In February, Cisco announced its latest innovation – Cisco Crosswork Network Automation – a new network automation portfolio for Service Providers.  Read Jonathan Davidson’s blog for an overview to understand our comprehensive approach to enabling a closed-loop, mass-scale automation solution. This multi-part blog series presents each of the five new pillars in the Cisco Crosswork portfolio.


Over recent weeks, we have taken you on a journey to learn more about the pillars in the Cisco Crosswork network automation portfolio. So far, we have presented:  Change Automation; Network Insights; Situation Manager; and Health Insights.  Today, we will conclude this blog series by taking a closer look at the final pillar – Cisco Crosswork Data Platform.



Big data analytics is transforming the way in which service providers derive insight from networks and compute infrastructure data as they move towards a future of closed-loop automation and self-healing infrastructures. These insights not only help enhance operational efficiency and end-user experience, they also improve the overall business outcomes when combined with additional sources of information on the subscriber profile, service performance, SLA and churn rates.

The challenge is two-fold though: firstly, how to deal with the data itself in terms of the increasing volume of network telemetry, disparate datasets, siloed OSS application stacks, and, secondly, how to deploy a high performance, horizontally-scalable, end-to-end data platform that can address a variety of use cases demanded by the customers. Big data solutions that are able to meet this challenge are undergoing rapid innovation themselves, and combining the latest, best-in-class, open source technologies into an integrated platform can be both complex and time-consuming.

In releasing Cisco Crosswork Data Platform, we are helping service providers effectively tackle these challenges and reducing the need for them to deal with underlying complexity.  Providing an integrated and open solution with reliable data collection, high-performance processing and scalable storage services, the platform enables data scientists and network experts to focus on deriving meaningful insight via their use cases rather than orchestrating and managing the underlying infrastructure. Now, you do not need to undertake an arduous task of building your own data platform and a cumbersome process of selecting the right open source components but can rely on Cisco Crosswork Data Platform as a hardened solution with integrated scale-out collection service. In addition, you can take advantage of a range of applications developed on the platform by the Cisco Partners, Cisco Advanced Services, or your own teams.

Today, service provider networks are experiencing accelerating data volume, data variety and velocity. This trend is caused by the increasing number of end-points, technology layers, proliferation of applications and diversity of use cases ranging from real-time automated remediation and optimization to longer-term service usage and subscriber churn analysis. The underlying datasets are equally diverse with different formats and schemas, both structured and unstructured and, in this multi-dimensional problem space, operators are looking for instant access to actionable insights.


Cisco Crosswork Data Platform is a network-aware big-data solution embracing the following architectural principles to meet the challenge:

  • Decoupled data aggregation (producers) from data analysis (consumers) – eliminates siloed OSS stacks, collects data once and makes this data available to multiple applications, improves overall operational productivity and reduces the cost of platform integration.
  • Real-time, stream and batch data processing – enables a wide range of use-cases, ranging from real-time health monitoring to long-term trend analysis and reporting.
  • Immutable dataset – allows raw data to be transformed on-demand to meet a variety of use-cases, improving ingress performance and richness of supported use-cases.
  • Simple, scalable, and extensible – provides horizontally-scalable compute and storage to support dynamic and fluctuating analytics workloads, delivering a cost-optimized solution
  • Open and Programmable – expedites application development with real-time query support and common data processing libraries, ensuring that apps can meet service-provider-specific requirements.

Figure 1: Crosswork Data Platform Capability Overview

Cisco Crosswork Data Platform offers an intuitive Console that provides comprehensive, real-time visibility into the health of platform components to substantially reduce the administrative effort and ensure optimized deployment. Contextual guidance helps you to quickly resolve any performance issue mitigating any impact on the hosted application components. In addition, Console allows you to deploy or undeploy application packages, manage the lifecycle of application components and define data retention policies.

Cisco® Crosswork Data Platform is an integrated offering that combines components of PNDATM and allows integration with data collection services to provide an end-to-end platform that enables network data analytics. It empowers operations teams to deal with data volumes, variety and velocity and enables high-performance analytics to help drive and automate operational decisions.

Visit us at “Unlock the Power of Data Zone” booth in the “World of Solutions” at Cisco Live 2018 US to experience Crosswork Data Platform in action. Would also encourage you to attend the Product and Solution Overview session at 8AM on Tuesday, June 11th.

The five new pillars of the Cisco Crosswork network automation solution enable Service Providers to accelerate their journey to a fully self-healing infrastructure.  We will continue to keep you up-to-date on the latest developments, so watch for future blogs and stay informed.

PNDATM is the “Platform for Network Data Analytics” established under the Linux Foundation as PNDA Project a Series of LF Projects, LLC


Jérémie Garnier

Software Engineer

Service Provider