NFD23 - Juniper
The path is clear, weel defined and execution follows the overall vision of an AI-driven enterprise that monitors and measure end-user experience (or is it the frame/packet experience?).
The hardest part of the troubleshooting process is usually to recreate the problem. We can’t fix what is not broken.
Mist can identify problems and automagically collect all the network states around the issue, including packet captures. Even more, the Mist assistant, Marvis, can suggest remediation actions and help the root cause analysis. Auto remediation will be the next natural step, I assume it will not take much time to see it in deploy in production with confidence, at least for the most common issues.
Cloud management is a key factor to consider for any company that has more than one site.
The Mist cloud can manage AP and switches and leverage Marvis assitance on all of them. This is nothing new compared to other vendors and became madatory to manage distributed networks.
During Calvin Remsburg ’s end-to-end automation demo he showed us how different tools can be integrated to create a complete solution.
The amount of work in the background to create a seamless process was not quantified, an automated workflow can definitely have a huge impact on the quality and effectiveness of the work of an Operations team, it should well worth the initial investment.
This is once again an example of how companies should consider developing their own automation workflow integrating open-source projects and vendor tools, investing in the internal skills of the teams.
Ansible, Netbox, Git and others, a mix of the best projects all working with Mist.
The full demo recording is available:
My overall impression of Juniper’s presentations was very positive. I didn’t felt the smell of overpromising typical of marketing pitches. The organic integration and evolution of Mist on top of existing Juniper products seem very coherent.
Juniper has a clear view of the roadmap and, very important, execution follows the vision.
On the other hand, every nice story has to face the reality: all these wonderful features only work with Juniper products, leading to the usual dilemma of vendor lock-in.
In my opinion, a vendor that can provide a well-integrated product line, with reasonable pricing and don’t try to sell more and more servers to manage the latest features, can be a reasonable candidate to build large parts of the network.
The risk of lock-in is largely compensated by the powerful tools made available to the Network Operations team.