Day 2 of Google Next conference started on a high-note. The keynote was very well executed and is markedly different than Day 1 keynote.
The keynote started with reinforcing Googles investment in network, infrastructure and operations. This is to be expected and is an important point to reinforce. The first product they discussed was Cloud spanner, which is one of the most exciting development in DB space in a while now. Its a good choice to use for production grade large scale applications. The pricing makes me concerned about its fit for dev/test and smaller scale applications. Its a problem GCP team has to figure out the best way to solve.
The second announcement was around their pricing. This had me literally jumping up and down with excitement. In addition to their sustained use discounts, they had announced committed use discounts. Unlike other cloud vendors, GCP has decided to apply thinking to pricing problem and have nailed it. With committed use, you don’t buy by instance types, but buy no of cores and memory and use them as you see fit. This is the most sensible way to rent instances. Google has separated themselves from rest of the cloud providers when it comes to pricing. I also think its irresponsible for a startup to not seriously consider GCP if they are doing only VM based compute.
The next announcements were around Identity Aware Proxy and API gateway. Google struggles a lot with Identity and it showed in the demo with the person having to resort to incognito windows. Google has a lot to learn from Microsoft when it comes to solving Identity problem.
They had some announcements around Microsoft environments, which were ok. Their data prep, Bigquery etc were decent as well. Security related announcements like DLP were decent, but needs to have an ecosystem around it to be of use.
One disappointment from the keynote is Google still do not understand what serverless means or its potential. They kept saying App Engine is the original serverless, which is bit of hogwash. This is an area where AWS will demolish GCP going into the future. GCP has to understand this if they want to play in the big leagues in future.
In the short term , GCP has positioned themselves to be a worthy contender. For VM based compute clouds, I would not be surprised if GCP starts beating AWS and become first choice for many startups.
Its good progress from where they were last year.
Thats all folks.