IBM Storage Scale System
In order to assist organisations in creating a global data platform that effectively manages file and object data for the most demanding artificial intelligence (AI), high-performance computing (HPC), analytics, and hybrid cloud workloads, IBM Storage Scale Systems are high-performance hardware platforms that run IBM Storage Scale software.
Large volumes of unstructured data, including as documents, audio, photos, videos, and other objects, can be managed quickly and easily by combining the capabilities of NVMe flash technology with IBM Storage Scale software.
Built on a massively parallel file system, Storage Scale System can be implemented on a variety of hardware platforms, such as virtual machines, ARM-based POSIX clients, IBM Power, IBM Z, x86, and Kubernetes. With support for NVIDIA GPUDirect Storage, it offers outstanding performance for GPU infrastructure.
IBM Storage Scale with Content Awareness
Discover your organisation data’s hidden worth
Their chatbots and other AI helpers have limited economic relevance since the huge language models that underpin them are trained using very little enterprise data. This problem is solved by IBM Storage Scale’s content-aware storage features, which uncover the semantic meaning concealed in unstructured data to enable AI assistants to produce more intelligent responses on their own.
IBM Storage Scale System 6000
Extreme performance is the goal of the Storage Scale System 6000. It has the following features:
- A single 4U node that maximises uptime by having redundant hardware and active-active controllers;
- Throughput of up to 330 gigabytes per second (GB/S) with little latency;
- With the NVMeoF performance tier, up to 13 million IOPS; and
- Maximum effective capacity of 3.4PBe in a typical 4U rack area.
The Storage Scale System 6000 is designed to expand to petabytes of capacity and thousands of nodes. It uses IBM Storage Scale RAID erasure coding, which offers data efficiency, reliable high performance, intelligent monitoring, and dynamic adjustment of the IBM Storage Scale system and data. It also mitigates storage hardware failures. Containerised software speeds up and streamlines the maintenance process by delivering updates and installations.
The NVIDIA GPUdirect Storage protocol, which allows a direct data channel between GPU memory and local or remote storage, such as NVMe or NVMe over Fabric (NVMe-oF), is supported by Storage Scale System 6000 to speed up the processing of globally distributed data.
The Storage Scale System 6000 has linearly scalable performance and employs a straightforward building-block methodology. In other words, a group of ten Storage Scale System 6000 systems can process more than 3 terabytes per second. Up to nine SAS hard disc drive expansion enclosures are supported.
IBM Storage Scale System 3500
Customers who need an entry-level or mid-level system for less demanding product scenarios might consider the IBM Storage Scale System 3500. With the exception of size and capacity, it shares many of the same features as the Storage Scale System 6000.
Advantages
Boost analytics and AI
Install high-performance hardware to handle several tasks at once.
Expand to a thousand nodes
For the most demanding applications, create a global data platform.
Connect and make all of your data more efficient
Use global data abstraction services to link disparate data sources and locales.
Features
Boost apps
Use TB/s parallel access to huge files to optimise data. For quicker access to small files, use the new turbo tier feature, which can perform up to tens of millions of operations per second.
Erasure Coding in IBM
Recover data faster in minutes as opposed to hours and with less space than data replication. For commodity servers, IBM Storage Scale Erasure Code Edition is a high-performance scale-out storage solution.
Data that is lockable and audit trails
Get access to a thorough record of all system actions and activities pertaining to data or reports.
Cybersecurity
With IBM Storage Scale protected copy, you can swiftly recover from a cyberattack and protect your file data.
Data reduction and tiering depending on policies
By releasing main storage, you can improve performance and move data between storage tiers.
International cooperation and sharing
Utilise sophisticated caching and AFM distributed routing to enable low-latency read and write access to data from any location in the world.
Energy efficiency combined with sustainability
With IBM Storage Scale System, you may meet sustainability objectives by adhering to IBM’s dedication to environmental leadership in all aspects of its business operations, including design, technology use, and operations.
