Why I joined SingleStore
Observations and Experiences from my Career in IT
Abstract
In this article, I discuss some of my observations and experiences in IT and my frustrations around the challenge of learning new technologies. Through my career in IT, I have learned some valuable lessons, and these have led me to my new role as a Technical Evangelist, where I help others learn to make technology work for them.
Introduction
I recently joined the Developer Relations team at SingleStore. In this article, I wanted to summarise my observations and experiences in IT and share my main reason for joining SingleStore.
First, a bit of history and context. In my career in IT, I originally started as a developer working on real-time systems and Relational Databases. However, the first year as a developer was one of the most painful in my career. Having graduated and thinking that I would work on the latest and greatest technologies with my first employer, I was disappointed to find myself supporting and maintaining very old software that few people around me knew anything about, using a programming language that I had never heard of. However, with the benefits of hindsight, I later realised that those early experiences were beneficial. They taught me to be a self-starter, to chase up and ask colleagues for help, read any and all documentation, book time on shared resources/testbeds, code and test, and document everything that I learned. In short, my goal in the end was to make it easy for someone else to pick up the project from me without having to go through the pain that I did.
Since that first role, I have worn many hats and worked in many different roles, such as consultancy, product strategy, evangelism, technical writing and technical training. These roles have been with several Blue Chip companies and Big Data startups. As time has passed, I have also observed the speed of innovations in IT and the growth, maturity and acceptance of Cloud Computing.
Object-Oriented Database Management Systems
In Database Systems, my area of interest, new technologies have come and sometimes gone. Think back to the late 1980s and early 1990s and the growth of Object-Oriented Database Management Systems (OODBMSs). At the time, it was the next big thing. It achieved some market success with customers using the technology in production environments. For example, some case studies can be found in Object Databases in Practice and Succeeding with Object Databases. Analysts also predicted that the technology would achieve strong growth using graphs that looked something like Figure 1.
Today, however, very few of those vendors survive. Some vendors were acquired, some fell by the wayside, and one or two are still around today. There was much talk of this technology being the next generation of Database Systems, replacing Relational Database Systems. It never happened. However, the disruption and innovations that OODBMSs brought caused the Relational Database vendors to add new capabilities. There was also the emergence of Object-Relational Database Systems.
XML Database Management Systems
Fast forward to ten years later and lather, rinse, and repeat. At that time, XML was the next big thing. XML Database Systems achieved some market success with customers using the technology in production environments. For example, some case studies can be found in XML Data Management: Native XML and XML-Enabled Database Systems. Analysts also predicted that the technology would achieve strong growth using graphs that looked something like Figure 2.
But this time, there was an extensive choice of XML Database products. Ron Bourret, for example, used to maintain a website with a vast list of XML Database Products. Some of the OODBMS vendors also repositioned their products as XML servers. There was much talk of this technology being the next generation of Database Systems, replacing Relational Database Systems. It never happened. However, the savvy Relational Database vendors again innovated and added XML support to their products.
NoSQL Database Management Systems
Fast forward another ten years and lather, rinse, and repeat. This time, NoSQL was the new hot technology. Analysts also predicted that the technology would achieve strong growth using graphs that looked something like Figure 3.
Whilst NoSQL has gained some traction in the market, the overall market share of this technology remains very small compared to the Relational Database market. Becoming proficient with these products also takes time. For example:
- Building an Application upon Riak — Part 1
- Cassandra Summit 2013
- JPMorgan consolidates derivative trade systems with NoSQL database
In a book by Roger Sessions titled COM and DCOM: Microsoft’s Vision for Distributed Objects, he used the example of a large water tank to represent corporate data. He compared OODBMSs to Relational Database Systems, and we could extend this to NoSQL against Relational Database Systems. If the water tank had the dimensions of 200 feet (width) by 100 feet (depth) by 9 miles (length), the quantity of information currently stored in NoSQL Database Systems would not quench a thirst on a hot day. However, the disruption by NoSQL has again caused the Relational Database vendors to innovate and respond.
By now, you will have realised that there is a particular pattern in the different Database technologies we’ve discussed so far. New Database technologies emerged, and there were predictions about the imminent demise of Relational Database Systems. However, Relational Database vendors adapted, and this technology still accounts for the vast majority of Database installations today. Relational vendors have also added significant new capabilities, and products can be easily deployed in cloud environments.
Cloud Computing
Today, the cloud is also where many companies choose to run their IT operations; pay someone else to use their resources and let them take care of the plumbing and infrastructure. This allows business owners to focus on what they do best. However, the choice of Big Data and related technologies available today is mind-boggling, as discussed in The 2020 Data & AI Landscape. There is far too much that developers must learn.
It would be great to provision systems in the cloud and scale up and scale down, as required. To use the power of distributed computing to get valuable insights into data and help business owners make important decisions much faster. This could be achieved using a product built upon the mature, well-understood and solid foundations of Relational technology but with newly added capabilities to meet needs such as working with JSON documents, time-series data, and fast analytics. SingleStoreDB can do all of this and more. One product that can meet multiple needs, saving developers frustration, time and effort. Figure 4 shows the high-level SingleStoreDB architecture.
Summary
All of this leads me back to the original reason for writing this article. Having been a developer, I have gone through many frustrations working with various technologies over the years. Part of our mission on the Developer Relations team at SingleStore is to help developers get up to speed on our technology. We want to make the journey using our technology as smooth, productive, and enjoyable as possible. And so, the desire to help others work effectively with a unique product was my primary reason to apply for a role at SingleStore.
Next Steps
Check out the resources on the new SingleStore Developer website. Sign-up for a free account, Select a Deployment Option and get US$500 in Free Credits. Work through the free Self-Paced Training material. Have questions? If so, check out the Forums.
If you’d like to contact me, please reach out using the social media links on my Personal website.
Acknowledgements
Many thanks to my colleague Linda Jackson at SingleStore for her comments and suggestions on an earlier draft version of this article.
