Bandwidth management for ISPs and operators
Bandwidth management for ISPs: see which applications use your links
Bandwidth management for ISPs starts with a plain question: which applications are using the capacity you pay for, and when? Interkey Intelligent DPI classifies traffic by application in real time, including encrypted traffic, without decrypting it. Live View shows the result as network-level and application-level trends, so capacity and plan decisions follow real demand. Where you decide to act, policies are applied per application, subscriber group or cell, under rules that you define and approve. This page answers how an ISP network team reads the picture of its links; for the choice of policy action, see traffic shaping.
A Saudi product, built by Interkey
For network, capacity and product teams at ISPs and telecom operators in Saudi Arabia. Reviewed November 2026.
- What it is:a DPI platform that shows which applications use your bandwidth and applies the policies you define.
- What it does:classifies traffic by application in real time, reports trends in Live View, and applies policy per application, subscriber group or cell.
- Who it is for:network, capacity-planning and product teams at ISPs and operators.
- Where it works:fixed and mobile networks, on your own infrastructure; it can operate entirely in the Kingdom.
What you can see, and what you can do about it
Bandwidth management has two halves: knowing what is on the network, and deciding what to do about it. Most of the value comes from the first half, because a capacity plan, a plan design or a congestion fix is only as good as the picture behind it.
| Question an ISP asks | What the platform provides |
|---|---|
| Which applications fill our links at peak? | Network-level and application-level traffic trends in Live View, so capacity decisions follow real demand. |
| Is the traffic we count really what is on the network? | Classification by application for encrypted and tunnelled flows too, so reports do not undercount them. |
| Where is the pressure: one cell, one group, one application? | Policies and reports scoped per application, subscriber group or cell. |
| Is a rule doing what we approved? | Policy execution feedback in Live View, and triggered policies recorded as metadata. |
| What would a rule have done? | Report-only mode shows the effect of a rule before it acts on traffic. |
Visibility first: knowing what uses the bandwidth
Port numbers no longer tell an ISP what is on a link. Encryption removes the payload signal, and many applications share the same common ports, so counting by port or by payload alone either misses traffic or puts it in the wrong class.
Interkey Intelligent DPI classifies from flow behaviour: handshake structure, packet sizing and timing, and endpoint behaviour. An AI traffic model built into the engine core combines these features, and it is retrained automatically as applications change. Live View then shows the result: a full-network view with drill-down to application level, trend analysis, and latency measured in seconds.
These reports describe classes of traffic and subscriber groups. The platform is built for network governance and regulatory compliance, not for surveillance of individuals. Its output is metadata about applications, traffic volumes and triggered policies.
Governing bandwidth with operator-defined policy
- You write the rules.What counts as fair use, which plans a rule covers and what happens when it triggers are your decisions, made within applicable regulation. The platform applies the policies you define.
- Scope.Policies act per application, subscriber group or cell, so a rule applies where the pressure or the requirement actually is.
- Timing.Policies are deployed and adjusted in real time.
- Placement.A policy acts on traffic only in an inline deployment. Out-of-band, the platform classifies traffic and reports without sitting in the forwarding path.
- Actions.The policy actions include real-time policy deployment and adjustment, service-chain control, traffic shaping, and blocking where your policy and applicable regulation require it. Which actions your deployment uses is defined with your team during scoping and documented in the integration plan. See Traffic shaping for telecom operators.
For rules tied to plans and allowances, see Fair usage policy enforcement for telecom operators.
A practical way to start
1. Classify out-of-band.The platform receives a copy of traffic and shows which applications use the bandwidth, without touching the forwarding path.
2. Measure on your traffic.Check detection and false-positive rates for the classes your decisions depend on. See Measuring DPI classification accuracy on your traffic.
3. Write the rules down.For each rule: the traffic classes, the scope, the action, the basis and an owner. Test each one in report-only mode.
4. Decide on inline.Enforcement follows only after classification has proven itself and a failover test has passed for your topology.
Capacity planning, plan design and congestion analysis can all use the same classification. For the congestion case, see Mobile network congestion by application.
Governance for bandwidth rules
- A policy catalogue.Each rule has its traffic classes, scope, action, commercial or regulatory basis and an owner.
- An approval route.Someone signs off the move from report-only to enforcement, on stated evidence.
- Data rules.Retention and access for the metadata follow your data-governance rules and the contract.
- Change control.Rule changes and model updates enter the live network through your change-control process.
Reading Live View for a bandwidth question
Bandwidth questions rarely need every report. Match the question to the view. For the question "what is on my links right now", use the full-network view with drill-down to application level. For "when and where does an application drive load", use the network-level and application-level trends over time, and compare busy hours across cells. For "is the rule I approved doing its job", use policy execution feedback. For "can I trust what I am seeing", remember that Live View latency is measured in seconds, so a change in traffic shows up quickly, and that classification results depend on your own traffic mix, so check detection and false-positive rates on it first. Write down the question, the view and the decision that depends on it before opening the reports. Teams that start from the reports tend to collect charts; teams that start from the question tend to reach a capacity or plan decision.
If you are looking for which policy action fits which job, read Traffic shaping for telecom operators.
Questions about bandwidth management for ISPs
Does the platform shape traffic?
Can we see which applications use our bandwidth without changing our network?
Does it see encrypted traffic?
Can a rule apply to one group of subscribers only?
Who sets the rules?
How is it priced?
Proof of concept in three phases
Written acceptance criteria.Agree which rules are tested, on which traffic and subscriber groups, how results are measured and what counts as a pass.
Out-of-band classification against verified traffic.Measure false positives as carefully as detection for every class the rules depend on.
Report-only policy, then the inline decision.Each rule shows what it would have done; enforcement follows only after classification has proven itself.
Discuss bandwidth management on your network
Tell us what you want to see on your links, which plans or groups matter, and your timeline. Our team in Riyadh replies in Arabic or English.
- Our engineers read your request
- We agree the scope of a technical session
- You see the product on a use case you choose