I never understand it when someone comes up with a great tool, then defaults it to work poorly… leaving the rest up to imagination. The Pandas dataframe has a great and underutilized tool… to_sql() . Lesson learned, always read the fine print I guess. I’m usually guilty of this myself… wondering why something in slow and sucks… and not taking time to read the documentation. Here are some musings on using the to_sql() in Pandas and how you should configure to not pull your hair out.
I want to interrupt your semi-regularly scheduled technical blog post for this public service announcement. I mean the url does say “confessions” does it not? For better or worse I’ve been thinking a lot lately about what it means to be a Data Engineer, what’s like to be a Data Engineer, and what makes a good Data Engineer. Just the life of a Data Engineer in general. The Battlefield of the Data Engineer is fought in the labyrinth of nested SQL queries. It rages to the depths of distributed computing clusters. It vies for victory on the crags and peaks of DevOps. It attacks for precious ground amid the chaos of the perfect OOP and Functional code bases. Phew… and all that just to keep your head above water.
Read more
Streams, streams, streams…. when will it ever end? It’s hard to keep up with all the messaging systems these days. GCP PubSub, AWS SQS, RabbitMQ, blah blah. Of course there is Kafka, hard to miss that name floating around in the interwebs. Since pretty much every system designed these days is a conglomerate of services… it’s probably a good idea to poke at things under the cover. Of course Apache Kafka is probably at the top of list of those open source streaming services. Today I’m going to attempt to install a Kafka cluster and push some messages around.
Read more
So what’s up with Apache Hive? It’s been around a long time…but all the sudden it seems like it’s requirement in every other job posting these days. “It’s not you… it’s me.” That’s what I would tell Hive if it suddenly materialized as Mr. Smith via the Matrix that I’m pretty sure is the new reality these days. I’ve been around Hadoop and Spark for awhile now and I feel like Hive is that weird 2nd cousin who shows up at Thanksgiving. You know you should like and be nice to him, but you’re not sure why. It seems like Hive sits in a strange world. It’s not a RDBMS, although it does ACID, but it’s touted as a Data Warehousing tool. Time to dig in.
Read more
Yikes, distributed geospatial data processing at scale. That has fun written all over it… not. There isn’t that many people doing it so StackOverflow isn’t that useful. Anyone who has been around geospatial data knows the tools like GDAL are notoriously hard to use and buggy… and that one’s probably the “best.” What to do when you want to process and explore large satellite datasets like Landsat and Modis? Terrabytes/petabytes of data, what are going to do, download it? The power of distributed processing with Apache Spark. The simplicity of using SQL to work on geospatial data. Put them together… rasterframes. What a beast.
Read more
Seriously. Haven’t you had enough of SSIS, SAP Data Services, Informatica, blah blah blah? It’s been the same old ETL process for the last 20 years. CSV files appear somewhere, some poor old aged and angry Developer soul in a cubicle pulls up the same old GUI ETL tool, maps a bunch of columns to some SQL Server, if you’re in a forward thinking shop…maybe Postgres. This is after painstakingly designing the Data Warehouse with good ole’ Kimball in mind. Data flows from some staging table to some facts and dimensions. Eventually some SQL queries are run and a Data Mart is produced summarizing a years worth of data for a crabby Sales or Product department. Brings a tear to my eye. And this is all because Apache Spark sounds scary to some people?
Read more
I’ve never seen so many posts about Apache Spark before, not sure if it’s 3.0, or because the world is burning down. I’ve written about Spark a few times, even 2 years ago, but it still seems to be steadily increasing in popularity, albeit still missing from many companies tech stacks. With the continued rise os AWS Glue and GCP DataProc, running Spark scripts without managing a cluster has never been easier. Granted, most people never work on datasets large enough to warrant the use of Spark.. and Pandas works fine for them. Also, very annoyingly it seems most videos/posts on Spark about shuffling/joins blah blah that make no sense to someone who doesn’t use Spark on daily basis, or they are so “Hello World” as to be useless in the real world. Let’s solve that problem by setting up our own Spark/Hadoop cluster and doing some “real” data things with it.
Read more
Man. Every time I open IntelliJ to write/learn some more Scala I have to take a deep breath. Yes, it’s been fun and good for my to my brain feel like in a Doctor Strange movie, but it’s also challenging and frustrating at times. One of the things I find myself doing a lot as a Data Engineering is HTTP stuff, mostly pulling files or data from APIs. Doing this work in Python is most enjoyable and easy, I’ve been curious to see how Scala handles HTTP stuffy stuff.
Read more
Has anyone else noticed how popular Apache Airflow and Kubernetes have become lately? There is no better tool than Airflow for Data Engineers to built approachable and maintainable data pipelines. I mean Python, a nice UI, dependency graphs/DAGs. What more could you want? There is also no better tool than Kubernetes for building scalable, flexible data pipelines and hosting apps. Like a match made in heaven. So why not deploy Airflow onto Kubernetes? This is what you wish your mom would have taught you. It’s actually so easy your mom could probably do it….maybe she did do it and just never told you?
Read more
Database design… hmmm. There is probably nothing more all over the board in tech. Data warehousing, analytics, OLTP… everyone with their own “defend this hill to the death” ideas. Kimball vs Inmon. Hmmm.. what to do, what to do? After defending my own hills to the death over the years and arguing over whiteboards I’ve come to a conclusion. The right answer is somewhere in the middle. Understanding a few basic design principals will help any data engineer master writing DDL for anything from a Data Warehouse to a high load OLTP systems… across all RDBMS platforms.
Read more