Advanced Computing in the Age of AI | Saturday, April 20, 2024

Cisco Expands Container Infrastructure Effort 

Cisco Systems is taking an infrastructure approach to container-based application deployment via an open source project designed to define operational specifications for networking, storage and computing.

Project Contiv unveiled by the networking giant earlier this month is designed to help automate policies for the micro-services used to deliver applications delivered via containers. Cisco (NASDAQ: CSCO) said the container infrastructure project combines application and operational "intent," that is, the specified action to be performed, into what is defined in the open-source project as the Contiv "cluster-wide" intent.

Among the infrastructure policies likely to be specified for container-based applications are network and security requirements, including encryption, firewalls and load balancers. In a blog post, Cisco's Balaji Sivasubramanian added that corporate and government compliance policies also must be specified along with allocating storage and computing resources for determining performance.

Cisco said the open source initiative currently includes networking and storage projects focused on Docker-based application container deployment. The networking element includes a "container network plug-in" to provide infrastructure and security policies for multitenant micro-services deployments. Production environments require greater isolation of application containers running on virtual machines and other cloud hosts as a way of securing applications.

Cisco's Project Contiv architecture.

Meanwhile, a volume plug-in controls Docker storage volumes on multitenant infrastructure, allocating persistent, distributed storage. The tool also includes "intent-based consumption" running on the Ceph open-source storage platform used to store data on a distributed computing cluster.

Sivasubramanian said Contiv would soon be part of Cisco's Project Mantl, a micro-services infrastructure effort consisting of a layered stack that handles system integration by leveraging tools like the Apache Mesos orchestration tool and Kubernetes for cluster management. Cisco promotes the micro-services infrastructure project as part of the "glue" to run hybrid clouds and cloud-native applications delivered via container technologies.

While noting the growing number of tools available to configure cloud and other infrastructure, the Cisco engineer added "there is still need for [administrators] to be able to specify the infrastructure operational policies around network, storage, security, compute for the containerized applications in an automated way and have those policies be implemented across infrastructure consistently."

Without them, "we could have resource contention between production and development applications or security violations between different applications [and/or] tenants and overall unpredictable application performance," Sivasubramanian added.

Hence, Cisco is offering a "better way for containerized applications to run in a shared infrastructure."

Among the other expected infrastructure policies for container-based applications are physical infrastructure policies centered on bandwidth limits and guarantees per container, latency requirements and others, Cisco predicts.

The recent embrace of application containers by large enterprise vendors such as Cisco and Microsoft also illustrates how container technology is gradually moving into the mainstream as customers look to scale their cloud operations. Scaling features derived from emerging micro-services like application host density and smaller memory, computing and storage footprints also will propel cloud adoption, Cisco is betting.

About the author: George Leopold

George Leopold has written about science and technology for more than 30 years, focusing on electronics and aerospace technology. He previously served as executive editor of Electronic Engineering Times. Leopold is the author of "Calculated Risk: The Supersonic Life and Times of Gus Grissom" (Purdue University Press, 2016).

EnterpriseAI