Network teams have a lot of goals to achieve this year. Automation skills and DevOps practices must be applied to speed up the DNS, DHCP, and IP address management aspects of continuous integration and continuous delivery pipelines. IP assignment and DNS management issues that are cropping up as workloads migrate to multi- and hybrid-cloud infrastructures have to be solved. More economical and agile ways to scale the infrastructure delivering these network services must be found while also being a good partner to other IT and business teams.
Achieving these myriad goals requires a shift from the traditional appliance-based approach for delivering network services. By delivering DNS, DHCP, and IP address management infrastructure as containerized software, network teams have more deployment flexibility and more opportunities for automation and integration.
Containers 101
Containers are portable software packages that bundle together an application and its related dependencies (configuration files, libraries, language versions, etc.). Everything that the application needs to run is decoupled from the target environment where the application will run.
Once deployed, the application processes within a container are isolated from other containers. Resource requirements and limits can be specified for each container, which eliminates resource contention in the target environment.
All of this adds up to versatility and portability for deployment and functional operation regardless of the physical or virtual target environment. Let us review some of the benefits of NS1’s container-based delivery approach.
Freedom to Deploy Anywhere
Containers can be deployed on any platform – public cloud, private cloud, physical devices, even on a virtual machine. This gives network teams dramatically more flexibility in choosing where and how their containerized software runs. Our customers need the deployment flexibility because enterprise network topologies can vary widely. Their environments also increasingly incorporate cloud-based datacenters, colocation facilities, and private-cloud network fabrics as various business units migrate their workloads.
NS1 takes the flexibility of multi- and hybrid-cloud deployment one step further – by simplifying the effort required to assure service connectivity between cloud-based applications and on-premise applications. Hybrid-cloud strategies can create a mess of multiple components for network teams to stitch together – resulting in connectivity that is anything but seamless. Platform agnostic deployment of NS1 to any cloud as the primary DNS service can eliminate the management of those multiple components.
Small Footprint
Another benefit of containers over virtual appliances is their lightweight footprint. Unlike virtual machines, all containers share the host system’s kernel. Consequently, they require minimal resources to operate. As a result, containers can be spun up on devices with smaller footprint than traditional hardware or virtual appliances.
This makes it more cost effective to deploy NS1 in places that wouldn’t have considered previously, but where by doing so can achieve significant improvement in user experience and application performance. Small, remote offices can have local DNS and/or DHCP services without the additional expense of a dedicated physical appliance. Moreover, organizations can also avoid the costs of shipping the hardware and the productivity hit of having an engineer travel to and from remote locations. We have even seen some of our more forward leaning customers deploy NS1 containers on switches that have native Docker container support. Not only does this save costs (and effort), but also improves performance by delivering DNS and/or DHCP servers closer to the network edge and application users.
Rapid Provisioning, Configuration, and Deployment
Container deployment times are reduced because there is no Guest OS to bootstrap, configure, and connect. There are no last-minute OS updates to install and configure.
This was the case at a global gaming company. They had a manual process to provision, configure, and validate a legacy virtual appliance for their testing environment, which caused feature delivery delays. With every aspect of NS1 configurable through our API, the customer automated their deployment pipeline with Jenkins. They could deploy our container, configure the network, and create the DNS records with configured application-specific traffic steering policies on the fly when they deploy the application.
Not only did NS1 deliver on the promise of infrastructure as code, but our lightweight containers dramatically reduced provisioning and certification times. Dev and test teams finally had automated provisioning of a sandbox environment within the timeframes they needed to deliver critical updates to users.
Easily Scale Up and Scale Down
Since container deployment is programmable and quick, it is easy to spin up and deactivate NS1 containers for short term purposes such as testing or batch jobs or to support temporary locations such as live events or conferences. In fact, some test/sandbox environments are so short lived that a recent survey indicates that 85% of all containers are live for less than 24 hours. So imagine the frustration when the time taken to deploy a legacy appliance is longer than the time it’s actually in operation.
The increasing use of microservices architectures for digital transformation also drives the ephemeral nature of enterprise production environments. Containerized microservices simplify and accelerate deployment to the point where autoscaling capabilities can become an intrinsic part of business applications. Individual services can be programmatically replicated or decommissioned to adjust capacity within minutes. This adds to the requirement for IP and DNS updates at higher rates.
NS1’s API is designed to interact with other infrastructure at much higher velocities than legacy DNS, DHCP and IP Address Management appliances. When combined with container-based deployment, our customers can support modern microservices-based applications with high-frequency updates and unlock flexible scaling.
Cost-Effective Scalability
Our approach redefines the economics of DNS, DHCP and IP Address Management software licensing for enterprises. It is easy to scale up network services for short timeframes or temporary locations without worrying about reaching appliance licensing limits. There is no need to incur appliance provisioning and deployment costs and delays to scale up for high volume periods such as Black Friday or year-end reporting.
Conclusion
As enterprises modernize their applications, network teams also need a modern infrastructure to deliver network services for those applications. Modern application delivery requires a DNS, DHCP, and IP Address Management solution that is platform-agnostic, cloud-native and scalable.
Our approach enables the transition from static network infrastructure with “big bang” upgrades to continuous delivery of software-defined networking services. Network teams can use the same automation tools and infrastructure constructs as their DevOps counterparts. Cumbersome appliance upgrade paths are replaced with deployment processes automated through APIs. The result? Enterprises can accelerate their deployment velocity as well as improve efficiency dramatically without having to incur the overhead of managing traditional appliances.
Please reach out to us so that we can help you in this journey.