Showing results for 
Search instead for 
Did you mean: 

Innovation, a way of life in Cisco Engineering.

Cisco Employee


Despite tens of thousands of developers and an equally large number of solutions being churned out at high velocity, Cisco continues to keep a constant eye on innovation and relentlessly finds better ways to solve problems.


Lay of the land – insight into Cisco Engineering:

 Cisco has a large number of DevOps activities associated with development and these require an enormous number of resources for running workloads. The resources encompass storage/compute/memory and associated ancillary costs (real estate footprint, electricity, etc). Moving to the cloud does not change the fundamentals of the challenge, since even those cloud workloads at the end of the day need to run on compute and consume electricity.


Cisco internal engineering smelt an opportunity for innovation:

Born out of a Cisco fueled engineering hackathon, this award winning solution with 2 patents filed, took resource optimization of cloud resource at Cisco to a new level.

We came up with something called TimeBox. A data driven resource optimizer, that:

  1. Understands intent.
  2. Provides recommendations.
  3. Monitors and heals workloads, as if on auto-pilot.
  4. Provide insight into workloads.
  5. One stop shop to discover TCO footprint to find a direct mapping to your dollar costs.


Sharing the recipe:




TimeBox, uses machine learning to understand intent of historic workload computations and then uses those to make recommendations of a better schedule. This schedule once tweaked, gets retrained for subsequent more sophisticated AI driven recommendations. The TimeBox works as a smart assistant, automatically answering for Cisco engineering the questions:

  1. What is the optimal resources required for a given workload?
  2. A system on auto-pilot, which monitors and heals aborted workloads.
  3. Knowing exactly what the Total Cost of Ownership is for the workloads.
  4. Prevents accidental hoarding of resources.


In a nutshell:

Scheduling and optimization of resources is not a new problem, but using genetic modelling based AI to solve this NP hard problem may just be. TimeBox can be pervasive with applications across any industry, where there are resources that undergo periodic consumption, there is a need for optimal capacity planning, workloads with large variety and associated variable characteristics and the efficiency of resource allocation is paramount.


Reach out to  , if you want to discuss if TimeBox, an internal engineering innovation can bring down the total Total Cost of Ownership for your cloud or legacy workload strategy, while giving you the satisfaction of being a responsible global citizen by reducing the associated carbon footprint of your cloud datacenters.


Content for Community-Ad

This widget could not be displayed.