CFD20: Juniper Networks: Digging into Details and Demos of Apstra and AI Lab for AI Data Centers

CFD20: Juniper Networks: Digging into Details and Demos of Apstra and AI Lab for AI Data Centers

This quick blog contains live blog notes of the next 3 presentations from Cloud Field Day 20. AI JVD’s (Juniper Validated Designs): build without all the research and details.

Part 1: Your Private AI Data Center, as Easy as Cloud (Nick Davey)?

The JVD’s start with documentation etc., but they can use Apstra to lab test then instantiate the actual datacenter. ?

Note: you can save versions off to GitHub or TerraForm Cloud. The latter is used to automate the automation, Juniper guided design: reduce the effort, automate Apstra workflows without API coding.

No hardware is needed (until you push “build“): Apstra uses virtual switches!

Juniper integrations “on top of” Apstra: Terraform, servicenow. Can make them talk to Apstra but also to each other. Order network infrastructure from high-level interface.

The Terraform providers for the Juniper AI JVD’s are posted on GitHub, etc. You can post to them to your Apstra instance without requiring hardware.

See the video for links for that plus other Terraform code.

Part 2: Demo How This Works

Watch the video when it's posted – screen captures can’t do it justice!

(So I’m not saying “RTFM” but “WTFV” – Watch The Fine Video!)?

Part 3: Design, Deploy, and Operate AI Clusters like a Pro (Jay Wilson)

Jay is a customer field team person. How to monitor and manage the environment.

Apstra does datacenters but is focused on fabrics, and can manage many fabrics and datacenters.

Jay reviewed the lab design diagram: GPU’s in middle, bottom. Top and right was storage. Left was management network.

“Think like a hyperscaler” with Apstra. It implements and validates INTENT. Apstra is the Source of Truth. Telemetry is also important for this. Apstra is used as used: no modification. ?

Jay then shared a QR code. And segued into a recorded demo (to avoid Murphy’s Law): Tailoring, Operating, and Developing and Instance.

Terse notes from that:

·????? Apstra GUI: 3 fabrics: Backend GPU, Storage, Frontend Management Fabric.

·????? This had customized telemetry added to it (since the OS in box doesn’t yet have Streaming gRPCs).

·????? A second aspect is that Apstra does fabrics but doesn’t adjust all the “nerd knobs”: so they added 41 configlets for focussed use situations.

·????? Custom dashboard for ECN’s, PFC’s, drops to monitor drops of interest. Across 152 interfaces.

·????? Widget: by leaf by interface reporting ECN’s, PFC’s. To understand where the congestion is, who is being told to back off.

·????? Real-time heatmap!

·????? Apstra tidbit: if there is an anomaly: you can’t acknowledge and hide it, you fix it or it continues to show up.

Discussion then added ?a lot of good Q&A and details: watch the recording!

#CFD20 #CloudFieldDay #JuniperNetworks #PeterWelcher #CCIE1773 #BlueAlly

?

要查看或添加评论,请登录

Peter Welcher的更多文章

  • Introduction to Microsegmentation

    Introduction to Microsegmentation

    This blog begins an introductory series of moderately long blogs, covering key aspects of Microsegmentation and Zero…

    3 条评论
  • Pete’s Take: Catchpoint at Cloud Field Day 22

    Pete’s Take: Catchpoint at Cloud Field Day 22

    Tech Field Day always produces such great technical content! However, it can be a challenge keeping up with it due to…

  • AI Ate My Blog on RoCEv2

    AI Ate My Blog on RoCEv2

    I acknowledge I’ve been a blog technology summarizer for quite a while. It served to help me broaden/solidify my skills…

  • AI Datacenter Switch Math

    AI Datacenter Switch Math

    Author: Pete Welcher, Coauthor: Brad Gregory This is blog #3 in a small series about Networking for AI Datacenters…

  • AI Requirements for Datacenter Networking

    AI Requirements for Datacenter Networking

    Author: Pete Welcher. Coauthor: Brad Gregory.

  • Quick Takes #2, February 2025

    Quick Takes #2, February 2025

    I’m working on some longer blogs that I hope to be able post in the next week or two. In the meantime, lots of exciting…

  • Quick Takes: February 2025

    Quick Takes: February 2025

    I’ve got some longer technical blogs in the works. For this week, it’s time again for some of my “Quick Takes”:…

  • Pete’s Take: Pain Points in Networking and IT

    Pete’s Take: Pain Points in Networking and IT

    It’s a new year, so time to look at how Networking and IT have been evolving. Ignoring the AI elephant in the room.

    1 条评论
  • Pete’s Take: Pondering NetOps/AIOps Strategy

    Pete’s Take: Pondering NetOps/AIOps Strategy

    What’s new in NetOps, including AIOps, and where are things heading? Some thoughts ..

    1 条评论
  • Pete's Take: AI/ML and Error

    Pete's Take: AI/ML and Error

    Artificial Intelligence (AI) has certainly received a lot of press lately. And achieved new levels of hype.

社区洞察

其他会员也浏览了