Skip to content

Fast Data tackles “data hunger” in the UAE and Saudi Arabia: chip idleness forces a restructuring of ultra-storage

Share
Fast Data tackles “data hunger” in the UAE and Saudi Arabia: chip idleness forces a restructuring of ultra-storage

Listen to this article

Read by Anchor

The Gulf race to acquire advanced graphics processors reveals a silent operational gap: while enterprises and cloud providers scramble to reserve high-performance computing capacity, these processors clash with the so-called “data hunger” phenomenon, as costly chips sit idle because traditional storage media cannot feed them unstructured data quickly enough. In this context, US-based Fast Data is intensifying its moves in the UAE and Saudi markets to shift from a storage provider to what it calls an artificial-intelligence operating system, drawing on its global support for more than three million GPUs and its strategic partnership with Core 42, a unit of G 42 in Abu Dhabi.

Danny Khouri, regional channel director for Fast Data Middle East, told the AI Everything forum in Abu Dhabi that owning graphics-processor infrastructure alone does not guarantee economic return if compute cycles remain idle due to slow back-end data platforms. He noted that the system, whose regional presence was launched by Haider Aziz, started early from an “universal storage” model to integrate unstructured data, the main nerve of AI applications, before evolving its software platform to manage the full data lifecycle from ingestion to insight extraction and serving swarms of intelligent agents.

These moves coincide with a growing focus on digital sovereignty and sovereign computing in the region, a trajectory that spans processors, networks, data layers and governance frameworks. To overcome the dilemma of moving sensitive data to public clouds, the company, together with Nvidia, developed the Data Enclave solution, which relies on confidential computing, hardware isolation and cryptographic proof. The technology lets organisations run advanced models in trusted local environments, keeping full control of encryption keys and data within their geographic borders, while model developers retain protection of their proprietary weights.

Limiting competition to processor procurement without addressing data-transfer bottlenecks turns billion-dollar investments into idle assets that drain operating budgets.Although regional demand remains largely driven by government initiatives and new cloud providers, the data indicate an approaching critical inflection point where organisations move from exploratory pilots to full-scale production. In Saudi Arabia, the company has opened a local office and is holding talks with new customers, leveraging its existing client base through its original manufacturing partnership with Hewlett Packard Enterprise to support sovereign-computing environments alongside accelerated compute-center projects.

This shift in computing architecture has direct implications for technology leaders and digital-transformation teams in banks and large enterprises across the Gulf, Egypt and the wider region. Building AI capabilities is no longer just about renting compute hours or buying processing cards; it requires a radical restructuring of isolated data warehouses across departments into a single ultra-fast access platform.Today's operational bet is shifting toward reducing the time to extract value and securing delegated work environments that can manage sensitive data without breaching national regulations.This reality forces organisations to revisit their capital-spending plans, moving team skills from merely scripting commands to engineering low-latency data flows and encrypting them before committing to long-term compute contracts.

Don't miss the next story

Subscribe for updates