Average Ratings 0 Ratings
Average Ratings 0 Ratings
Description
NVIDIA Magnum IO serves as the framework for efficient and intelligent I/O in data centers operating in parallel. It enhances the capabilities of storage, networking, and communications across multiple nodes and GPUs to support crucial applications, including large language models, recommendation systems, imaging, simulation, and scientific research. By leveraging storage I/O, network I/O, in-network compute, and effective I/O management, Magnum IO streamlines and accelerates data movement, access, and management in complex multi-GPU, multi-node environments. It is compatible with NVIDIA CUDA-X libraries, optimizing performance across various NVIDIA GPU and networking hardware configurations to ensure maximum throughput with minimal latency. In systems employing multiple GPUs and nodes, the traditional reliance on slow CPUs with single-thread performance can hinder efficient data access from both local and remote storage solutions. To counter this, storage I/O acceleration allows GPUs to bypass the CPU and system memory, directly accessing remote storage through 8x 200 Gb/s NICs, which enables a remarkable achievement of up to 1.6 TB/s in raw storage bandwidth. This innovation significantly enhances the overall operational efficiency of data-intensive applications.
Description
Sangfor aStor represents an innovative software-defined storage solution that consolidates block, file, and object storage into a cohesive, elastically scalable resource pool, utilizing a fully symmetrical distributed architecture to facilitate on-demand provisioning of high-performance and cost-effective storage tiers tailored to various service needs. It can be deployed as either an integrated hardware-software system or as standalone software, with the ability to scale from a minimal setup of three commodity x86 nodes to expansive cloud-scale clusters comprising thousands of nodes, allowing for EB-level capacity growth. The system's multi-node parallel processing and intelligent caching mechanisms—including RDMA, SSD hot-data caching, and layering—achieve exceptional throughput, IOPS, and performance with small I/O operations, significantly enhancing cache hit rates to 90% and improving small I/O processing by as much as 65%. Additionally, its distributed metadata management ensures the seamless handling of billions of files without any significant latency, making it a robust solution for modern storage challenges. Overall, Sangfor aStor stands out as a versatile and powerful option for organizations looking to optimize their storage infrastructure.
API Access
Has API
API Access
Has API
Integrations
Amazon S3
Apache Spark
CUDA
Microsoft Hyper-V
NVIDIA NetQ
NVIDIA virtual GPU
OpenStack
Swift
VMware Cloud
Integrations
Amazon S3
Apache Spark
CUDA
Microsoft Hyper-V
NVIDIA NetQ
NVIDIA virtual GPU
OpenStack
Swift
VMware Cloud
Pricing Details
No price information available.
Free Trial
Free Version
Pricing Details
No price information available.
Free Trial
Free Version
Deployment
Web-Based
On-Premises
iPhone App
iPad App
Android App
Windows
Mac
Linux
Chromebook
Deployment
Web-Based
On-Premises
iPhone App
iPad App
Android App
Windows
Mac
Linux
Chromebook
Customer Support
Business Hours
Live Rep (24/7)
Online Support
Customer Support
Business Hours
Live Rep (24/7)
Online Support
Types of Training
Training Docs
Webinars
Live Training (Online)
In Person
Types of Training
Training Docs
Webinars
Live Training (Online)
In Person
Vendor Details
Company Name
NVIDIA
Founded
1993
Country
United States
Website
www.nvidia.com/en-us/data-center/magnum-io/
Vendor Details
Company Name
Sangfor
Founded
2000
Country
China
Website
www.sangfor.com/cloud-and-infrastructure/products/astor-enterprise-data-storage-solution
Product Features
Data Center Management
Audit Trail
Behavior-Based Acceleration
Cross Reference System
Device Auto Discovery
Diagnostic Testing
Import / Export Data
JCL Management
Multi-Platform
Multi-User
Power Management
Sarbanes-Oxley Compliance