The Evolution Of Hadoop

Holden Ackerman has an interesting analysis of Qubole customers’ adoption of Hadoop 2:

In Qubole’s 2018 Data Activation Report, we did a deep-dive analysis of how companies are adopting and using different big data engines. As part of this research, we found some fascinating details about Hadoop that we will detail in the rest of this blog.

A common misconception in the market is that Hadoop is dying. However, when you hear people refer to this, they often mean “MapReduce” as a standalone resource manager and “HDFS” as being the primary storage component that is dying. Beyond this, Hadoop as a framework is a core base for the entire big data ecosystem (Apache Airflow, Apache Oozie, Apache Hbase, Apache Spark, Apache Storm, Apache Flink, Apache Pig, Apache Hive, Apache NiFi, Apache Kafka, Apache Sqoop…the list goes on).

I clipped this portion rather than the direct analysis because I think it’s an important point:  the Hadoop ecosystem is thriving as the matter of primary importance switches from what was important a decade ago (batch processing of large amounts of data on servers with direct attached storage) to what is important today (a combination of batch and streaming processing of large amounts of data on virtualized and often cloud-based servers with network-attached flash storage).

How Spark Works: RDDs And DAGs

Shubham Agarwal gets into the way that Spark translates operations on Resilient Distributed Datasets into actions:

When we do a transformation on any RDD, it gives us a new RDD. But it does not start the execution of those transformations. The execution is performed only when an action is performed on the new RDD and gives us a final result.

So once you perform any action on an RDD, Spark context gives your program to the driver.

The driver creates the DAG (directed acyclic graph) or execution plan (job) for your program. Once the DAG is created, the driver divides this DAG into a number of stages. These stages are then divided into smaller tasks and all the tasks are given to the executors for execution.

Click through for more details.

Tracking Down Long-Running xp_cmdshell Processes

Thomas Rushton investigates what’s taking so long with an xp_cmdshell call:

I wanted to know what he was up to, but the sql_text field only gives “xp_cmdshell”, not anything useful that might help to identify what went wrong.

So we have to go to Taskmanager on the server. On the “Process Details” page, you can select which detail columns you want to see. We want to see the Command Line, as that’ll tell us if it’s some manually-launched batch job that’s failed or something else going wrong.

An alternative to using the Task Manager is to open ProcMon, part of the Sysinternals toolset.  It takes a bit of getting used to, but is quite powerful once you know its ins and outs.

Creating Seaborn Plots With R

Abdul Majed Raja shows how to call Python from R and build plots using the Seaborn Python package:

The reticulate package provides a comprehensive set of tools for interoperability between Python and R. The package includes facilities for:

  • Calling Python from R in a variety of ways including R Markdown, sourcing Python scripts, importing Python modules, and using Python interactively within an R session.
  • Translation between R and Python objects (for example, between R and Pandas data frames, or between R matrices and NumPy arrays).
  • Flexible binding to different versions of Python including virtual environments and Conda environments.

Reticulate embeds a Python session within your R session, enabling seamless, high-performance interoperability.

The more common use of reticulate I’ve seen is running TensorFlow neural networks from R.

Getting The Current Date And Time In SQL Server

Randolph West shows a few functions which can retrieve current date and time information:

What do we mean by local date and time?

As discussed previously, SQL Server is not time zone aware, nor does it have to be. This is because the operating system that SQL Server runs on can have multiple custom regional settings depending on which user is logged into the server.

This holds true for the SQL Server service account as well, which is just another user on the operating system. When any of these functions is called, it is asking for the date and time from the operating system.

If you’re going to use DATETIME2 (which you generally should), take advantage of the precision that SYSUTCDATETIME() gives you over GETUTCDATE().

SQL Operations Studio April Release

Alan Yu announces the April release of SQL Operations Studio:

Highlights for this build include the following.

  • Public preview release of SQL Agent extension

  • Added new extensions and improved existing extensions

    • Improvements to Server Reports Extension
    • Release of SSMS Keymap extension
    • Release of AlwaysOn Insights extension
    • Release of MSSQL Instance Insights
    • Release of MSSQL Db Insights
  • Added Visual Studio Code 1.21 platform source code refresh

    • Improved large and protected file support for saving Admin protected and >256M files within SQL Ops Studio
    • Integrated Terminal splitting to work with multiple open terminals at once
    • Reduced installation on-disk file count footprint for faster installs and startup times
  • Continue to fix GitHub issues

There’s a lot in here.

Takeaways From Implementing Power BI Embedded

Meagan Longoria has some thoughts after a proof of concept using Power BI Embedded:

After making changes and testing your report, make sure to clear any slicer values before publishing, if you have row-level security on a field shown in a slicer and you leave values selected. The selected values will be shown to users when they view the report. For example, let’s say you have created a row-level security role that can only see Product A, but you can see everything, and you left Product A and Product B selected and deployed the report. A user who views the report next and is a member of that RLS role will see the two selected values in the slicer, even though they can’t see the data for Product B on the page. This may not be a big deal for an internal report. But now imagine this is for clients. You don’t want clients to see other clients in the list. This behavior is consistent in the Power BI web service and isn’t specific to embedding. It’s just important to remember this.

There are plenty of interesting notes here, so check it out if you’re thinking of a Power BI project.

Creating A Custom Calendar Table With Power Query

Matt Allington shows how to create a calendar table which allows users to set the start and end dates:

My approach to teaching people to use Power Query is to always use the UI where possible.  I first use the UI to do the hard work, then jump in and make small changes to the code created by the UI to meet any specific variations required.  Keep this concept in mind as you read this article.

I am going to use Power BI Desktop as the tool for this, but of course Power Query for Excel will work just as well and the process is identical.  In fact the calendar query at the end can easily be cut and pasted between Power BI and Power Query for Excel.

Check it out for another method for building calendar tables.  I tend to build them in SQL Server because that’s what I’m most familiar with, but it’s good to know a few different ways of doing this.

Azure Data Lake Alerting

Jose Lara shows how to send alerts if you hit a utilization threshold:

If you want to see the step-by-step guide to create a new Log Analytics alert, check out our recent blog post on creating Log Analytics Alerts.

For the alert signal logic, use the following values:

  • Use the query from the previous step

  • Set the sum of AUs to 50 as the threshold (you can use any number that reflects your own threshold)

  • Set the trigger to 0: whenever the threshold is breached

  • Set the period and frequency for 24 hours.

Read the whole thing if you use Azure Data Lake Analytics; an unexpectedly large bill is a tough thing to swallow.

Creating Map Plots With ggmap

Laura Ellis shows how to use the ggmap package to create choropleth maps in R:

In the last map, it was a bit tricky to see the density of the incidents because all the graphed points were sitting on top of each other.  In this scenario, we are going to make the data all one color and we are going to set the alpha variable which will make the dots transparent.  This helps display the density of points plotted.

Also note, we can re-use the base map created in the first step “p” to plot the new map.

Check it out.  This is an introduction to creating choropleths, making it a good start.


April 2018
« Mar