Delhi's e-governance infrastructure hit a practical wall this week when a routine audit of the Unified Digital Delhi portal flagged more than 40,000 duplicate image files spread across municipal and state government databases, slowing public-facing services and inflating storage costs at the Delhi Secretariat's central data centre in ITO. The cleanup operation, which began Monday, has already forced temporary outages on two citizen-service dashboards managed by the Delhi e-Governance Society.
The problem matters now because Phase 4 of the Delhi Metro expansion, combined with a push by the AAP government to digitise heritage records in Old Delhi's Shahjahanabad precinct, has driven a dramatic spike in the volume of photographs being uploaded to official servers. Construction progress images, environmental clearance documents with satellite snapshots, and Yamuna riverfront redevelopment files have all fed into a system that, until this week, had no automated deduplication protocol in place.
Where the Backlog Built Up
Ground zero for the redundancy problem is the Delhi Urban Shelter Improvement Board's document management system, which technicians at the Indraprastha Estate office identified as holding duplicate copies of more than 12,000 resettlement photographs — some files triplicated across three separate folders. The Delhi Pollution Control Committee's air quality monitoring archive, housed separately on servers in Lajpat Nagar, is also under review after investigators found identical sensor-location images uploaded repeatedly during the 2025 winter smog emergency response.
The Delhi Heritage Conservation Committee, which has been digitising records of protected structures in Mehrauli and Nizamuddin since early 2025, separately discovered that a contractor had been submitting the same before-and-after restoration photographs under different file names to meet monthly deliverable targets. That finding, confirmed this week by committee staff, has added a procurement compliance dimension to what began as a straightforward IT housekeeping exercise.
Officials at the Delhi e-Governance Society have not yet issued a formal public statement on the full scope of the audit. The organisation, which operates under the Department of Information Technology, confirmed through its public notice board that the Unified Digital Delhi portal's document upload module would be unavailable between 2 a.m. and 6 a.m. daily through July 11 as engineers run deduplication scripts.
What the Numbers Reveal
Storage costs are the sharpest argument for urgency. Delhi government cloud storage contracts, tendered through NIC India, are billed per gigabyte per month. Duplicate image files identified so far account for an estimated 2.3 terabytes of redundant data — a figure that, at standard NIC cloud rates, translates to a non-trivial recurring monthly overhead that auditors say has been accumulating since at least January 2024. Independent IT procurement analysts in Delhi who follow government tender documents note that image-heavy departments routinely underestimate storage bloat during portal design.
The deduplication drive is using an open-source hash-matching tool deployed by the National Informatics Centre, the same technology the Delhi government used in 2023 to clean up the Ration Card Management System's beneficiary photograph database ahead of the revised Public Distribution System rollout. That earlier exercise removed roughly 8,000 duplicate entries, according to figures published in the Delhi government's annual IT report for 2023–24.
Citizens who use the Unified Digital Delhi portal to track services — ranging from property mutation applications in areas like Rohini and Dwarka to birth certificate requests — are unlikely to notice disruption beyond the nightly maintenance windows. However, departments using the internal document management interface, including the Delhi Development Authority's planning cell at Vikas Minar, have been asked to pause bulk image uploads until July 12.
The practical advice from Delhi e-Governance Society technicians, according to the public notice posted this week, is straightforward: any department head who has been using manual workarounds to upload duplicate supporting images for pending applications should resubmit clean, single-copy files after July 12 when the new deduplication layer goes live. Going forward, the portal will automatically reject any image whose hash value matches an existing file already on the server — a measure that should prevent the same problem from compounding further as the Yamuna cleanup and Metro Phase 4 documentation keep growing.