Author: Jeremy Morgan
Hello Cloud Gurus! We are coming to you from re:Invent 2022 in Las Vegas!
It is currently Tuesday afternoon, as we’re filming this, so we’ve already had two keynotes, so far. Last night was Monday Night Live, with Peter Desantis. And this morning we heard from AWS CEO Adam Selipsky.
Now, this special episode is to get you caught up on the top announcements from the last two days. But we’ve also already done some Livestreams and YouTube Shorts where we both gave our personal highlights and dug into the announcements a bit more.
But let’s dive into what we’ve heard so far! Let’s start with a really fun one! AWS has been working very hard, behind the scenes to obliterate Lambda cold starts. And they have succeeded! Check out their blog post on SnapStart for more info, but the quick version is that they’ve used a combination of smaller Firecracker VMs, and snapshotting post-initialization functions, plus really efficient snapshot transfers. And that has resulted in cold starts that Peter Desantis claimed are virtually indistinguishable from warm starts! Which means that Lambda is absolutely perfect for spiky workloads!
The new Amazon DataZone lets you share, search and discover data at scale. You can now get insights and access to data no matter where it’s stored. You can view and manage data across organizational boundaries, to enable fast collaboration with other teams.
Amazon Security Lake is a new service—currently in preview—to pull together all your security-related data into a Data Lake. It grabs info from AWS services like CloudTrail, VPC FlowLogs, GuardDuty, and Inspector. But it also connects to on-premises data through Security Partners.
Spark developers who work in Sagemaker, AWS Glue, and similar services use 3rd party connectors to work with Redshift. Today AWS announced full Apache Spark integration with Redshift. Developers can now build Apache spark applications for Redshift, with ease.
A big announcement, Opensearch is now serverless!! Well, to be clear, we now have, quote, “the preview release of a new serverless option for Amazon OpenSearch Service”. So you can use OpenSearch without managing cluster sizes or instance types or any of that jazz. Much better!
ML Powered forecasting with Q is a new natural language querying for analyzing business data. Users can now view business performance forecasts quickly and easily. It allows folks who aren’t analysts or data scientists to grab the information they need.
With the new SimSpace Weaver service, you can run simulations at scale using EC2 instances. Operations planners often run simulations for things like responses to a disaster, traffic control during an event, and many other applications. Traditionally these simulations were run on a single machine, but SimSpace Weaver can run on multiple EC2 instances and scaled up quickly.
Guard Duty will soon offer runtime threat detection at the container level. While Guard Duty already protects EKS, it will now go deeper into the container level to ensure your applications run as securely as possible.
There are tons of new EC2 instance types available such as Gravitron ones but we’re not going to read them all out! You might want to go and check out the AWS posts if you were hoping for a specific one.
Well, alright! That was a bunch of great stuff, but re:Invent 2022 is by no means over, yet! Stay tuned right here as we continue our coverage.
And, as always…
Keep Being Awesome, Cloud Gurus!