GLPI vs iTop: which to choose

GLPI vs iTop: when each platform is the better choice

Choose GLPI when you want ITIL tickets, problems, changes and asset management in one product, with inventory collected by the GLPI Agent, and you want the option of vendor-backed support from Teclib' through its GLPI Network subscriptions. GLPI fits IT teams whose main question is "what do we own, who uses it, and how do we serve it?" and who want the service desk and the asset register to share one data model from day one. Its native agent inventory has been available since version 10, so a team with a large Windows and Linux estate can start collecting hardware and software data without first adding a separate tool.

Choose iTop when CMDB modeling and change management are the first priorities and your agent count fits one of its editions. Combodo positions the free AGPL Community edition for up to 10 agents, with the paid Atelier edition for up to 50 agents and Forge for more than 50, available as SaaS or on premises. If your team will spend most of its design time on configuration item classes, relationships and change approval workflows, iTop is built around that model. Confirm with Combodo which edition your agent count requires before you plan the rollout, because the tier boundary decides the licensing conversation.

The decision turns on four points. First, discovery: GLPI's network discovery and SNMP inventory need the separate GLPI Inventory plugin, so list every device type you must find without an agent and test that path before committing. Second, agent count, which determines whether iTop's Community tier is enough. Third, the change approval process: map your required approvers, emergency change path and CAB workflow against each product's change module in a proof of concept. Fourth, the support model: GLPI subscription prices and iTop paid edition prices are not covered here, so request written quotes from Teclib' and Combodo that name the support levels, response expectations and upgrade services included.

Which platform fits which IT scenario

This table maps common GLPI vs iTop situations to the platform that usually fits better and the condition that makes it true. Read each row as a starting hypothesis, then confirm the listed items in your own environment before shortlisting.

Scenario	Better fit	Why and what to confirm
Service desk that mainly needs ticketing plus endpoint hardware and software inventory	GLPI	GLPI handles ITIL tickets, problems and changes alongside asset management, with native inventory from the GLPI Agent since version 10. Confirm that agent-based inventory covers your endpoint operating systems and that your ticket workflows fit GLPI's ITIL model.
Team that must model many interdependent configuration items and relationships	iTop	iTop is built around a CMDB with a configurable data model. Confirm how many CI classes and relationship types you need, and test a representative slice of your data model before committing.
Network with heavy SNMP or network discovery needs	Depends on the discovery tooling you can run	GLPI needs the separate GLPI Inventory plugin for network discovery and SNMP inventory. Confirm which discovery features iTop offers in your edition and whether they match your device mix and scan scope.
Estate that will grow past 10 or 50 agents	iTop, if agent tiers match your growth	Combodo positions the free Community edition for up to 10 agents, Atelier for up to 50 and Forge for more than 50. Confirm the agent count you will reach and the current edition terms with Combodo, since tiers drive licensing cost.
Organization that wants vendor support under a paid subscription	GLPI	Teclib' sells GLPI Network subscriptions and cloud hosting. Confirm the support levels, response terms and whether Combodo's paid editions give you the support model you need.
Team with strong in-house Linux and PHP skills that will run the platform itself	Either, decided by data model and discovery needs	Both are self-hosted PHP applications, so existing skills transfer, but the choice should follow scope. Confirm upgrade procedures, backup approach and the effort to maintain custom fields, plugins or data model changes for your release cycle.

How GLPI and iTop are built and deployed, and what that means for operations

GLPI and iTop both target ITIL service management with a configuration database, but they reach that goal differently. The cards below cover how each product is packaged, where discovery comes from, how each is offered for self-hosting, and which release and upgrade facts your operations team should verify before committing.

GLPI as a single product

GLPI bundles ITIL tickets, problems, changes and asset management in one application maintained by Teclib'. Native inventory comes from the GLPI Agent, which has been part of the product since version 10, so basic computer inventory does not require a second product. Check that the agent deployment model, reporting needs and plugin set in your environment match what the single codebase provides before you design around it.

