DCN April 2017 - Page 41

hybrid cloud A world of difference The hype and promise of the cloud can often be far removed from reality. There can be a world of difference between spinning up a new compute instance and actually getting a working instance on to a production network, into service, and in sync with the traditional infrastructure. If, for example, the automation of core DDI network services in the cloud lags behind compute and storage processes, the roll out of applications can be delayed, and the number of inconsistencies in network policies can increase. Without a central resource with which to manage DNS and IP addresses in the cloud, IT teams may find that the view they have of their networks, virtual cloud platforms (VPCs), IP addresses, and the DNS records being assigned is incomplete and out of date. What’s more, the need to use multiple platforms in a hybrid environment will lead to a lack of correlation and consistency with regards to common resources such as DNS zones and networks. While the server team may be responsible for the virtualisation component, for example, all the network aspects may be handled by a different team altogether. The network team may, therefore, have no visibility into the creation and destruction of virtual machine (VM) resources, making it difficult for them to be linked to automated set-up and configuration tasks. Lacking this visibility, and without any accurate and up to date information on which IP addresses and DNS records have been assigned to which VMs at any particular time, networking teams may feel there’s little point in even attempting to meet audit and security compliance policies. Most server admins will have access to information on the various different factors such as ‘The hype and promise of the cloud can often be far removed from reality.’ applications, locations, and users that need to be tracked for VMs and networks, IP addresses, and DNS zones. It’s likely, however, that most networking teams won’t have access to this information and, with a reliance on manual methods to create and destroy VMs, their responses will be slow. The time it can take for the manual provisioning of IP addresses and DNS records in a virtual environment can be a hindrance to the promise of rapid delivery – arguably one of the main attractions of cloud deployment. Fundamentally, virtualised resources can’t be seen on the network without a DNS entry, and if they can’t be seen then they can’t be used. It’s all very well for an organisation to be able to provision resources in just a matter of minutes, but it takes days or even weeks before these resources are in use, then its investment in cloud technology is basically worthless. On top of this, the risk of unreliable DDI management causing a potentially costly network outage represents a serious threat to an organisation, which could have an impact far beyond the network itself, as illustrated by recent news headlines. Strong and secure foundation If they are to perform effectively, those organisations running critical workloads in the cloud, or spanning multiple different geographical locations, will require the scalability and resilience offered by a foundation of highly available, secure DDI services. Enabling this requires a centralised platform that can deliver greater levels of automation and visibility, helping to control, secure, and analyse next generation cloud environments and data centres. Automating DNS provisioning as part of existing workflows, for example, and in the clean-up following the destruction of VMs can eliminate unnecessary trouble tickets and time-consuming manual processes. This will then eliminate the most common bottleneck affecting the successful implementation and rollout of cloud applications, and will allow hybrid cloud deployments to be optimised to meet an organisation’s particular needs. Consolidating different terms and naming conventions via a single, unified console will allow the identification and tracking of virtual machines and components across disparate platforms and cloud environments, thereby avoiding blind spots and incongruent views within the network. Automatically documenting the destruction of a VM, cleaning up its DNS record, and releasing its IP address will ensure that all appropriate information is accurate and up to date. And, by distributing authorisation for permission, IT teams are able to maintain oversight and control of the hybrid cloud as it evolves while, at the same time, empowering the workload of individual employees. As more organisations transition to a hybrid cloud environment, it’s never been more importan