This past week I was fortunate to attend the 2019 NewYork Serverlessconf — a 3-day long event with speakers, workshops and a hackathon focused on cloud-based technologies. Serverlessconf started a fears ago, the earliest one listed (https://serverlessconf.io/) was in 2016, and began in New York, with a second and third held later that year in Tokyo and London respectively. In subsequent years, Serverlessconf has also been held in Austin, Paris, and San Francisco. On the website, Serverlessconf is described as “a community led conference focused on sharing experiences building applications using serverless architectures. Serverless architectures enable developers to express their creativity and focus on user needs instead of spending time managing infrastructure and servers.”
For my part, I was only attending the conference portion, as that’s what my ticket was for, but also probably for the best as I’m just dipping my toes in serverless tech — so it was a great opportunity to learn more about ~the cloud~ without the pressure of being a newcomer that might exist in a workshop or hackathon. Although, I did hear in one presentation that someone with no prior experience did the hackathon last year and had a marvelous time, and considering what a warm, welcoming, and fun environment it was, I wouldn’t be surprised if there were more in the same boat. Well, hindsight is 20/20, and since it was such a great conference — I’m definitely planning to attend the hackathon and workshop portions next year.
There were a number of impressive speakers on the docket, and as with any conference, it’s simply impossible to see them all. As I was only able to attend the second day of the conference due to a prior commitment, I wanted to take advantage as much as possible. So, I got there bright and early, and did my best to be a sponge.
All the speakers had great energy, and there were a wide variety of topics. There were a wide range of topics covered, with material ranging from novice to supreme overlord, but it was structured so that there were always at least two talks going on at once. Each talk was roughly 30 minutes give or take, and having worked festivals and events before, I was incredibly impressed by the flow and management of the day’s agenda — they really did a great job with timing, so kudos to the organizers, because I know how difficult it is to execute an operation of that magnitude with so many things going on, and to make it seem seamless!
I started my day with “Serverless: An ops experience or a programming model?” from Donna Malayeri, a product manager with Google Cloud Platform. With the discussion of what serverless is at it’s core with relation to the original launch model and the current evolved state and a look at serverless modularization, It was a great intro. Although I hadn’t yet started taking notes (regrets!) it got me excited for the day,
Next I listened to Sam Kroonenburg, the founder of A Cloud Guru (a key partner of Serverlessconf) discuss “Serverless: From one function to 43 Microservices”. Sam founded A Cloud guru in 2015 and explored the reasons he decided on building a Serverless platform, and walked through each phase of his journey — from starting with himself as the only developer, to a team of 5, to multiple teams of multiple people, how the company grew and changed, implications of certain decisions, and what’s on the horizon.
Afterward Capital One’s Jonathan Altman gave a talk “Serverless Lessons Learned and Our Best Practices”, which served more or less as an overview/summation of other different things that had been discussed at the conference or currently relevant goings-on in the serverless community.
Following that, I sat in on “Detecting outages at scale” from Sander van de Graaf of Ookla / Downdetector, who detailed the process of outages and his team’s process at downdetector to scale their services with the latest Facebook outage.
I was a little late but caught the latter two thirds of Farrah Campbell and Danielle Heberling from Stackery illustrating their journey to learn Serverless, what they’ve learned and how they’ve grown in the Serverless environment and community, what they’ve created together and how they’ve leveraged their skill sets to accentuate their working partnership.
I was super stoked to attend the “Serverless State” talk by Tim Wagner, inventor of AWS Lambda, who showed the possibilities of serverless technology and how to tackle areas that still need work to enhance applications and increase serverless ability. He noted that 15 years ago when he was first discussing the possibilities of serverless to a high level sales college at Amazon, he was asked “what can’t this serverless technology do?”, to which he responded “Well, probably never will this be used for video transcoding” before launching into Berkeley’s creation of one of the best video transcoders available right now…created with serverless technology. He ended his talk unveiling Serverless peer-to-peer networking for Lambda with Reliable communication between Lambda functions before smashing an actual server with an axe. It was a great talk all around and a great reminder that limitations are never concrete.
H^2 