Network discovery needs a plugin

In GLPI, network discovery and SNMP inventory are not part of the core agent path. They depend on the separate GLPI Inventory plugin, which adds a second component to install, configure, patch and test. Ask Teclib' or your integrator which plugin versions match each GLPI release, and include the plugin in your upgrade plan and test scope.

iTop as an AGPL ITSM and CMDB

iTop is an AGPL-3.0 ITSM and CMDB platform from Combodo. Its configuration model is the center of the product, so the data model, lifecycle rules and CMDB relationships are where most design effort goes. Confirm which discovery connectors and agents your environment needs, since the vendor documentation and edition terms determine what ships with the product.

Hosting choice from the vendor

Combodo offers iTop as SaaS or on premises, so an on-premises deployment inside your own cloud account, data center or isolated network is a supported path. GLPI is self-hosted in your infrastructure by default, with Teclib' also selling cloud hosting. Decide early where configuration data, inventory and attachments must live, since that choice shapes backup, access and audit evidence.

Release cadence in the supplied versions

GLPI published 10.0.28 on September 30, 2026, 11.0.11 on October 1, 2026 and 12.0.0 on October 7, 2026, so three major lines are maintained at once. iTop published 3.3.0 on September 15, 2026 and 3.2.3-3 on October 16, 2026. Treat parallel branches as a planning input: pick the line you will run, know its end-of-support date, and schedule upgrades against that date.

Confirm upgrade paths and runtime versions

Before you choose a version, ask each vendor for the supported upgrade path between your current release and the target, the database engines and versions supported, and the supported PHP version for each line. These requirements change between major releases and determine whether an in-place upgrade or a parallel rebuild is needed. Record the answers in writing, and test the upgrade with a copy of your production CMDB before go-live.

CMDB depth, change management and discovery compared

This table compares GLPI and iTop on licensing, CMDB approach, ITIL process coverage, inventory, discovery, deployment, editions and vendor support. Where the supplied facts stop short of the detail you need, the cell says to confirm with the vendor, and the sample data test later on this page explains how to check it yourself. Read each row as a question to answer before you shortlist either platform.

Capability	GLPI	iTop
License	GPL-3.0-or-later	AGPL-3.0
Data model and CMDB approach	Assets and inventory items linked to ITIL objects. Confirm with vendor how relationship types and CI classes can be extended.	Configuration items modeled as classes with relationships. Confirm with vendor how far the class model can be customized without core changes.
ITIL tickets, problems and changes	Handles ITIL tickets, problems and changes. Confirm with vendor whether change approval workflows can be configured per change type.	ITSM and CMDB platform. Confirm with vendor which change approval and validation steps ship by default and which need configuration.
Asset management	Asset management is part of the core product, linked to tickets and changes.	CMDB is the core of the product. Confirm with vendor how asset lifecycle states are modeled.
Native agent inventory	Native inventory from the GLPI Agent, available since version 10.	Agent-based inventory is tied to the Community, Atelier and Forge agent tiers. Confirm with vendor which inventory agent and version to use.
Network discovery and SNMP	Needs the separate GLPI Inventory plugin. Confirm with vendor its supported devices and version compatibility.	Confirm with vendor which discovery mechanisms and device types are included in each edition.
Deployment options	Self-hosted in your own infrastructure. Teclib' also offers cloud hosting. Confirm with vendor which hosting options fit your data residency needs.	SaaS or on premises, as stated by Combodo. Confirm with vendor which hosting model is supported for your edition.
Editions and agent limits	GLPI is open source. Paid Teclib' GLPI Network subscriptions are sold. Confirm with vendor what subscriptions include.	Free Community edition positioned for up to 10 agents. Paid Atelier for up to 50 agents and Forge for more than 50. Confirm agent counting rules with vendor.
Vendor support	Teclib' sells GLPI Network subscriptions and cloud hosting. Confirm support response terms in writing before you rely on them.	Combodo sells Atelier and Forge tiers. Confirm support scope and response terms for each tier in writing.

