Infortrend Hybrid Flash Storage Powers AI and HPC with Speed and Scale at Lower Cost

Infortrend Hybrid Flash Storage Powers AI and HPC with Speed and Scale at Lower Cost

COMUNICATO STAMPA – CONTENUTO PROMOZIONALE

TAIPEI, Feb. 24, 2026 /PRNewswire/ — Infortrend Technology, Inc. (TWSE: 2495), a leading provider of enterprise storage and AI solutions, today announced that its flagship EonStor GS 5024U hybrid flash storage accelerates critical workloads and provides multi-PB capacity while saving costs, addressing the growing demands of AI training, high-performance computing (HPC), and intelligent surveillance. 

Bridging Performance and Scale for Modern Data Demands 

As data volumes continue to grow, enterprises in AI, HPC research, and surveillance analytics face a dilemma: These applications demand both high-speed access and massive capacity; however, scaling all-flash storage to a multi-petabyte level is costly, especially considering that peak performance is required only for a subset of data, while HDD-only systems cannot deliver the performance required for active workloads. Infortrend bridges this gap with the EonStor GS 5024U hybrid flash storage solution, featuring unmatched performance for active datasets and massive capacity for cold data at significantly lower cost. 

EonStor GS 5024U Storage: Flagship Performance with Multi-PB Capacity 

The EonStor GS family’s highest-performance unified storage system, EonStor GS 5024U, delivers up to 125 GB/s throughput and 2.4 million IOPS for performance-demanding workloads. It supports both all-flash and hybrid configurations. In a hybrid configuration, GS 5024U can scale up to 20 PB, providing a flexible platform for both hot and cold datasets. This hybrid setup and support for economical QLC SSDs help to lower the costs by over 50%[1] compared to all-flash deployments. These capabilities ensure that GS 5024U can efficiently meet the demands of large-scale, performance-critical data environments. 

Maximized Efficiency Across Applications 

Active datasets, such as AI training data, HPC computation inputs, or high-resolution surveillance streams, are stored on the NVMe SSD tier to ensure high-speed access and maintain compute or analytic efficiency. Less frequently accessed data and processed results can be automatically tiered to HDDs, reducing all-flash storage costs while maintaining accessibility for large-scale datasets. 

“GS 5024U combines high-speed access for active workloads with cost-efficient capacity for large datasets through a hybrid storage architecture. The solution can deliver up to 100 GB/s throughput using HDDs only, further optimizing costs for large-scale projects,” said Frank Lee, Senior Director of Product Planning at Infortrend Technology. 

Learn more: EonStor GS 5024U  

About Infortrend 

Infortrend (TWSE: 2495) has been developing and manufacturing storage solutions since 1993. With a strong emphasis on in-house design, testing, and manufacturing, Infortrend storage delivers performance and scalability with the latest standards, user-friendly data services, personal after-sales support, and unrivaled value. For more information, please visit www.infortrend.com
 

Infortrend® and EonStor® are trademarks or registered trademarks of Infortrend Technology, Inc.; other trademarks are the property of their respective owners. 

View original content:https://www.prnewswire.co.uk/news-releases/infortrend-hybrid-flash-storage-powers-ai-and-hpc-with-speed-and-scale-at-lower-cost-302687537.html
 

Copyright 2026 PR Newswire. All Rights Reserved. 

COMUNICATO STAMPA – CONTENUTO PROMOZIONALE: Immediapress è un servizio di diffusione di comunicati stampa in testo originale redatto direttamente dall’ente che lo emette. L’Adnkronos e Immediapress non sono responsabili per i contenuti dei comunicati trasmessi

immediapress

Hey, Ciao 👋
Piacere di conoscerti.

Iscriviti per ricevere i contenuti principali nella tua casella di posta, ogni settimana.

Non inviamo spam! Leggi la nostra Informativa sulla privacy per avere maggiori informazioni.

Comments

No comments yet. Why don’t you start the discussion?

Lascia un commento

Il tuo indirizzo email non sarà pubblicato. I campi obbligatori sono contrassegnati *