Over a year ago, I got my initial exposure to Google Cloud through a local meetup event, after which I shared my 1st impression on Google Cloud Platform on Medium. Since that I got a few new visits to that article every week. Last week (2018/09/26) I was so happy to join the very 1st GCP Summit hosted in Sydney (and likely the 1st on south hemisphere). It’s about the right time to reflect and share my fresher impression, again, on the Google Cloud Platform.
It all started with a moderate rain in the morning. Luckily by the time I arrived, the rain became fairly light. The event venue was at Carriageworks, a short walk from Redfern station.
Keynote
After going through the e-ticket scan, lanyard and swag bag collection process, I entered the main hall. After quickly going through a line of partner stands, I found a seat at the deep side of the iconic building to hear the already-started keynote presentation. A bit out of my expectation, the lady on stage was not sharing new features of GCP product family, instead, she was talking about some cool features of G-suite product suite e.g. Gmail Smart Compose, Google Sheet AI-enabled response to an average number question based on exported data from Salesforce.
Breakout sessions
After the keynote and a quick break, I went directly to Bay 25 to follow the “Accelerate Application Development” track, as most of my interested talks were on this track. There were another 3 tracks, covering distinct areas of interests i.e. “Modernize Infrastructure”, “Transform how Teams Work”, and “Create Intelligence from Data”. Interestingly, all rooms (or sections) are called “Bay”, because the venue are actually former railway carriages plus blacksmith workshops.
The 1st talk in Bay 25 was “CI/CD Pipelines in the Cloud”. Nothing too sexy in the talk partly because CI/CD has become the new norm for modern software organizations, also there are plenty of managed CI/CD offerings available in the market.
The follow-on talk in Bay 25 was “Building Serverless Application”, given by Chris Broadfoot. The interesting part of this talk was classifying Google App Engine (GAE) as one “less customizable” Serverless offering. As noted GAE had been around for 10 years time, proving itself as one of the longest-living GCP products. The fast and automatic spin-up of hundreds of GAE instances to support the constantly increasing request load in the demo was visually amazing!
Lunch time followed after the previous 2 breakout sessions. It’s probably not such a great experience being surrounded by loads of people, with an extra-long zip-zag queue in front of you. Some waitresses were trying to persuade people moving to an alternative catering location, or lined up into the right queues to get desired type of food options. I chose to walk back to the breakout space and spent some time handling business emails to avoid the crowd. After I went back to the catering half an hour later, there’s barely no queue, only a handful of people left which made my lunch efficient enough.
The first afternoon sessions in each Bay were reserved for GCP partners to share their experiences of Google Cloud adoption. I skipped it and rejoined Bay 25 for the 2nd session — “Using APIs to Deliver Connected Experiences and Grow Your Businesses”, presented by John Rethans, Head of Digital Transformation Strategy, Apigee. Apigee was acquired by Google in 2016. John also invited one of their local customers — Macquarie Bank onto stage to talk about API ecosystem and the crucial role API plays in modern Microservices architecture design.
The next talk in Bay 25 was my targeted talk as it focused on discussing Database options in GCP — “Not All Databases are Created Equal: How to Choose the Base Option for Your Needs”. The author Elizabeth Wu layout a “fleet” of database offerings in GCP, from Cloud Spanner (her favorite), Cloud SQL, Cloud Datastore, to the new Cloud MemoryStore. One can easily become puzzled why there are so many database choices from a single cloud provider, if without certain database background. I think Elizabeth did a great job of providing audiences the various factors you need to take into account while choosing database(s) for your application. The Firebase Realtime Database (RTDB) was actually interesting. Google even built another product called “what???”that combines the best of both Cloud Datastore and Firebase RTDB.
The last breakout session talk I joined was in Bay 17, “Connectivity Deep Dive and Best Practices”. It turned out to be of great technical depth, meeting my expectation of such a 400-level talk, except that I started feeling a bit struggle to follow in the middle of the talk, without the prior knowledge of e.g. differences between global and shared VPCs.
Food and beverage
Making attendees feel comfortable is crucial to achieve a successful tech conference. Food and beverage offerings were generally good, with enough abundance and quality. Coffee cups and water bottles were both reusable and recyclable, which was in alignment with Google’s public image of going green and environmental protection.
One-month free GCP courses
At the summit, it was well advertised people could take one-month Coursera GCP courses for free. I took one of the courses a few months ago with great experience, so I highly recommend you to go and have a check.
Final word
The 2nd impression I got from this summit on Google Cloud Platform was — capable and confident. The summit itself was neither too noisy (not so easy to achieve considering the amount of attendees) nor too quiet (when means the engagement was effective). The presentations about GCP product suites gave people confidence that GCP is well equipped with capabilities to make their service offerings better. The opening ecosystem built offline is the cornerstone that Google, as a company, leverages to enhance their advantages in the promising ML and AI areas, which are expected to be able to benefit people’s daily life in the coming future.
