Press Release

FOR IMMEDIATE RELEASE                    28th August, 2026

The Hidden Energy Leak in Global Computing: How a 50-Year-Old Software Habit is Burning Power — and How to Fix It

WARRACKNABEAL, AUSTRALIA — As artificial intelligence, cloud computing, and massive data centres drive global energy demand to unprecedented highs, an open-source software project is drawing attention to a pervasive, invisible source of computational waste: the way computers handle everyday text.

The project, FastSafeStrings, demonstrates that standard programming conventions established in the early 1970s force computers to waste up to 90% of their processing effort simply searching for the end of words and sentences. By modernising this fundamental mechanism, FastSafeStrings delivers performance improvements of up to 10× to 18× on high-volume record processing while dramatically reducing the electrical energy consumed by servers and mainframes.

The "Reading with a Blindfold" Problem

Whenever a computer processes a line of text—such as an address, a financial transaction, or a database record—the widely used C programming language uses a convention called a "null-terminated string." Because the computer does not keep track of how long the text is, it must scan every single character one-by-one until it hits an invisible stop marker (\0).

When millions or billions of records are processed in enterprise workflows, this scanning happens repeatedly for the exact same data: once to read it, once to measure it, once to append it, and once to output it.

FastSafeStrings replaces this legacy habit with a descriptor-based approach (known as a "Dope Vector"). By attaching the length directly to the text, the CPU knows instantly how long the data is without redundant scanning.

In addition, Linux and Windows systems can read and write IBM Mainframe Variable Blocked files approximately 10 times faster than conventional Windows and Linux files.  These files can be read and written directly on IBM systems.

Key Real-World Benefits

• Slashing Computing Energy & Emissions: In enterprise benchmarks transforming 5,000,000 records, standard processing required up to 9 to 11 seconds of heavy CPU time. Under FastSafeStrings, the exact same task completed in roughly half a second—a massive reduction in CPU cycles that directly lowers server power draw, cooling demands, and carbon emissions.

• Built-in Cyber Immunity: Legacy string scanning is the root cause behind infamous buffer-overflow security bugs (including vulnerabilities like Heartbleed). FastSafeStrings builds strict capacity limits into every operation by default, providing an immediate upgrade path aligned with international cyber directives (such as the US CISA 2026 "Secure by Design" memory-safety roadmap).

• Cross-Platform & Drop-In Ready: The open-source library operates across Linux, Windows, and enterprise mainframes (IBM z/OS), allowing existing mission-critical systems to adopt it incrementally without requiring multi-million-dollar software rewrites.

FastSafeStrings is published under an open-source licence and is available for developers, enterprises, and researchers worldwide.

Project Links & Resources:

• Website / Overview: https://www.makingabeautifulworld.com/climate/fastsafestrings

• Open Source Repository & Benchmarks: https://github.com/clemcl/FastSafeStrings

• Media Contact: Clement Clarke — clemclarke@gmail.com