Tony Flath is an Edmonton, Canada-based tech influencer who specializes in various fields including cloud computing, cybersecurity, IoT and artificial intelligence to name a few. I had the opportunity to sit down with him and discuss the evolution of cloud technology which is progressively utilizing advanced AI frameworks.
Phil Siarri: Hi Tony, how are you? Can you tell our audience a little bit about your professional experience?
Tony Flath: Hi Phil, I’m doing great, thanks! I hope that you are doing well too. My professional experience has been an interesting one, from years of sales of complex solutions to 15 years of human resources management practitioner experience with a CHRP designation to 2016. Now working as a telecommunications and technology advisor with specialty in cloud and cybersecurity. As well over the last 5 years I’ve focused my social media efforts as “@TmanSpeaks” across social media where I am recognized as a technology influencer for cloud, cybersecurity, and artificial intelligence.
PS: What was your first introduction to cloud computing?
TF: Years back I worked with Ceridian Canada first as an HR product consultant and implemented complex HR systems with some pre-sales to account executives and solution consultants in the sales organization. I first learned on-premises and then was involved in the first HR SaaS before anyone even knew what cloud was. These years gave me great insight into devOps application development and cloud.
PS: Back then, did you think “the cloud” would become such a massive sector?
TF: Hell no! I certainly did see that shared infrastructure, platform, and application just made sense and devOps mentality meant you could build and innovate along the way. If anything this made me look at the world of work in this way and is certainly part of the reason I became a cloud disruptor.
PS: Can you describe what a multicloud strategy is and how it can be beneficial to a given organization?
TF: There are many clouds now, from public to private, to running on virtualized software “on-prem”. First there was the notion that hybrid cloud was required to manage cloud from one ‘single pane of glass’, one GUI to control all, this at first was a lot of hype without good control. Containers and Kubernetes have been showing great advancement for true multicloud control. Just look at IBM’s pending purchase of Red Hat and their release last year of the world’s first multicloud manager. Other public cloud providers like Amazon and Google have certainly showed more development and services for multicloud. What multicloud offers an organization is the ability to manage many workloads across many platforms with governance and devOps controls providing cost savings, increased controls, & application development enablement.
PS: According to you, what are promising, relatively untapped arenas for cloud providers in 2019 (and beyond)?
TF: IoT and edge computing are going to further cement multicloud with fog computing at the edge. IoT with rich Analytics and Big Data will continue to fuel machine learning and deep learning. Where I see huge opportunity is automation of code where artifical intelligence tools will know what data trends mean to customers and auto develop new code to better support application. There has already been evidence that AI neural networks are exceeding human capabilities. 5G will rock IoT with massive app development and at some point desktop and mobile will become one.
PS: Can you expand on as to what AI can bring to cloud computing?
TF: AI is already bringing some game to cloud computing. First, let’s look at cloud cybersecurity. If you look at leading edge SIEM solutions like LogRhythm and Splunk, these solutions use “threat intelligence” obtained through machine learning and artificial intelligence to automate and use predictive analytics to detect possible threats as they happen based on patterns and traits known to contribute to cybercrime efforts. Microsoft’s recent announcement of Azure Sentinel, a complete cloud automated SIEM with cybersecurity professionals constantly monitoring, detecting and isolating in literally seconds. Another thing that AI can bring to cloud is better orchestration and control of a mass number of workloads running across clouds — a multicloud that not only enables better management of various cloud platforms but automates and controls those platforms for efficiencies and cost saving. A good example of this is Densify, that offers automated cloud and container workload management which results in significant savings.
PS: What are important considerations when it comes to building and scaling AI-enabled cloud solutions?
TF: First and foremost, you need to have a good reason for why you are looking to cloud AI before just jumping into the “how am I going to do this?”… don’t just throw tech at it! Look at the economic outcomes from both cost savings and solution development enablement and then do your homework by looking at solutions and making sure to look at a proof of concept as many solutions are early days and you want to test and then scale deployment.
This interview was originally published on Nuadox.com on March 29, 2019 as “Tony Flath on the evolution from cloud to AI-enabled multicloud”. The text has been slightly shortened to improve clarity.