TWiGCP “Firebase Summit, Node.js 10 in AppEngine, and 7TB of persistent NVRAM”
Here are the main announcements for this past week :
“What’s new at Firebase Summit 2018” (firebase.googleblog.com). Formal support coming in beta, new Firebase Management API, improved ML Kit, a new local Cloud Firestore emulator, a Data Studio template, and more! Oh, and the team also celebrated 1.5 million monthly-active Firebase apps!
“Node.js 10 available for App Engine, in lockstep with Long Term Support” (Google blog). Where idiomatic means understanding developers including their software cadence.
“Available first on Google Cloud: Intel Optane DC Persistent Memory” (Google Cloud blog). That’s up to 7 TB of Persistent NVRAM!
“Bringing enterprise network security controls to your Kubernetes clusters on GKE” (Google blog). Private clusters, Master authorized networks, and Shared VPC.
“Kubeflow 0.3 Simplifies Setup & Improves ML Development” (kubeflow.org). New deployment tool kfctl, hyperparameter tuning, PyTorch operator parity with TensorFlow, more TFX libraries, new examples with XGBoost, GPU inference, etc.
From my still favorite “Customers are best to talk about GCP” department :
- Optiva Inc. Case Study (cloud.google.com) #CloudSpanner
- How Streak built a graph database on Cloud Spanner to wrangle billions of emails (Google blog)
From the “How-to” department :
- How 20th Century Fox uses ML to predict a movie audience (Google blog)
- Cutting costs with Google Kubernetes Engine: using the cluster autoscaler and Preemptible VMs (Google blog)
- Exploring container security: running and connecting to HashiCorp Vault on Kubernetes (Google blog)
- Serverless from the ground up: Connecting Cloud Functions with a database (Part 3) (Google blog)
- Run Apache Spark and Apache Hadoop workloads with the flexibility and predictability of Cloud Dataproc (Google blog)
- Integrating Google Cloud Build with JFrog Artifactory (Google blog)
- Classifying congressional bills with machine learning (medium.com)
- OpenCensus and Firestore Native (medium.com)
From the “will gRPC-Web change your perspective on REST?” department :
- Envoy and gRPC-Web: a fresh new alternative to REST (blog.envoyproxy.io)
From the “there is still so much to be invented” department :
- TensorFlow code and pre-trained models for BERT (github.com)
- Introducing AdaNet: Fast and Flexible AutoML with Learning Guarantees (ai.googleblog.com)
From the “what’s a TWiGCP without BigQuery?” department :
- Python PyPI stats in BigQuery: Reclustered (towardsdatascience.com)
- Modern data warehousing with BigQuery: a Q&A with Engineering Director Jordan Tigani (Google blog)
From the “we need your input” department :
- Michael McDonald asks “what would you use a Cloud Functions API Gateway for?” (forms.google.com) Chime in !
From the “the more you know” department :
- Getting to know the Google Cloud Healthcare API: Part 1 (Google blog)
- Container Runtimes Part 3: High-Level Runtimes (ianlewis.org)
From the “Beta, GA, or what?” department :
- [GA] Cloud SDK 223.0.0
- [GA] Stackdriver Trace consumption-based pricing
- [Beta] local emulator for Cloud Firestore
- [Beta] Video Intelligence API — Object tracking, Text detection (OCR)
- [2.8] Apache Beam 2.8.0
- [0.3] Kubeflow 0.3
From the “all things multimedia” department :
- Kubernetes Tutorial: How to Failover MySQL on Google Kubernetes Engine (GKE) (youtube.com)
- Getting Started with TensorFlow.js (youtube.com)
- Skaffold — Local Kubernetes Development Made Easy (youtube.com)
- All the videos from the Firebase Summit 2018 (youtube.com)
- GCP Podcast #153 — Bazel with Tony Aiuto (gcppodcast.com)
- Kubernetes Podcast #27 — Evolution of the Kubernetes Community, with Sarah Novotny (kubernetespodcast.com)
- Google Doodle — Halloween 2018 (google.com) #GCP #OpenMatch
This week’s picture is the 12th second of the Logan trailer and is taken from the “How 20th Century Fox uses ML to predict a movie audience” post.
That is all for this week!