Arelle Performance Opimization

[Mike D]

Hi, all,

I generally use Arelle to validate DORA Register of Information packages.  While our register is not per se complex (about 200 vendors and 400 contracts), it takes a LONG time to validate (on the order of several days), even on a dedicated VM that isn't doing much else.  

This is running on a Windows 11 machine via the CLI, though the performance on the UI is similar.  I run a single command to load my local versions of the taxonomy packages, load my target file, and validate.  

Wondering if someone can help me understand either what environment specs I can tweak to shorten the runtime, or if they have any advice on whether there are any config flags I can pass in that will optimize the resources.  

Currently running on a dedicated VM (VMWare 7.1) with

• Win 11 Enterprise
• Intel Zeon Gold 6342 CUP @ 2.8Ghz 2.79Ghz (2 processors)
• 32.0 GB Installed Ram

What I'm observing via Task Manager is that I'm currently consuming about 25-26% of processor capacity, and about 2.5 GB of RAM.  This is nowhere close to maxing our resources.  

I believe I've been told in the past that this runs single threaded, so maybe not a surprise it's not maxing the CPU, but somewhat surprised the memory use is so low compared to what's available (70% of RAM is currently reporting unused via TaskManager).  Network and disk usage on this VM are minimal, and nothing else is really running. 

Any tips appreciated.  

Mike

[Austin Matherne]

Hi Mike,

Thanks for the detailed context. You're correct that Arelle is single-threaded, so the ~25% CPU usage you're seeing is consistent with maxing out one core. Adding more cores or RAM won't improve single-instance validation times if the bottleneck is compute.

Arelle has an option to surface what operations are consuming the most time (`--collectProfileStats`). This won't speed things up directly, but if a small number of formula rules account for most of the runtime, we may be able to optimize them.

Happy to dig into specifics if you want to share the specific validation command you're running along with the output from `--collectProfileStats`.

Kind regards,
Austin Matherne

[Sam Ray, CPA US, Dip IFRS ACCA UK, SAP Certified]

Guys, it’s fascinating — and slightly hilarious — how much complexity one must navigate in the Age of AI.

 I’ve been around since the early XBRL days, working with Charlie Hoffman and Louis Matherne back in 2008. 

Now here we are, deep in the AI era, and for a fleeting moment I found myself wondering whether XBRL has quietly passed its expiration date(:-

… like a bag of coffee beans with a ‘Best Before’ label.

 Turns out even standards can make you check the pantry.

--

---
You received this message because you are subscribed to the Google Groups "Arelle-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to arelle-users...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/arelle-users/896bcef5-2e89-483f-a37c-e1e5cdca624an%40googlegroups.com.

[Sam Ray, CPA US, Dip IFRS ACCA UK, SAP Certified]

I asked Microsoft CoPilot this question

UNDER US GAAP OR IFRS FOR DEPRECIATION WHAT IS THE USEFUL LIFE OF A TECHNOLOGY FOR FINANCIAL REPORTING COMPLIANCE TO REGULATORY AGENCIES OF A COUNTRY

This is the answer:

You’re asking a very precise accounting question, so let’s cut straight to the authoritative answer:

📘 US GAAP and IFRS do not prescribe a specific useful life for “technology used for financial‑reporting compliance.”

Neither framework provides a table or mandated number of years for this type of asset.
Instead, both require management to estimate the useful life based on expected economic benefit, obsolescence, and regulatory factors.

Below is the clean, authoritative breakdown.

🧩 1. Under US GAAP

Relevant Standards

• ASC 360 – Property, Plant & Equipment
• ASC 350 – Intangibles (including software)

Key Principle

Useful life must reflect:

• Expected use
• Technological obsolescence
• Regulatory or legal limits
• Vendor support lifecycle
• Integration with other systems

Common practice (not mandated)

For financial‑reporting or compliance technology:

• 3–7 years is typical in practice
• 3–5 years for software
• 5–7 years for more stable on‑prem systems

But these are industry norms, not GAAP rules.

🧩 2. Under IFRS

Relevant Standard

• IAS 38 – Intangible Assets

Key Principle

Useful life is:

• Finite → amortize over the period of expected benefit
• Indefinite → no amortization, but annual impairment testing

IAS 38 requires considering:

• Expected usage
• Technical, technological, or regulatory obsolescence
• Stability of the regulatory environment
• Vendor support and upgrade cycles

Common practice

Similar to US GAAP:

• 3–7 years for compliance/reporting technology
• Shorter if regulatory rules change frequently
• Longer if the system is stable and vendor‑supported for many years

🧭 3. What This Means for “Technology for Financial‑Reporting Compliance”

This category includes:

• XBRL filing systems
• Regulatory reporting engines
• Financial close & compliance platforms
• Country‑specific statutory reporting tools

These are typically intangible assets (unless hardware is involved).

Useful life is determined by:

• How often the country’s regulations change
• How often the vendor updates or retires versions
• Whether the system is cloud‑based (SaaS → usually expensed, not depreciated)
• Integration with ERP upgrade cycles
• Expected economic benefit period

🎯 4. The Practical Answer

US GAAP:

No prescribed useful life. Management estimates it.
Common range: 3–7 years.

IFRS:

No prescribed useful life. Management estimates it.
Common range: 3–7 years.

Why no fixed number?

Because regulatory technology varies widely by:

• Country
• Vendor
• Deployment model
• Regulatory stability
• Technological pace

[Herman Fischer]

Sam,

Your question is very timely and exactly the motivation for the OIM Working Group effort to define what started as the “OIM Taxonomy” and is now an XBRL Model for moving ahead.  We’ve started with what is being modeled in current taxonomies, how well the current syntax supports the reporting and consumption communities and how consumable it is by AI.  

While the "coffee beans" noted below have passed their expiration date, those beans (Xlink, XML Schema, inconsistent modeling approaches) carried important semantics which can be modeled and represented in ways that will be usable for the future.

Please join us in the various communities on this effort.  XII has the OIM Working Group.  XBRL.US has XTAC. 

   Herm Fischer

   Chair, OIM Working Group

From: arelle...@googlegroups.com <arelle...@googlegroups.com> on behalf of Sam Ray, CPA US, Dip IFRS ACCA UK, SAP Certified <somnath...@u.northwestern.edu>
Date: Sunday, February 1, 2026 at 11:02 AM
To: arelle...@googlegroups.com <arelle...@googlegroups.com>
Subject: Re: Arelle Performance Opimization

Be Careful With This Message From ("Sam Ray, CPA US, Dip IFRS ACCA UK, SAP Certified" <somnath...@u.northwestern.edu>) Learn More Potential Impersonation The sender's identity could not be verified and someone may be impersonating the sender. Take caution when interacting with this message.

To view this discussion visit https://groups.google.com/d/msgid/arelle-users/CAML%2BC4jdECbqtdLaZ0bLNoZQU9oWUT%3Dr1T-YSxERkb%3DSbaqzbA%40mail.gmail.com.