How has the recent turmoil within the OpenAI offices changed your plans to use GPT in a business process or product in 2024?
Increased uncertainty means we are more likely to evaluate alternative AI chatbots and LLMs.
No change in plans, though we will keep an eye on the situation.
With Sam Altman back in charge, we are more likely to go all-in with GPT and LLMs.
What recent turmoil?

Microsoft Azure Container Instances are Now Ready for Production Workloads

Apr 25th, 2018 1:10pm by
Featued image for: Microsoft Azure Container Instances are Now Ready for Production Workloads

Previewed last year, Microsoft’s Azure Container Instances (ACI) is now ready for production usage, according to the company.

Microsoft touts ACI as a “serverless” service, one that offers the container, either of the Linux or Windows variety, as a basic compute primitive, much like a virtual machine (VM) is a primitive. With these “serverless containers,” the end user does not have to worry about provisioning or managing the underlying the infrastructure, explained Microsoft Principal Program Manager Gabe Monroy.  Amazon Web Services launched a similar service later last year, Fargate.

In the initial ACI preview, Microsoft had found that customers were using the services for batch processing, continuous integration, and event-driven computing, Monroy said. The per-second billing model also lent itself well to burst workloads. To this end, Microsoft lowered the pricing and eliminated the fee for initializing a container. Current prices are:

New Price (US West) Old Price (Global)
Create fee None $0.0025 per Instance created
vCPU per second $0.000012 $0.00001250
Memory (GB) per second $0.000004 $0.00001250

The company is hoping that the lower prices will attract more long-running workloads, such as running web servers. A single container running on a single vCPU and 1GB of memory would cost approximately $41.47 per (30 day) month.

Microsoft promises an uptime service level agreement of 99.9 percent for any container group. Each container is secured and isolated through a VM hypervisor. The service is across 6 regions: West US, West US 2, East US, West Europe, North Europe, and Southeast Asia.

The company also has added some new features with this general availability (GA) release, including

  • The addition of the Execute (EXEC) command to provide an interactive shell for containers.
  • The ability to Monitor CPU and memory utilization of Linux containers.
  • The ability to give each container a reliable endpoint address through a DNS name label.
  • The Azure portal has been upgraded with metric alerts and the ability view container logs/events.
  • Container state and log output can be read through the Azure CLI 2.0.
  • A set of container restart policies.
  • Multiple volume types can be used for mounting to Linux containers, including Azure FilesgitRepoemptyDir, and secret volumes.

For those using the Kubernetes, ACI can be controlled by the open source container orchestration engine, thanks to the ACI Connector for Kubernetes. “The ACI Connector allows you to bridge the world of ACI containers with the world of Kubernetes. You can drive all that stuff through the portable API Connector,” Monroy said.

And the Virtual Kubelet released in December, allows ACI containers to easily hook into other Azure resources. “It shows up as a Node in Kubernetes. It’s not really a node, it is a bridge to a service. But you can schedule to it like any other node,” Monroy said.

Microsoft is a sponsor of The New Stack.

Group Created with Sketch.
TNS owner Insight Partners is an investor in: The New Stack.
THE NEW STACK UPDATE A newsletter digest of the week’s most important stories & analyses.