Insights and Updates

Read the HarperDB teams’ thoughts on database technology topics such as SQL, NOSQL, Edge Computing, IoT, Industry 4.0, and more

Part 1: Our HarperDB HTAP Database Origin Story

Posted by Kaylan Stock on March 21, 2018
Kaylan Stock
Find me on:
Harperdb6 origins blog part1.jpg
After finishing college, Fred thought he knew his career path.  The trajectory seemed clear but after several years working in medical research,  Fred chose to pursue a different avenue. This change was the catalyst for events that later on would help define and champion the team at HarperDB. 
Fred first discovered his interest in the big data landscape while working in research in the Department of Infectious Diseases at the Cleveland Clinic Foundation,  presenting extensive data sets and analytics around his research.  After working in the medical field, Fred decided to explore other interests which ultimately led him to a sales executive role in software technology. He sold computer aided design and manufacturing technologies that also encompassed product data management and product life-cycle management tools.  Through this work he started to not only view big data as interesting but also gained a deep understanding for the importance of big data in a business landscape. 
 He started as one of the first employees of Phizzle Inc. in 2006 where the focus was consumer engagement through SMS campaigns.  One day in 2011, by now he had moved to Colorado, Fred was flying home from the Phizzle Corporate office in San Francisco. 
 Stephen and I were both at the ticket counter [at the San Francisco Airport] and Stephen was checking his golf travel bag which was this huge bulky metal contraption that he used…and Stephen just looked very disheveled and out of sorts….” 
 Come to find out, that same disheveled man was seated next to Fred on his flight in the very same row. They sparked a conversation, where Fred officially met Stephen Goldberg.  Stephen was at the time doing Salesforce implementations for Fortune 500 companies, but his interest in computers can be traced back to middle school. He spent a lot of time in the library with other kids like himself, doing stupid” things like plugging power supplies into each other and exploding computers. Back in the early 1990’s there wasn’t a computer staff at the school, so Stephen and his friends were allowed to skip class and get extra credit for being the unofficial computer help desk. This was his first experience working with computers other than for fun.
 As his interest flourished, his father noticed a raw talent emerging, so he sent Stephen out to California to work with his Uncle at HurlNet in the summer of 97 when he was 13. HurlNet provided the first ISP billing software. Stephen had the opportunity to work alongside seasoned developers and learn to program. He was learning everything from Linux to HTML and beyond. That summer with his uncle really opened the door to coding for Stephen. He continued his advancement in tech when he did an internship with Sequoia Software at 16 years old, where he helped prototype a dynamic app to that was one of the first uses of XML.
 When Stephen entered into college, he had decided that he didn’t want to study computer science, out of fear of being a dork”, so he ignored his natural and acquired talents and studied Theology with a plan to attend law school. After graduation, he struggled to find work and decided that graduate school wasn’t for him, landing him back in the computer software world once more. He bounced around for several years at a few different software companies including Red Hat. 
In 2009 Stephen had moved to Colorado and was working for Statera as a consultant. He was assigned a project for Core Site,  to do a Salesforce integration. Once the integration was complete he was asked to hand over the project to one of the in house developers. That developer was Kyle Bernhardy. After a month long stint of being holed up in a room together, Kyle and Stephen both realized one thing. That they wanted to work together for the rest of their careers. 
As Stephen put it,   [Kyle and I] really liked each other, Kyle was an unbelievable architect [and] unstoppable when solving a problem….[there] have been multiple times I wanted to quit throughout the years but Kyle always pushes me to a solution.”
Kyle, on the other hand, really appreciated Stephen’s broader view point of what technology could be. Over the years, Kyle considered himself a lone wolf having worked for a lot of small companies, in one case being the only in house developer.  He became self reliant and found he didn’t enjoy working with any other developers until he met Stephen. 
“[Stephen] was one of the first people I met in my career where, we just meshed at a level of how we think, how we approach solutions, but also…we don’t have the exact same competencies..” 
Where Kyle is a bit more technical, Stephen has that broader vision and ability to see the landscape as a whole. Over the years they have learned and grown together and found that they bring a great balance to each other. 
 Fast forward to Fred and Stephen on their flight back to Denver in 2011. Stephen was flying back from giving a talk at Salesforce Dreamforce,  he had just left Statera to start his own consulting firm, KloudRoot. Stephen took an interest in what Phizzle was doing around consumer engagement. After hitting it off on their flight, Fred knew he wanted to work with Stephen further, so Phizzle hired KloudRoot as consultants. 
 In the beginning of 2012, Stephen brought Kyle in to join him at Kloudroot, where they worked on different Salesforce integrations. As they settled into their new routines, Stephen and Kyle realized they needed help with complex operating systems like configuration and system admin based tasks. They started interviewing potential  system admins to join the team. 
Enter Zach Fowler.  He came in profusely sweating, in a ridiculous corduroy jacket, corduroy shirt and a wool tie completely terrified and nervous.  When Stephen first saw him he wasn’t sure what to think. 
 We asked him some questions and his answers were mind blowing, I couldn’t tell if he was right or not, just listening to him was amazing. I knew that he was a genius  and that he had a very different perspective on life and technology  and that perspective for the last 6 years really helped Kyle and I to do things in a very different way because…Zach always really thinks about technology at a incredibly deep level as well as an incredibly futuristic point of view….”
KloudRoot continued to provide consulting services for Phizzle for about a year and when Phizzle decided they needed to revamp their platform and bring the development team in house they made a strategic move in December of 2012 to merge KloudRoot into Phizzle. That merger included Stephen, Zach, and Kyle. Fred knew he wanted to work with them specifically.
“I wanted to work with [Stephen, Zach, and Kyle] because I saw a very cohesive team and extremely intelligent team, one that, you could tell, really respected each other and genuinely liked each other…..” 
The Kloudroot team had years of combined experience and they really understood a vast amount of technologies, and for Fred it was refreshing to add that visionary and technical outlook to Phizzle.  For the next four years, Stephen led the engineering team at Phizzle, they had many successes and learned a lot about big data, databases, and middleware.  Then in early 2016 Kyle, Zach, and Stephen were out in Palo Alto for a project for Phizzle.  After another long exhausting day they were hanging out in their tiny Airbnb and they started talking.
Stephen described how the concept for HarperDB started to form, We were creating this thing at Phizzle…it wasn’t related to Phizzle’s core business [of consumer engagement] but we were trying to solve some technology problems deeper in the stack around big data and to run Phizzle’s business we needed all of this big data architecture and it was just getting more complicated… trying to improve the [database] solutions around that….trying to make them run better and we were trying to make middleware products to sync all this data together [and] make [the data] scale, we kept fine tuning the system and we [finally] looked at each other and said….fundamentally we are just putting a band aide on the solution, we aren’t really solving the problem…” 
 That night in Palo Alto, the guys realized, what if it didn’t have to be this way. The database industry today is so fragmented with multiple stores and things are overly specialized. They decided to zoom out batting around the concept of one  single database that could do the transactional rates of a noSQL database but also have the reporting capabilities of a SQL data store. They spent that night up kicking the idea back and forth, a centralized datastore, something simplistic and modern, creating a small footprint that even a junior developer could spin up in minutes. 
Kyle described what went through his head that night, In those moments it was very freeing to just talk about and….in that…to me it was like, oh we are just kicking an idea around but then after a couple hours of us getting into it, we had this moment of looking at each other across this kitchen table and just looking each other in the eyes and being like this is a great idea!” 
The guys would fill Fred in on their idea when they traveled home to Denver a few days later, but then would sit on it for over a year before HarperDB was founded. 

To be continued....

Topics: Our Story