As connectivity costs drop and more devices smarten up, it’s clear the Internet of Things (IoT) is leading the pack in terms of disruptive technology trends. The IoT is growing and evolving at a rate many are struggling to keep up with. With more devices come more communications, and more strain on the network: that burden is especially heavy on Service Providers. But, with that burden comes opportunity. The IoT ‘gold rush’ will predominantly be realised in the delivery of contextual insight and intelligence, requiring fully automated, massively scalable and highly secured datacentres at its heart. The datacentre is the new factory for IoT, and co-location and hosting companies are incredibly well placed to deliver on that.
The Problem with the Internet of Things
The main stumbling block to fully-fledged adoption of IoT is the bandwidth required to accommodate so many ‘smart’ objects connecting to the network. Accommodating this scalability and transformation is a battle in itself, but with increased scale comes a heightened security threat. The network has a larger ‘surface area’ than ever before, which means increased areas of risk and more complex rules to regulate them. Security and rapid, agile scaling are the essential considerations in coping with the unpredictable traffic growth of IoT. These events will continue to grow as more devices get connected, and at the same time the events will become more complex and diverse. Ever-bigger data will explode the volume and the demands on the network. The firewall – physical or virtual – must be able to take immediate action based on known or emerging intelligence. It must identify attacks accurately and act quickly, while providing enough scale to accommodate massive bandwidth and a huge session count. The vast demands of IoT necessitate high performance and high scale, but they must be delivered hand in hand with low latency.
The vast amount of information and events to be analysed mean scalable datacentre networks are a must. Simplification and automation is critical. It’s much easier to scale up and out when the network can do most of the work itself. To enable this, the datacentre must be able to communicate with public and third party clouds, and the things themselves, to make the most of available resources. When every end point becomes a perimeter, micro-segmentation comes into its own, providing security in an increasingly complex environment. Running applications in isolation enables increased security because different customers or contexts can be segregated. Hence providing tiered access at a device level is essential. With such variation of workloads on one network, it needs a sophisticated understanding of priority, security and levels of access. Both mobile and physical networks need to ‘outsmart’ the devices connected to them. Of course, more devices mean more bandwidth, which means more strain and unpredictability for routers and switches. Fortunately, there are a number of ways to reengineer the network to cope:
Path computation element (PCE) implemented over an SDN control plane – to handle the unpredictable growth of traffic and requirements for path optimisation.
Network edges and service cores – to support automated, policy-driven service control and orchestration.
Gateway solutions – to support multiple services, automatically routing traffic and reducing network cost and complexity.
Traffic treatment – determined on a per-device and / or per-application basis, to enable customisable and differentiated services at scale.
Backhaul portion – robust enough to manage the increased traffic and data; efficient and optimised through seamless MPLS, and zero touch provisioning.
And not forgetting…
Hardware – hardened, fanless devices are essential to deployment in far away, hard to reach and often harsh locations.
Security should be guided by business need. Policies must be established and then built into transparent, virtualised network architecture at different levels from physical to the virtual, rather than considered a bolt on or afterthought. Automation will play a huge role in supporting the scale and security demanded by IoT growth. Repetitive tasks must be ‘taught’ to the network, based on business policies, to ensure rapid reaction and resolution to threats. The very process of accommodating IoT is a huge undertaking and Service Providers must be able to make the transition at a manageable pace. We partner with Juniper to provide automated, scalable and secure network solutions that offer agility, performance and value. Tags: Internet of Things, IoT, juniper networks, Security
Machine Learning is Critical to Cyber Security
6 Security Measures for Enterprise Campus Networks
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.