Week of June 26th
“…And I think to myself… What a wonderful world 🌎 .”
Happy Coconut🥥 Day!
Recently, I had been spending so much time ⏰ in GCP land☁️ that it started to feel like it was my second home 🏡 . However, it was time for a little data sabbatical. I needed to visit a land of mysticism✨ and intrigue. A place where developers can roam freely and where data can be flexible, semi-structured, hierarchical nature, and can be easily scaled horizontally… A place not bound to the rigidness of relational tables but a domicile of flexible documents. We would journey to the world of MongoDB🍃.
Ok, so we have been there before, but we needed a refresher. It had been about 6 years since we first became acquainted with this technological phenomenon. Besides we hadn’t played around too much with some of the company’s past innovations like MongoDB🍃 Compass 🧭 a sleek visual environment that allows you to analyze and understand the contents of your data in MongoDB🍃 and MongoDB🍃 Atlas☁️ the managed service used to provision, maintain and scale MongoDB🍃 clusters of instances that is conveniently offered on AWS, Azure and GCP.
To assist us on getting started would be our old comrade in arms, Pinal Dave from SQLAuthority fame. Pinal had put together an outstanding condensed course on Foundations of Document Databases with MongoDB
So, this is where we would begin. The course commences with an introduction on NoSQL (Not Just SQL) databases and some the advantages of a Document Database like an Intuitive Data Model, dynamic Schema and distributed Scalable Database. Then he gives us comprehensible explanation to the CAP (Consistency, Availability and Partition Tolerance) Theorem and that only 2/3 are necessary. MongoDB🍃 fits in under the CP variety while compromising on availability. Next, the following key 🔑 points are made in relation to MongoDB🍃
- All write operations in MongoDB🍃 are atomic on the level of a single document
- If the collection does not currently exist, the insert operator will create one in the collection.
Next, Pinal takes use through a few quick and easy steps on how to get setup with MongoDB Atlas☁️. Once, our fully managed MongoDB🍃 Cluster was fired🔥 up it was time to navigate our collections with MongoDB Compass🧭
For much of the rest of the course, we would concentrate on CRUD (Create, Read, Update, Delete) operations in MongoDB🍃 through both Compass🧭 and the CLI (Mongo Shell).
The course would also present us a terse walk through with syntax and elucidation on Read Concerns and Write Concerns in MongoDB🍃
Allows to control the consistency and isolation properties of the data read from replica sets and replica set shards
- Local – (No guarantee the data has applied to all Replicas) – Primary
- Available – (No guarantee the data has applied to all Replicas) – Secondary
- Majority – (default) acknowledged by a majority to all Replicas
- Linearizable – All successful acknowledged by a majority to all Replicas before Read (Query might have to wait)
- Snapshot – Used with multi document projection data from the majority to all Replicas
Level of acknowledgement requested from MongoDB🍃 for write operations
w:1 – Ack from primary
w:0 – No ack
w(n) – Primary + (n-1) secondary
Timeout: Time limit to prevent write operations from blocking indefinitely
As lasting point on database writes (UD), all write operations in MongoDB🍃 are atomic on the level of a single document. In other words, if you are updating a single document or if you are updating multiple documents in a collection at any time every single update is just atomic at a single document.
Closing time 🕰 … Time 🕰 for you to go out go out into the world 🌎 … Closing time 🕰…
Finally, dénouements of the course are on Common SQL Concepts and Semantics to MongoDB🍃 including some of the major differences between a typical RDBS and MongoDB🍃which can be represented by the table below:
|SQL||MQL (Mongo Query Language)|
|Predefined Schema||Dynamic Schema|
|Relational Keys||No foreign key|
|ACID Properties||CAP theorem|
Sadly 😢, that was it for Foundations of Document Databases with MongoDB. This left us clambering for more. Fortunately, Nuri Halperin happily😊 delivered and then some… Nuri a MongoDB🍃 Guru 🧙♂️ and a Love❤️ 👨⚕️ of sorts (known for creating the wildly popular jdate.com platform) put together a series of timeless MongoDB🍃 courses that have managed to stand up through the test of time 🕰. In which, I might add is not an easy feat when it comes to a burgeoning technologies like MongoDB🍃.
Part 1: Introduction course Introduction to MongoDB in-depth look at both the Mongo Shell (CLI) and CRUD Syntax and Indexing
Part 2: MongoDB Administration takes a deep dive into MongoDB🍃 administration key concepts i.e. installation, configuration, Security, Backup/Restores, Monitoring, High Availability and Performance
Nuri like Pinal, discusses some of the challenges found in Relational Databases like Impedance mismatch and the need for Object-relational mapping (ORM) for developers. He demonstrates how MongoDB🍃 solves these challenges through its schema-less approach and no relationships required model. In addition, he touches on to how MongoDB🍃 lends itself nicely to data polymorphism.
Next he takes us through the MongoDB architecture a collection humongous arrays that utilizes memory mapped BSON (Binary script object notation) or simply “Binary JSON” (Java Script object notation) files. MongoDB🍃 intuitively leverages the OS to handle the loading of data and saving to disk which allows the engine to center on speed, optimization, and stability.
MongoDBs🍃 main mission is to just serve up data quickly and efficiently. Next, Nuri takes through the Mongo Shell (CLI) which basically is just a Java☕️Script interpreter that allows you to interactively get insight into the MongoDB🍃 Server. Further he discusses indexes, types of indexes, and how paramount indexes are in MongoDB🍃 for practical usability.
Lastly, Nuri takes through MongoDB🍃 Replication which uses the simple to configure but highly scalable replica sets. This is how MongoDB🍃 achieves “Eventual Consistency”, Automatic Failover, and Automatic recovery… And this is just the introduction of the Part 2 of the course…
Like trying to watch all 3 parts of The Lord of Rings 💍 (Director’s Edition) trilogy in a single helping, it’s just wasn’t possible to complete all of Nuri’s two part sequel in a single week but we did get through most of it. 😊 This also left us with a little bit more on our plate🍽 as we continue through our Mongo Journey…
This Week’s Log
Out of the tree 🌳 of life I just picked me a plum… You came along and everything started’ in to hum 🎶… Still it’s a real good bet… The best is yet to come
Below are some topics I am considering for my voyage next week:
- More with Nuri and MongoDB
- Cosmos DB
- More with Google Cloud Path
- Working with Parquet files
- JDBC Drivers
- More on Machine Learning
- ONTAP Cluster Fundamentals
- Data Visualization Tools (i.e. Looker)
- Additional ETL Solutions (Stitch, FiveTran)
- Process and Transforming data/Explore data through ML (i.e. Databricks)
Stay safe and Be well –