The most valuable word in software is 'Finished'

We live in a world where the new "Shiny Thing" is announced every week. But the same fundamental problems exist. The need for speed, scale and accuracy became apparent at the dawn of the internet. They still matter now, and there's a never ending race between our ability to meet these requirements and the ever increasing demands expected of us. In such as situation having access to decades of experience is useful, if you want to actually get to 'Finished'
When you face real challenges, you need to engage a real expert, which is where I come in.

For my full work history, see my LinkedIn Profile. In summary:

• 35 years industry experience.
• Expertise in RDBMS work, OLTP systems, Big Data and Streaming.
• 4 patents on conflict management for geo-replicated database applications.

I have been a DBA, manager of DBAs, DB architect, support manager, troubleshooter, consultant, Java developer (entire application), MIS developer, big data developer, OLTP system tuner, benchmark runner, demo application developer, sales engineer, product marketer, stand in sales person, author of blogs and white papers, webinar host and conference speaker.

Consulting Services and areas of expertise 

OLTP Systems

OLTP has become much more complicated over the last decade, as we've moved on from human users and timescales, and now live in a world where 90% of all requests are by APIs, devices or Microservices. I have experience with production systems running at up to 750,000 transactions per second.

Big Data Processing

Stream Processing

Kafka has replaced older technologies such as SCP and SFTP, but in the process has created a requirement to process streaming events with 100% reliability within a short time of their creation.  In addition, the IoT is massively increasing data volumes.   

Database Technology Selection

Picking the right database technology can be a huge challenge. Every vendor promises the sun, moon and stars. But the actual problem is that it's surprisingly hard to tell up front what your  requirements are, and you tend to caught on the horns of a dilemma -  spend too much time evaluating and you get stuck in "analysis paralysis", too short a time and you might pick the wrong thing. 

Benchmarking & TCO measurement

Sadly, the most common outcome of the first benchmark run you do is the realization that you should have done it differently!  Establishing real world costs and limits is much more complicated than spawning a cluster, throwing traffic at it, and looking at your hyperscaler bill.

Active-Active Geo-Replication

Active-Active Geo-Replication is the "Bond Villain" requirement for databases - everybody talks about it, and wants it, but you'd usually have to be mad to even try! While implementing it is a lot easier than it used to be, it's still a major challenge. I have 4 patents in this area, and am happy to give advice. 

LinkedIn - Career history

A Vast Oasis Of Useless Information - Personal blog

Github - Various bits of code I've done