What’s new with Google Cloud Platform this month? We get a Google Cloud certification for database engineers, in beta, and new product, BigLake, helping manage data lakes and data warehouses across GCP, AWS, and Azure. We also take a look at updates to Cloud Spanner, GCE (Google Compute Engine) VM Suspending, Media CDN, and upcoming Google Summits for security and machine learning. Let’s go!
Your keys to a better career
Get started with ACG today to transform your career with courses and real hands-on labs in AWS, Microsoft Azure, Google Cloud, and beyond.
Professional Cloud Database Engineer BETA
If you work with structured data in Google Cloud Platform, or if you want to, then you might be interested in Google’s new Professional Cloud Database Engineer certification, now in beta.
As Google describes it, a person with this qualification “designs, creates, manages, and troubleshoots Google Cloud databases used by applications to store and retrieve data”. And they “should be comfortable translating business and technical requirements into scalable and cost-effective database solutions”.
The beta exam costs less now than it will when it becomes generally available, and you might get some exclusive swag for participating. The tradeoff is that it’s a longer exam (four hours!) and that you won’t get your results for maybe a couple of months.
GCE VM Suspending now GA
If you’re still using VMs directly, instead of higher-level services, then this latest piece of GCP news might interest you: you can now suspend and resume Compute Engine instances through the ACPI S3 signal! When instances are in the suspended state, you can pay less for premium OS licenses, and you don’t pay at all for cores and RAM. You will still have to pay storage costs for the instance memory though.
One of the use cases for this is to pre-warm some slow-starting instances so you can use them later to scale out. Google claims that their way has some advantages over how this works on other clouds, but they might also be forgetting that some other clouds have let you do this since 2018. Oops!
Upcoming Google Cloud Platform online summits for security and applied ML
Mark your calendars, because Google is running a Security Summit on May 17th and an Applied ML Summit on June 9th, both online. We are all responsible for security, and machine learning is only good to us if we can apply it effectively to solve problems, so these two days could offer you valuable insights and help you make meaningful progress.
Improved Speech-to-Text (STT) AI
Google Cloud Platform is already pretty awesome at converting speech to text through their STT API, but it has just gotten even better. The latest models are available to us under some new tags. The “latest short” tag is for models that are tuned to short phrases like voice commands, while “latest long” is for longer-form speech, such as the video version of this blog! Did we use it to write this post? You’ll never know!
Media CDN generally available
Given that YouTube is pretty decent at handling video at scale, it’s nice to see that Google’s new Media CDN service, which is already generally available, will let us take advantage of the same technology YouTube uses to deliver video globally with very low latency. Without any effort on your part, it supports HTTP/3, TLS 1.3, and BBR congestion control.
Cloud Spanner Change Streams coming soon
Cloud Spanner is Google’s global and infinitely scalable relational database, but it’s not the only place you’ll ever want to have your data. That’s why the upcoming Cloud Spanner Change Streams functionality will be so interesting.
To start with, you can use this to replicate Cloud Spanner data to all manner of other storage, such as BigQuery or Cloud Storage. You can also process the impact of data changes on dependent systems through something like Dataflow, or trigger whatever you want with PubSub. If you only care about some changes, you can even filter this down by table or column.
Thankfully, this has no need for any extra infrastructure – you just turn it on and your existing Cloud Spanner instance does what you need!
New BigLake product unifies data lakes and warehouses across clouds
Recently announced, BigLake is Google’s new product for managing data lakes and warehouses across GCP, AWS, and Azure. Google recognizes that organizations might have lots of data on both Azure and AWS, in addition to their GCP-based data, that they’d like to manage and leverage consistently in one place.
So data might get stored on S3, or GCS, or Azure Blob Storage in open formats like Parquet or ORC, and then developers and data analysts only have to deal with BigLake. By the way this is more secure because least privilege doesn’t include any of the raw data. Then BigLake will farm those queries out to BigQuery on GCP, and BigQuery Omni on AWS and Azure.
Get more GCP news goodness in you!
Looking to learn more about cloud and GCP? Check out our rotating line-up of free courses, which are updated every month. (There’s no credit card required!)