What is free and what is paid in GLPI and iTop

GLPI is licensed under GPL-3.0-or-later and maintained by Teclib', which sells GLPI Network subscriptions and cloud hosting. The open-source software carries no license fee, so recurring spending in a GLPI program usually goes to vendor subscriptions, hosting, and the staff or partners who install, configure, and operate the platform. The subscription price and its contents are not supplied here, so request them in writing from Teclib' before you build a budget.

Core license: GPL-3.0-or-later covers the GLPI source code you download and run in your own infrastructure.
Vendor subscription: GLPI Network is sold by Teclib'. Ask which support hours, response terms, and services are included, and whether any feature you plan to use depends on the subscription.
Cloud hosting: Teclib' also sells hosting. If you choose it, data location, backup responsibility, and exit terms move into the vendor contract.
Plugins: Network discovery and SNMP inventory need the separate GLPI Inventory plugin, so confirm its license and support status on its own.

Confirm in writing which GLPI Network services are priced separately from the core software and whether the plugins you need carry terms of their own. Start from the GLPI project site when you compare the vendor's published offers with your written quote.

What drives the running cost of GLPI and iTop

The running cost of GLPI and iTop is set less by the software license than by who hosts it, how far the CMDB and discovery reach, and how many agents or users the tier must cover. Both platforms are open source and self-hosted in your own cloud account, data center or air-gapped network, so the infrastructure bill, the database and web-tier hosting, and staff time are paid to you or your provider, not to a license vendor. GLPI is GPL-3.0-or-later and iTop is AGPL-3.0, so neither requires a per-seat fee to run the Community code. Where a vendor is paid is in support, subscriptions and tiers: Teclib' sells GLPI Network subscriptions and cloud hosting, and Combodo positions iTop's free Community edition for up to 10 agents, with paid Atelier for up to 50 agents and Forge for more than 50, in SaaS or on premises. Confirm current tier names, limits and fees on each vendor's own pricing pages before you budget.

Some costs are one-time and some recur. The one-time work is designing the data model (classes, relations and required fields in iTop, or asset types and fields in GLPI), tuning discovery scope, and building integrations with identity, monitoring and the service desk. Those costs are driven by how many asset classes you model and how many systems feed the CMDB. The recurring work is tier or subscription fees, vendor or partner support, upgrades across major versions, and maintenance of the agents. For GLPI, native inventory comes from the GLPI Agent since version 10, but network discovery and SNMP inventory need the separate GLPI Inventory plugin. That plugin is an extra component to test on each upgrade, to keep compatible with the core, and to document in your change process. For iTop, the agent count is the main tier lever: once your managed agents pass 10 or 50, the paid tier changes. Count agents the way the vendor does before you select a tier, and model growth for the next two or three years.

When you request a quote from DeployOpen or any other implementer, ask for a written scope that states the platforms and versions in scope, the number of agents or assets at go-live and expected growth, the discovery scope (networks, SNMP devices, cloud accounts and endpoints), the integrations to build and test, and the training and change process design included. Ask for the support level and response expectations separately from the implementation, and state whether the vendor subscription is quoted or left to you. Ask for upgrade assumptions: which major versions are expected during the term, who tests the GLPI Inventory plugin or iTop extensions against them, and whether backups and rollback are part of the retainer. DeployOpen does not publish a price list, so a quote follows from this written scope and should be compared line by line with the same assumptions.

Where GLPI and iTop projects usually run long

Most GLPI and iTop delays come from decisions made before any data moves: which configuration item classes exist, how clean the current asset records are, and how far discovery reaches. The cards below cover the risks that tend to stretch schedules, with the preparation that shortens each one.

CMDB data model agreement

Teams often start importing before they agree which configuration item classes they need and how those classes relate. In iTop, the class model and relationships shape change impact and reporting, so changing them late forces rework in forms, reports and integrations. In GLPI, the asset types, dropdowns and relations need the same agreement. Prepare a written list of classes, owners, mandatory fields and relationships, and test it against a few real service examples before configuring either tool.

