Advanced Computing in the Age of AI | Monday, June 24, 2024

VAST Data and Arista Networks Partner to Simplify Scaling AI Workloads for Enterprise AI Infrastructure 

May 28, 2024 -- VAST Data today announced a collaboration with Arista Networks, a leading provider of cloud networking solutions, to offer optimized AI infrastructure combining low latency, high-bandwidth lossless ethernet switching with a performant, secure data platform that can seamlessly scale to meet the needs of modern enterprises and large GPU cloud service providers.

Enterprises are increasingly implementing AI for business-critical tasks, highlighting the need for extremely reliable, secure, and high-performance infrastructure that can be deployed rapidly and seamlessly scale as AI initiatives expand. The VAST Data Platform's innovative Disaggregated Shared-Everything (DASE) architecture, paired with Arista Networks' high-speed, low-latency networking ensures enterprises maintain resilient data access for optimized GPU-accelerated computing.

“As enterprises continue to adopt and integrate Gen-AI across their businesses, modern AI applications require an accelerated computing network that can interconnect hundreds and thousands of GPUs, with fast access to massive unstructured data sets distributed across the globe,” said Martin Hull, VP/GM Cloud and AI Networking. “We are consolidating and optimizing AI infrastructure to empower enterprises to unlock the full potential of AI.”

Key benefits for joint customers of Arista Networks and VAST Data include:

  • Simple, Robust Enterprise Functionality: VAST and Arista’s network and data management composability enables a best-of-breed approach to resilient, easy to manage, multi-tenant, scalable, high-performance, and Zero-Trust based AI infrastructure that meets strict enterprise security and compliance standards.
  • Efficient Infrastructure for AI Workloads: The VAST Data Platform optimizes organizations’ data sets using proprietary Similarity data reduction and compression capabilities that help significantly reduce power consumption and improve efficiency. Paired with Arista’s data-driven, client-to-cloud networking for AI environments, businesses benefit from efficient networking infrastructure built to support their AI workloads.
  • End-to-End, Full Stack Observability: Ensuring that data is flowing from edge-to-core-to-cloud in order to meet the needs of data-hungry algorithms and services is a critical job that can be easily managed using Arista and VAST Data’s complementary observability tools.
  • Flexible Scalability: With VAST and Arista, organizations have the flexibility to scale as their AI and data needs grow, offering low latency, high-bandwidth lossless ethernet switching, and disaggregated infrastructure that can scale storage and performance independently as their business requires.

“Today’s new AI workloads require a new type of data infrastructure, built on top of a radically different architecture than before,” said John Mao, Vice President, Strategic Alliances at VAST Data. “With VAST and Arista, organizations can consolidate enterprise and AI workloads onto a unified, multi-tenant infrastructure designed to streamline their AI data pipelines, simplify management, and reduce complexity - allowing them to focus on deriving value from their entire data set and accelerate innovation.”

As part of this partnership, VAST Data has qualified select Arista Ethernet switches for use in production deployments at scale, today.

About VAST Data

VAST Data is the data platform company built for the AI era. As the new standard for enterprise AI infrastructure, organizations trust the VAST Data Platform to serve their most data-intensive computing needs. VAST Data empowers enterprises to unlock the full potential of their data by providing AI infrastructure that is simple, scalable, and architected from the ground up to power deep learning and GPU-accelerated data centers and clouds. Launched in 2019, VAST Data is the fastest growing data infrastructure company in history.


Source: VAST Data

EnterpriseAI