Quantum has announced enhancements to its ActiveScale object storage platform, focusing on accelerated access to cold data in Glacier-class archive tiers. The company reports that these updates enable tape-based archives to function as responsive and query-ready data lakes, specifically designed for artificial intelligence (AI), analytics, and high-performance computing use cases at exabyte scale.
Central to the update is the new Ranged Restore feature that allows data center operators and other customers to restore only the necessary byte ranges from large objects stored in tape-based Glacier-class tiers. Previously, restoring cold data could require retrieving full objects—often hundreds of gigabytes—resulting in long wait times and increased operational costs. Quantum claims its ActiveScale platform now enables selective restore of data segments, reducing retrieval times, computational load, and egress overhead. Quantum is the only vendor currently offering this customized Amazon Simple Storage Service (S3) Glacier extension for tape-based cold storage, according to the announcement.
The updated platform also introduces a redesigned restore engine that delivers more than five times faster throughput for small object restores compared to prior versions. According to Quantum, these performance gains are achieved by batching and intelligently ordering restore requests, which can significantly speed up high-volume retrieval workflows. AI model training pipelines, data validation jobs, compliance lookups, and pipeline-driven restores are cited among scenarios that benefit directly from these improvements.
Geoff Barrall, chief product officer of Quantum, stated, “Cold data is no longer offline data,” adding, “By eliminating the legacy limits of Glacier-class archives, ActiveScale turns tape into an active asset — fast, API-accessible, and ready for AI and analytics at scale.”
The new ActiveScale enhancements are available immediately. The company targets enterprises operating large, archive-driven data centers and other environments with demanding AI and analytics workloads.
Source: Quantum
