Komprise Awarded Elastic Shares Patent to Accelerate AI Ingestion, Metadata Extraction & Data Mobilization for Unstructured Data

Komprise, a active player in analytics-driven unstructured data management, announces a new patent that addresses common issues with under-utilization of expensive resources such as GPU, memory and network when processing large unstructured datasets.GPU clusters, which are generally more efficient than CPUs, still show average utilization of only 50%, according to 2025 Cornell University research. Even during active jobs, GPUs are idle 14-76% of the time. As IT budgets are already stretched to accommodate 20%+ annual unstructured data growth and as CIOs face new AI infrastructure requirements, they need to tackle cost-efficiency more than ever today.

Typical load balancing partitioning provides one-time static balancing at the beginning of the job and therefore can’t account for uneven work. Instead, the Komprise Elastic Shares technology continually and dynamically reallocates resources to ensure near linear speed-up at scale.

This reduces waste, lowers costs and accelerates data mobilization tasks intrinsic to AI data workflows and ingestion, metadata extraction, data migrations, data tiering and sensitive data management.

Komprise Elastic Shares: Optimizing Expensive Data Center Resources
Komprise, with Elastic Shares, overcomes three limitations of load balancing approaches:

• Dynamic partitioning ensures expensive resources get assigned new tasks as soon as the resources become available
• Komprise can process datasets without prior knowledge of their size and structure, which is essential for data streaming to AI
• Komprise automatically rebalances resource allocation to address unstructured data hierarchies of unknown branch densities

The Elastic Shares patent (# US12566637B2) adds to the company’s growing portfolio of innovations and prior patents pertaining to unstructured data management:

• Transparent Move Technology (TMT) : This patented technology eliminates user disruption even when maximizing tiering across vendors and architectures. TMT cuts 70%+ costs and extends storage capacity while also avoiding the dreaded rehydration penalty
• Komprise Intelligent AI Ingest:  Ingestion engine delivers 2X-faster data flows to AI while lowering storage, transfer and compute costs
• Komprise Hypertransfer: This data migration innovation brings 25X faster migrations for SMB data
• Komprise Elastic Data Migration: Komprise delivers 27X faster migrations for NFS data

“The patents are validation of our focus in overcoming current limitations that keep organizations from fully harnessing the power of unstructured data with AI,” said Mike Peercy, CTO and co-founder, Komprise. “With Elastic Shares, our customers can fully utilize precious compute, memory, and network resources to gain near-linear scaling and the best competitive advantage.”