Cleaning existing asset data

Duplicate records and stale entries are usually the biggest hidden cost. Spreadsheets, old CMDB exports and inventory tools often disagree on hostnames, serial numbers and owners. Deduplicating before import is slower than importing and then fixing, but it avoids having two records for one server in a live CMDB. Ask who owns the source of truth for each asset type, and agree matching keys (serial, asset tag, hostname) before any bulk load.

Agent rollout across mixed estates

Inventory depends on agents reaching the central server from every operating system in scope. GLPI includes native inventory through the GLPI Agent, so the rollout effort sits mostly in packaging, deployment tooling and network paths to the server. For iTop, the agent route and any inventory connectors must be confirmed for each platform you run. Prepare a list of operating systems, versions, segments and proxy or firewall rules, and pilot on one group per platform before wide deployment.

Discovery and SNMP scope

Network discovery and SNMP inventory are where scope creep starts. In GLPI, these capabilities need the separate GLPI Inventory plugin, so it must be installed, versioned and tested as its own component. Scan ranges, SNMP credentials, community strings and device types must be approved by the network team before discovery runs. Decide which devices belong in the CMDB at all, since unmanaged scans can flood the database with objects nobody owns. Confirm the equivalent discovery path for iTop with the vendor before you commit.

Change workflow and approvals

Change management is where a tool must match how the organization already works. GLPI covers ITIL changes and problems natively, so the work is mapping existing approval tiers, emergency paths and CAB rules into its workflow. iTop has its own change and request model, and the approval design must be rebuilt in its terms. Gather current change templates, approver roles and exception paths, then review them with the people who run CAB before configuration starts, since changing approval logic after go-live disrupts trust quickly.

Identity, email and monitoring links

Integrations with the directory, the mail system and the monitoring stack are often scoped last and take the longest to test. User and group sync decides who can log in and who can be assigned tickets, while mail ingestion and notifications depend on mailbox rules and TLS settings. Monitoring feeds create or update CIs, which can conflict with manual edits. List each integration, its owner, the data direction and the authoritative system for each field, then confirm the supported connectors with the vendor for your exact versions.

Moving from GLPI to iTop, or the reverse, step by step

This sequence moves asset, configuration item and ticket history from one ITSM suite to the other with counts reconciled at every stage, so the team can cut over knowing what was carried, what was rebuilt and what was left behind. Each step ends with a check you can verify before moving on.

