Here is another dev-ops overview of the different types of services mesh systems, their designs, strengths, weaknesses and approaches. A few of the more-popular meshes are missing such as TwistLock. And my concerns about simplicity and performance are not well-addressed. But it's a good starting point to learn the basics. As a beta user of istio, I still recommend it; for example you can have some pods ignore the mesh but log to istio's prometheus to align all events and timestamps.
Rob Wilson (Roobert) takes us on a tour of his cluster scaler that enables rapid, dynamic creation, destruction, and scaling of development environments for large development organizations that want to be more efficient with their cloud spend. Cloud functions and Terraform are used. Robert's approach is very simple and clever.
Rick rants about the ten things hie hates most about PostgreSQL while admitting it is the best storage system you can use by default. It's a great in-depth look at the issues you will face if you push this database past its limits as Rick has done.
("Atomic Deployment" aka flip a symbolic link as we did in the 1990s)
Slack has written an overview of how they deploy the slack service we all use at least 12 times each day. It's surprising more teams do not have a few dozen versions of service software on their production servers and symbolic links to the current running version as we had in 1994.
Check out the finalists list. If you are looking for good science fiction to read during the lock-down consider the stories on that list. Also, the conference will be webcast but only to registered attendees. Lois Bujold will be the grand master.
Take advantage of deserted roads. Send out teams of two to four workers - trained in hygiene and safety - to fix potholes, repair empty schools. Get furloughed people to work. Fix infrastructure; stimulate the economy.