IBM License Metric Tool (ILMT) is the foundation of IBM's deployment measurement, compliance validation, and audit process. Understanding ILMT — how it works, what data it collects, how to validate accuracy, and how to defend audit findings — is critical to managing IBM audit exposure. This guide covers ILMT configuration, data validation, audit response, and the negotiation tactics around ILMT-based findings.
IBM License Metric Tool (ILMT) is automated software that runs on servers where IBM software is deployed. ILMT's purpose is to measure and report hardware specifications, deployed software products, and usage metrics to assess compliance with IBM licence terms.
ILMT is not neutral software. It is designed, controlled, and maintained by IBM to capture data in ways that support IBM's audit and enforcement process. ILMT collects far more data than necessary for simple compliance measurement — it collects detailed information about hardware, operating systems, competing software products, and deployments. Much of this data has little to do with actual IBM software compliance.
ILMT automatically sends collected data to IBM (unless explicitly configured otherwise). This creates information asymmetry: IBM knows what your deployment looks like before you conduct your own audit or analysis. This is why ILMT data is so powerful in IBM's hands — they see your environment through the lens of data they control and interpret.
ILMT is not a neutral measurement tool. It is designed to maximize IBM's audit leverage. Control ILMT configuration, validate data accuracy, and never allow ILMT data to be the sole basis for IBM's audit findings. Demand independent verification and your own deployment measurement.
ILMT collects six categories of data:
Want independent help negotiating better terms? We rank the top advisory firms across 14 vendor categories — free matching, no commitment.
Processor type, count, cores, hyperthreading configuration, memory, storage. This data determines PVU requirements and is the most important for licensing calculations.
Which IBM products are installed, versions, and deployment dates. ILMT detects IBM software even if it's not actively running.
OS type, version, partitioning configuration, virtual machine settings. This determines whether IBM software can be partially licensed (e.g., via hard partitioning) or must be licensed for full system capacity.
ILMT reports on Oracle, Microsoft, SAP, and other vendor products deployed. IBM uses this to identify opportunities to expand IBM footprint and develop account strategy.
Database activity, transaction counts, peak usage. ILMT measures peak usage (often 1–2 hours per quarter when system load is highest) rather than average usage. This inflates actual usage and compliance requirements.
System connectivity, backup/disaster recovery environments, inter-system replication. IBM uses this to identify systems that might be subject to licensing (e.g., active standby systems used for disaster recovery).
ILMT data is frequently inaccurate. Common errors include:
ILMT sometimes double-counts processors in virtualised environments, counts vCPUs as if they were full processors, or misidentifies processor types. A system with 128 vCPUs allocated across 16 physical processors might be counted by ILMT as 128 physical processors, inflating PVU requirements by 8x.
ILMT instances on clustered systems or systems with multiple network interfaces sometimes report the same physical system multiple times. A single database server in a 4-node cluster might be reported 4 times, overstating capacity 4x.
ILMT reports peak usage during brief periods of high load (typically 1–2 hours per quarter). This is not a representative measure of actual usage. Peak usage can be 3–5x average usage. Using peak as the licensing basis inflates requirements significantly.
Passive disaster recovery systems (that run 1–2 times per year for testing) are sometimes counted by ILMT as active deployments requiring licensing. If a DR system is identical in capacity to the production system, ILMT can double the total licensing requirement.
ILMT sometimes detects software that's installed but not licensed (e.g., Oracle's free XE edition installed on the same server as DB2). These false positives complicate audit analysis.
ILMT's default configuration maximises IBM's audit leverage. Proper configuration requires intentional effort:
Get the IT Negotiation Playbook — free
Used by 4,200+ IT directors and procurement leads. Oracle, Microsoft, SAP, Cloud — all covered.
Configure ILMT to send data only to your internal systems, not to IBM directly. This gives you the opportunity to validate and correct data before it reaches IBM. Many organisations disable automatic transmission to IBM servers entirely.
ILMT's default is to measure all systems. If you have specific systems where IBM software is deployed, configure ILMT to measure only those systems. This reduces noise and false positives.
Configure ILMT to report average usage, not peak usage, where applicable. Some products allow this configuration; others do not. Verify your product's capabilities.
Document exactly how ILMT is configured, which systems are measured, and what assumptions are embedded in the measurement. This documentation becomes critical if an audit dispute arises.
Before relying on ILMT data for compliance or audit response, validate accuracy by cross-referencing with your own records:
Compare ILMT processor counts with physical hardware inventories, BIOS settings, and operating system reports. Verify that vCPUs are not being counted as physical processors.
Identify any duplicate system reporting and consolidate. If ILMT shows the same system reported multiple times, keep only one instance.
Review all detected IBM software and verify that instances are actually licensed (not evaluation licenses, free editions, or test deployments). Remove false positives.
Compare ILMT usage metrics against your understanding of actual system usage. If ILMT reports peak usage of 10,000 transactions per hour but you know average usage is 2,000 transactions per hour, the peak data is useful context but should not be the sole basis for licensing.
If IBM initiates an audit, ILMT data will be a primary source. Your response strategy should focus on three elements:
Ask IBM to provide detailed documentation of how ILMT was configured, which systems were measured, what measurement assumptions were made, and how peak vs. average usage was calculated. Most ILMT audits lack detailed methodology documentation — lack of documentation is a defense point.
Work with your IT team and a specialist advisor to challenge specific inaccuracies in ILMT data. Document each error with alternative measurements and explain why ILMT's numbers are incorrect. Accumulating 10–20 specific data point challenges significantly weakens IBM's audit position.
Suggest independent measurement by a third-party auditor rather than accepting ILMT as the sole source. Independent verification often identifies errors ILMT missed and provides credible alternative numbers for negotiation.
ILMT-based audit findings are negotiable. Key negotiation points:
Push IBM to acknowledge that peak usage is not representative of normal operating capacity and that licensing should be based on average usage. This can reduce findings 30–50% if peak usage is significantly higher than average.
If DR systems are included in audit findings, argue that passive systems used only for testing should not be licensed or should receive a significant discount. Most contracts do not require full licensing of inactive DR systems.
Correct any ILMT errors (duplicate systems, processor miscounts, etc.) and recalculate findings on the corrected data. This often reduces claims 15–25%.
Even if the PVU count cannot be challenged successfully, push for lower per-PVU pricing in the settlement. Settlements often include both unit count reduction and price reduction.
Whether you need to configure ILMT properly, validate data accuracy, or defend audit findings — specialist advisory expertise protects your compliance position and reduces audit exposure.