Export and audit the source data
Your team, with the source instance administrator, exports computers, network devices, software, contracts, tickets, problems, changes and their relationships from GLPI or iTop into files you can open outside the application. Record the object counts per type, the number of open tickets and the number of inventory sources that feed the database, such as the GLPI Agent or the GLPI Inventory plugin on the GLPI side. Stale records, duplicate hostnames and orphaned software entries are flagged now, not after the move. The step is done when you have a signed-off inventory of counts and known data quality problems. Confirm with each vendor which export formats and API endpoints the current release supports, since the supplied facts do not describe a migration tool for either product.
Decide which classes and relationships carry over
Architects and service owners decide which configuration item classes must exist in the target model, such as servers, virtual machines, applications, databases and network devices, and which relationships matter for impact analysis, for example application to server or server to network device. GLPI's asset types and iTop's CI classes rarely map one to one, so write a mapping table with a column for each side and a column for items that will be dropped or merged. Items that exist only for inventory noise, such as transient processes or duplicate software versions, are marked for exclusion. The step is done when every source type has a named target or an explicit decision to exclude it, approved by the CMDB owner.
Map ticket states, change types and categories
Service desk leads map each ticket status, incident category, problem and change type in the source to a status, category or change class in the target, and record which approval steps each change type required. ITIL-style workflows differ in names and transitions, so states that have no equivalent need a decision: merge, rename or retire. Historical closed tickets usually keep their final status; open tickets need a deliberate mapping because workflow rules and SLA timers may not carry over. Expect the change approval logic to be rebuilt by hand in the target and tested with sample changes. The step is done when a written mapping has been reviewed against three real examples of each ticket and change type.
Build the target model and a test instance
The platform administrator installs the target product on a separate test environment, configures the CI classes, lifecycle states, ticket workflows, notification rules and authentication against your directory, then applies the mapping from the previous steps. Record each configuration choice in a document so it can be repeated in production. Confirm the supported installation requirements and edition limits with the vendor before provisioning, because the facts supplied for this page cover licensing tiers for iTop but not the GLPI or iTop hosting prerequisites in detail. The step is done when the test instance runs, accepts a manually created ticket and change, and the configuration document matches what is in the system.
Load a sample set and reconcile counts
The data engineer loads a representative sample, including every class, a few hundred relationships and the open tickets, into the test instance using the import path confirmed with the vendor. Compare record counts, relationship counts and status totals between the source export and the target, and inspect a sample of records field by field. Anything that did not load is logged with its reason, and rebuilt by hand where it matters. Expect custom fields, attachments and historical audit trails to need separate handling. The step is done when counts match or each difference has a documented explanation and an owner.
Re-point agents and discovery sources
The infrastructure team changes the inventory targets on endpoints and network discovery jobs so that they report to the new platform. On the GLPI side, agent inventory uses the GLPI Agent, and network discovery and SNMP inventory depend on the separate GLPI Inventory plugin, so that plugin path needs its own check on the target. Move a pilot group first, confirm that the target receives a fresh inventory record for each device, and check that duplicates are merged rather than created. The step is done when the pilot devices report correctly, the discovery schedule runs, and the old sources are disabled for that group only.
Run a parallel period
For a defined period agreed in your change plan, both systems run side by side. Service desk staff log new tickets in the target, while the source remains read-only for reference, and each day the team compares new tickets, changes and inventory counts across the two systems. Discrepancies are fixed in the mapping or the discovery configuration, not patched by hand in one system only. The period length is a decision for your change board and is not supplied here. The step is done when two consecutive reporting cycles show no unexplained differences and the service desk confirms the target workflows work for daily use.
Cut over and retire the old instance
The change board approves the cutover window, the source is set to read-only, a final delta export is loaded into the target, and users, groups and integrations are switched to the new endpoints. Keep an archived, read-only copy of the old database and file store under your retention policy, with a documented restore test. Retire the old instance only after the archive is verified and the support contract for the old platform has been formally ended or moved to read-only support. The step is done when the old instance is archived, its access is removed, and the post-cutover checklist is signed by the service owner.

Questions buyers ask about GLPI vs iTop

How much does it cost to run GLPI or iTop?

The software licenses are not the main cost, and no price for running either platform is supplied here, so treat any figure as something to get from a written quote. GLPI is GPL-3.0-or-later and has a paid option through Teclib' GLPI Network subscriptions and cloud hosting. iTop is AGPL-3.0, with Combodo offering a free Community edition positioned for up to 10 agents and paid Atelier (up to 50 agents) and Forge (more than 50) tiers, available as SaaS or on premises. The recurring costs usually come from subscriptions or tiers, hosting and infrastructure, staff time, and patching. The one-time costs are planning, configuration, data cleanup and integration work. Ask each vendor for the exact tier, agent count, support level and renewal terms, and ask a partner for a fixed-scope quote covering the work around them.

How long does a GLPI or iTop migration take?

No migration duration is supplied for either platform, so the timeline depends on the work you commission. The main drivers are the number of assets and their data quality, how many existing ticketing or CMDB sources must be moved, the complexity of your change and approval workflows, how many integrations need rebuilding, and how many teams must be trained. Agent rollout across endpoints and any network discovery design add further time, especially in segmented networks. Ask for a written scope with deliverables and a timeline before work starts. DeployOpen plans, implements, customizes and integrates self-hosted platforms in your own infrastructure, and most engagements are fixed-scope, so the timeline is agreed in writing rather than assumed.

Can an internal team run GLPI or iTop without outside help?

