IBM has launched the IBM Storage Scale System 6000, a cloud-scale global data platform designed for data-intensive and artificial intelligence (AI) workloads. This new system is part of the IBM Storage for Data and AI portfolio, which offers advanced storage solutions for modern data environments.
Gartner, a leading research and advisory company, has recognized IBM’s expertise in distributed file systems and object storage. For the seventh year in a row, IBM has been named a leader in the 2022 Gartner Magic Quadrant for Distributed File Systems and Object Storage. This recognition emphasizes IBM’s commitment to delivering high-performance storage solutions tailored to today’s data-driven world.
Introducing the IBM Storage Scale System 6000
The IBM Storage Scale System 6000 is a high-performance system that provides up to 7 million IOPs and up to 256 GB/s throughput for read-only workloads per system in a 4U footprint. This system is designed to aggregate data from multiple sources in near real-time, optimizing performance for GPU workloads. It is particularly well-suited for storing semi-structured and unstructured data, such as video, imagery, text, and instrumentation data.
In the future, the IBM Storage Scale System 6000 will incorporate IBM FlashCore Modules (FCM) in the first half of 2024. This addition will bring capacity efficiency with a 70% lower cost and 53% less energy per terabyte. The system also includes powerful inline hardware-accelerated data compression and encryption to ensure data security.
Other articles you may find interesting about AI and related technologies:
5 misconceptions about artificial intelligence (AI)
Understanding Multimodal Artificial Intelligence (AI)
The Role of Artificial Intelligence in Antivirus Software
Apple plans to invest $4.75 billion in Artificial Intelligence in 2024
Is Artificial Intelligence a threat or a major breakthrough?
Get acquainted with conversational artificial intelligence
“With our current Storage Scale Systems 3500, we are assisting in reducing the time to discover and increasing research productivity in a variety of scientific disciplines. For AI research involving medical image analysis, we have reduced access latency by up to 60% compared to our previous storage infrastructure. For genomics and complex fluid dynamics workloads, we have increased throughput by up to 70%,” said Jake Carroll, Chief Technology Officer, Research Computing Centre, The University of Queensland, Australia. “We have all the advantages of a high-speed parallel file system inside our supercomputing resources with the data management transparency and global data access provided by the IBM Storage Scale software.”
Carroll added, “IBM’s Storage Scale System 6000 should be a game-changer for us. With the specifications I have seen, by doubling performance and improving efficiency, we will be able to ask our scientific research questions with higher throughput but at a lower TCO and lower power consumption per IOP.”
The IBM Storage Scale System 6000 is designed to be flexible and supports various multi-vendor storage options, including AWS, Azure, IBM Cloud, other public clouds, and IBM Storage Tape. This compatibility with different storage options makes it a versatile solution for various data storage needs.
When compared to its competitors, the IBM Storage Scale System 6000 offers impressive performance. It provides faster data access with more than 2.5 times the GB/s throughput and twice the IOPs performance of leading competitors in the market. This makes it a powerful tool for organizations that require fast and efficient access to large volumes of data.
The IBM Storage Scale System 6000 is already being used in practical applications. For example, the University of Queensland has utilized the IBM Storage Scale global data platform and IBM Storage Scale System for research on applied AI for neurodegenerative diseases and vaccine technologies. This demonstrates the system’s ability to support complex and data-intensive research projects.
The Storage Scale System 6000 also integrates with NVIDIA technology, supporting NVIDIA Magnum IOTM GPUDirect Storage (GDS). This provides a direct path between GPU memory and storage, designed to enhance performance with data movement IO when GDS is enabled.
The IBM Storage Scale System 6000 is a powerful, flexible, and high-performance storage solution created to meet the demands of data-intensive and AI workloads. With its impressive performance, potential for future expansion, and compatibility with various storage options, this system is well-equipped to support the data requirements of modern organizations.