Orisium DAM Tested on 100TB Live Media Archive: Four Days, Ten Years of Content, One Searchable Result
Orisium completed a four-day live test of its Enterprise DAM on over 100TB of localised media content spanning a decade, sorting people, transcriptions, images, and documents across multiple brands into a fully searchable, indexed library.
Orisium has completed a four-day live test of its Enterprise DAM on a localised server containing over 100 terabytes of media content spanning more than ten years of production. The test processed content across multiple brands simultaneously, filtering people through facial recognition and tagging, transcribing audio from thousands of video files, and indexing images, documents, and mixed media formats. The result was a fully sorted, brand-segmented, person-identified, timeline-organised, content-type-indexed library searchable in seconds.
100TB. Ten years. Four days. Searchable in seconds.
What 100TB of Unstructured Media Looks Like
One hundred terabytes of media content is not unusual for organisations that have been producing content for a decade or more. It represents hundreds of thousands of video files, millions of images, tens of thousands of documents, presentations, design files, and audio recordings. For most organisations, this content lives across a patchwork of hard drives, NAS devices, cloud storage accounts, and archival systems with little consistency in naming, tagging, or organisation.
The test server reflected this reality. Content had been collected from multiple production teams working for different brands over different time periods. Folder structures were inconsistent. File names ranged from descriptive to meaningless. Metadata was sparse where it existed at all. Duplicate files were scattered across directories. In short, it was a typical enterprise media archive: valuable content buried under years of organic, ungoverned accumulation.
The Four-Day Process
The DAM's processing pipeline worked through the archive methodically over four days:
- Day 1: Ingestion and deduplication. The system scanned the full archive, identified unique files, catalogued duplicates, and began extracting technical metadata from every asset. Over 340,000 unique media files were identified from approximately 520,000 total files on the server.
- Day 2: AI analysis and facial recognition. Computer vision models processed images and video frames, identifying faces, objects, locations, and scene types. The system built a facial recognition index linking specific individuals across thousands of assets from different years and different brands.
- Day 3: Transcription and document indexing. Audio tracks from video files were transcribed to searchable text. Documents, presentations, and PDFs were OCR-processed and full-text indexed. The system cross-referenced spoken names with identified faces to build richer people profiles.
- Day 4: Brand segmentation and final indexing. Content was automatically assigned to the correct brand based on visual brand markers, metadata clues, folder origin, and content context. The full search index was built, tested, and optimised. By end of day, the entire archive was queryable.
DAM at the Forefront of Brand Content and Marketing
For marketing teams, the ability to search and retrieve any asset instantly transforms how campaigns are built. Instead of commissioning new photography because nobody can find the shots from last year's event, teams can surface exactly the right image in seconds. Brand consistency improves because every team member has access to the same governed library of approved assets. Campaign turnaround times shrink because the content already exists. It just needs to be found.
For HR and corporate communications, the same principle applies. Training videos, onboarding materials, executive headshots, event footage, and internal communications content are all searchable by person, date, topic, or content type. When an executive moves to a new role, every piece of content featuring them is instantly accessible for updates, redistribution, or archival.
In the current content era, organisations do not suffer from a shortage of content. They suffer from an inability to find, reuse, and repurpose what they have already created. DAM turns dormant archives into active, revenue-generating content libraries.
Content Reuse: Making Buried Footage Live Again
One of the most striking outcomes of the 100TB test was the volume of high-quality content that had been effectively lost. Professionally produced video from events five or six years ago, fully usable in today's campaigns, was sitting untouched in nested folder structures that no current team member knew existed. Product photography from discontinued lines contained lifestyle shots perfectly suited for new campaigns. Interview footage with industry experts contained insights still relevant to current market positioning.
The DAM does not just organise content. It resurrects it. By making every asset discoverable, it gives organisations the ability to extract ongoing value from their entire content history, not just the assets created in the last quarter.
From Enterprise Archives to Athlete Careers
The same technology that processed 100TB of corporate media scales directly to the Sotnem.com platform, where athletes accumulate years of training footage, competition recordings, and coaching notes across their entire sporting careers. A 15-year-old karting driver who has been on Sotnem since their first race will, by the time they reach professional motorsport, have thousands of video files, hundreds of telemetry sessions, and years of coaching feedback stored in the platform.
The DAM's search and organisation capabilities ensure that any of those assets can be surfaced instantly. "Show me every wet-weather qualifying session from 2024" or "Find the race where I set my fastest sector 2 at Bathurst" become simple queries that return results in seconds, regardless of how much content has accumulated over the years.
For coaches working with new athletes, the ability to review a complete performance history, organised by date, track, condition, and result, provides context that would be impossible to assemble manually. The athlete's entire development journey is searchable, sortable, and immediately accessible.
What Comes Next
The 100TB test validated the DAM's ability to process, organise, and index content at a scale that mirrors real enterprise deployments. Orisium is now preparing for production rollouts with organisations across media and entertainment, corporate communications, and sports, each bringing their own archive sizes, brand structures, and search requirements. The four-day timeline demonstrated that even the largest archives can be transformed from chaotic storage into searchable intelligence in under a week.