Yes, some teams do, but it depends on skills you already have and how much change you are ready to govern. Running either platform well needs administrators who understand the database, web server and upgrade process, staff who can design CMDB classes and relationships, and someone who owns the ITIL change and problem processes the tool enforces. Discovery is a specialist task: GLPI uses its agent for inventory and needs the GLPI Inventory plugin for network discovery and SNMP, and iTop depends on its own agent and integration choices. Before deciding, confirm your upgrade cadence, backup testing and who responds to incidents at night. If those gaps exist, a fixed-scope implementation or a retainer for upgrades and support can cover them.

What does DeployOpen need from our team to start?

DeployOpen needs enough access and decisions to scope the work accurately before any build starts. Typically that means access to the target infrastructure, such as cloud accounts, data centers or air-gapped networks where the platform will run, plus read access to the current asset, ticketing or CMDB data for sample extracts. You should also name process owners for incident, problem, change and asset management, since those workflows drive configuration. For agent-based inventory, you will need approval for the rollout plan, the endpoints in scope and the change window. Security and compliance contacts should confirm evidence requirements early. Engagements are agreed in writing, so the list of access, samples and approvals becomes part of the scope.

Which vendor support model fits a regulated environment?

Vendor support and implementation partner support answer different questions, and regulated teams usually need both. Vendor subscriptions, such as Teclib' GLPI Network or Combodo's Atelier and Forge tiers, cover product-level support and access to the vendor's release channel. An implementation partner plans, configures, hardens, documents and supports the platform in your environment, prepares controls and evidence for compliance assessments, and runs security testing. DeployOpen does not issue audit reports, attestations or certifications; independent auditors and certification bodies do that. Ask each vendor what support level, response commitments and upgrade paths their subscription includes, and ask your auditor which evidence they will need from the platform.

Do iTop's agent tiers change as our estate grows?

Yes, iTop's positioning is tied to agent count. Combodo describes the free AGPL Community edition for up to 10 agents, paid Atelier for up to 50 agents and Forge for more than 50, available as SaaS or on premises. If your agent count moves across a tier boundary, the edition you need changes, so model growth before you commit. Confirm with Combodo how agents are counted, whether counting includes all devices or only managed agents, how renewals and upgrades are priced, and what happens to data or configuration when you move between tiers. Your procurement team should ask for those terms in writing.

Is GLPI Inventory required for network discovery?

Yes, for network discovery and SNMP inventory, the GLPI documentation says you need the separate GLPI Inventory plugin. GLPI's native inventory comes from the GLPI Agent, which has been part of the platform since version 10, so endpoint inventory works without the plugin. Plan for the plugin if you need to discover switches, routers, printers or other SNMP-managed devices. Confirm the plugin version that matches your GLPI release and test discovery in a segment before a wide rollout, since scan scope, credentials and network paths decide how complete the results are.

How DeployOpen helps choose, implement or migrate between GLPI and iTop

DeployOpen is a Chainscore Labs company that plans, implements, customizes, integrates, hardens, documents and supports self-hosted open-source platforms inside the client's own infrastructure. For GLPI vs iTop, the work starts with the written question of which suite fits your CMDB depth, discovery approach and change process, then moves to a fixed-scope build or migration.

How we scope

Agree in writing which CMDB classes, ITIL processes and inventory sources the engagement covers
Confirm which discovery method you need, including whether network and SNMP inventory is in scope
Confirm current license tier, agent count and edition needs before any build work is quoted

How we build

Configure the chosen suite in your own cloud account, data center or air-gapped network
Map asset, ticket, problem and change workflows to the target platform's data model
Test integrations with your identity provider, directory and monitoring tools before go-live

How we hand over

Deliver documentation of configuration, data mappings, operating procedures and rollback steps
Provide migration reconciliation records showing which CI and ticket data moved and was verified
Train named administrators on daily operation, upgrades and backup and restore procedures

How we support

Offer ongoing support, upgrades and continuous testing as a retainer after handover
Track upstream releases for the chosen suite and plan upgrades against your change calendar
Prepare control evidence for compliance assessments, while independent auditors issue any formal reports

GLPI vs iTop