage (human readable)\n",
"!s3cmd du -H s3://my-bucket-name/\n",
"\n",
"# Sync local (source) to s3 bucket (destination)\n",
"[S3DistCp](http://docs.aws.amazon.com/ElasticMapReduce/latest/DeveloperGuide/UsingEMR_s3distcp.html) is an extension of DistCp that is optimized to work with Amazon S3. S3DistCp is useful for combining smaller files and aggregate them together, taking in a pattern and target file to combine smaller input files to larger ones. S3DistCp can also be used to transfer large volumes of data from S3 to your Hadoop cluster."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"To run S3DistCp with the EMR command line, ensure you are using the proper version of Ruby:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"!rvm --default ruby-1.8.7-p374"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The EMR command line below executes the following:\n",
"* Create a master node and slave nodes of type m1.small\n",
"* Runs S3DistCp on the source bucket location and concatenates files that match the date regular expression, resulting in files that are roughly 1024 MB or 1 GB\n",
"For further optimization, compression can be helpful to save on AWS storage and bandwidth costs, to speed up the S3 to/from EMR transfer, and to reduce disk I/O. Note that compressed files are not easy to split for Hadoop. For example, Hadoop uses a single mapper per GZIP file, as it does not know about file boundaries.\n",
"\n",
"What type of compression should you use?\n",
"\n",
"* Time sensitive job: Snappy or LZO\n",
"* Large amounts of data: GZIP\n",
"* General purpose: GZIP, as it’s supported by most platforms\n",
"\n",
"You can specify the compression codec (gzip, lzo, snappy, or none) to use for copied files with S3DistCp with –outputCodec. If no value is specified, files are copied with no compression change. The code below sets the compression to lzo:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"--outputCodec,lzo"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<h2 id=\"redshift\">Redshift</h2>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Copy values from the given S3 location containing CSV files to a Redshift cluster:"
"The CANCEL command will not abort a transaction. To abort or roll back a transaction, you must use the ABORT or ROLLBACK command. To cancel a query associated with a transaction, first cancel the query then abort the transaction.\n",
"\n",
"If the query that you canceled is associated with a transaction, use the ABORT or ROLLBACK. command to cancel the transaction and discard any changes made to the data:"
"When you create a table, you can specify one or more columns as the sort key. Amazon Redshift stores your data on disk in sorted order according to the sort key. How your data is sorted has an important effect on disk I/O, columnar compression, and query performance.\n",
"\n",
"Choose sort keys for based on these best practices:\n",
"\n",
"If recent data is queried most frequently, specify the timestamp column as the leading column for the sort key.\n",
"\n",
"If you do frequent range filtering or equality filtering on one column, specify that column as the sort key.\n",
"\n",
"If you frequently join a (dimension) table, specify the join column as the sort key."
"When you create a table, you designate one of three distribution styles: KEY, ALL, or EVEN.\n",
"\n",
"**KEY distribution**\n",
"\n",
"The rows are distributed according to the values in one column. The leader node will attempt to place matching values on the same node slice. If you distribute a pair of tables on the joining keys, the leader node collocates the rows on the slices according to the values in the joining columns so that matching values from the common columns are physically stored together.\n",
"\n",
"**ALL distribution**\n",
"\n",
"A copy of the entire table is distributed to every node. Where EVEN distribution or KEY distribution place only a portion of a table's rows on each node, ALL distribution ensures that every row is collocated for every join that the table participates in.\n",
"\n",
"**EVEN distribution**\n",
"\n",
"The rows are distributed across the slices in a round-robin fashion, regardless of the values in any particular column. EVEN distribution is appropriate when a table does not participate in joins or when there is not a clear choice between KEY distribution and ALL distribution. EVEN distribution is the default distribution style."
