Skip to content
← Solutions

Platform Fabrics

Five programmable fabrics form the foundation of every Clevertek solution — explore the building blocks.

Why we organise the portfolio by fabric

A fabric is the run-and-operate layer beneath a group of products. Where a single product solves one requirement, a fabric is the shared machinery that several products draw on, so that adding a capability does not mean adding another vendor, another console and another escalation path.

Clevertek organises its portfolio into five fabrics. The Network Fabric is the transport that carries traffic between sites, clouds and carriers. The Cloud Fabric is the compute, storage, data and platform layer that workloads run on. The Security Fabric is the control set applied across users, devices, networks and data. The Interaction Fabric is the voice, video, messaging and contact-centre estate your customers and staff actually touch. The IoT Fabric is the device-to-data path for connected operations. Products sit on one or more fabrics, which is why a single engagement can start with connectivity and extend into security, cloud or collaboration without a re-architecture.

How the fabrics relate to a solution

Solutions on this site are organised three ways to match how the requirement arrives. By need answers an outcome, such as moving to a secure network architecture or standing up a hybrid workplace. By industry answers a sector, such as banking or manufacturing, where the regulatory and operational context narrows the design space. By fabric answers the underlying platform question: which shared layer does this capability belong to, and what else already runs there.

The practical consequence is that a change made on one fabric is reviewed against the others. Adding a site to the Network Fabric is checked against the security policy that governs it and the cloud region it needs to reach. Extending the Cloud Fabric with a GPU workload is checked against the data classification that workload carries. That review is what keeps the estate from becoming a set of independent systems that happen to share a rack.

What each fabric carries

The five fabrics are not interchangeable layers of the same stack. Each one carries a different class of requirement, which is what makes the entry point a decision you can reason about.

Network Fabric

Carries every byte that moves between a site, a cloud region, a data centre and a carrier. Its products are dedicated internet access, private lines and MPLS, IP transit, SD-WAN, multi-cloud connectivity, managed Wi-Fi and LAN, and carrier aggregation. Choose this entry point when the requirement is a new site, a degraded circuit, a migration that needs a predictable path, or a WAN estate that has accumulated one contract per location.

Cloud Fabric

Carries the compute, storage and data platforms that workloads run on, across on-premise infrastructure, public cloud tenancy and edge locations. Its products are enterprise cloud, the AI cloud platform, GPU as a service, sovereign GPU cloud, compute storage and managed Kubernetes, multi-cloud management, migration and modernisation, DevOps, disaster recovery, and the AWS, Google Cloud, Azure and Microsoft 365 service lines. Choose this entry point when the requirement is where a workload should live, what it costs to run there, or how it gets there without a rebuild.

Security Fabric

Carries the controls applied to users, devices, networks and data. Its products are zero-trust network access, advanced network security, managed detection and response, cloud security posture, OT security, security assessment and consulting, and backup with ransomware resilience. Choose this entry point when the requirement is a compliance obligation, an audit finding, an incident, or a decision about how remote access should work now that the perimeter has stopped meaning anything.

Interaction Fabric

Carries the conversations your customers and staff have, across voice, video, messaging and contact centre. Its products are cloud voice and telephony, unified communications, contact centre as a service, programmable communications, and compliance recording. Choose this entry point when the requirement is a telephony estate reaching end of hardware life, a contact centre that cannot route by skill, or a workforce that needs one identity across every device it works from.

IoT Fabric

Carries data from physical equipment into the systems that act on it, across connectivity, protocol translation, edge processing and device management. Its products are IoT connectivity, industry solutions for industrial protocols, connected solutions, and edge compute. Choose this entry point when the requirement is visibility into plant or fleet that currently reports nothing, a maintenance model that reacts instead of predicting, or a compliance obligation to record operational data that nobody is capturing today.

Choosing where to start

The entry point is usually set by the requirement rather than by preference. A new office, a fibre cut, or a merger that adds twenty sites puts you on the Network Fabric because transport is the blocker. A model that will not train on the hardware available puts you on the Cloud Fabric. An audit letter with a date on it puts you on the Security Fabric. A PBX that the vendor will no longer support puts you on the Interaction Fabric. A production line whose downtime is currently diagnosed by walking the floor puts you on the IoT Fabric.

Where two fabrics could answer the same requirement, the deciding question is which one the change has to be reviewed against. Moving a workload to public cloud is a Cloud Fabric decision, but if the workload carries regulated data, the Security Fabric sets the constraint on where it can go, and the Network Fabric sets how it gets there without crossing the public internet. Naming the controlling fabric first is what stops a migration project from discovering a residency constraint in its final week.

What stays the same as you extend

Extending from one fabric to the next does not restart the relationship. Monitoring, alerting and reporting stay in the same place, so the estate you already monitor does not acquire a second dashboard that nobody opens. Change control stays in the same queue, which is what lets a routing change and a firewall rule be reviewed against each other rather than shipped independently and reconciled during the next incident. Commercial terms extend product by product, so an added capability is a decision about that capability rather than a renegotiation of everything already running.

The same applies in reverse. Where a fabric turns out not to fit, the products on it are handed back or moved without unpicking the rest. That is the practical test of whether a portfolio is genuinely modular: whether a customer can leave a part of it without leaving all of it.

Design on the right fabric

We will map your workload to the right products across the fabric.

Talk to us