Gaming industry under DDoS attack. Get DDoS protection now. Start onboarding

Products

  1. Home
  2. Blog
  3. Introducing FAST Object Storage: low-latency, S3-compatible storage built for AI workloads
AI
News

Introducing FAST Object Storage: low-latency, S3-compatible storage built for AI workloads

  • April 20, 2026
  • 2 min read
Introducing FAST Object Storage: low-latency, S3-compatible storage built for AI workloads

We're launching FAST, a new S3-compatible Object Storage type purpose-built for performance-intensive and AI workloads. It's built on VAST Data's industry-leading, all-flash storage platform, purpose-designed for high-throughput, low-latency data access at scale.

 

Standard object storage was designed for scale and cost-efficiency, and works effectively for a wide range of workloads such as backups and archives. But if you're running AI pipelines, ingesting IoT telemetry at high object rates, or serving latency-sensitive applications, waiting 200+ ms for a small file read can become a noticeable bottleneck.

 

Gcore FAST Object Storage removes that bottleneck. It combines lower latency, higher throughput, and the bandwidth to handle massive traffic from GPU clusters, all with rock-solid consistency. Because FAST is deployed exactly where our GPU infrastructure lives, it's a natural companion for AI pipelines that need high-speed data access close to compute.

 

FAST is already live in our AI regions in Portugal (Sines-2, Sines-3) and the UK (London-2).

Performance metrics

FAST delivers a dramatic performance leap over Standard (CEPH) storage, particularly for small-file and latency-sensitive workloads.

 

MetricImprovement
TTFB (small objects)~35× faster
Small-file throughputUp to 27× faster
Mixed workload throughput2.2× faster
Large-file throughput1.9× faster

 

Beyond raw speed, FAST shows rock-solid consistency with minimal variance between peak and sustained throughput. No spikes, no surprises.

Who should use FAST

  • AI/ML teams. Training datasets, embedding stores, and massive small-file collections are exactly what FAST was built for. Paired with Gcore's GPU infrastructure in the same locations, you get a tight data-to-compute loop without network hops across regions.
  • IoT and telemetry platforms. High object-rate ingestion is where standard storage struggles most and FAST shines brightest.
  • Any latency-sensitive S3 workload in Southern Europe or the UK. If your application calls for single-digit-millisecond first-byte latency and you're operating in or near these regions, FAST is a straightforward upgrade.

Pricing

FAST pricing is consistent across all three locations.

 

 EURUSD
Storage€0.06364/GiB/mo$0.07/GiB/mo
Egress€0.0013/GB$0.00143/GB
IngressFreeFree
Requests€0.03/10K$0.033/10K

Get started

FAST Object Storage is available now. It's fully S3-compatible, so if your tooling already speaks S3, migration is easy. Simply point your client at a new endpoint and go.

 

To provision FAST Object Storage, log in to the Gcore Customer Portal or reach out if you need a hand getting set up.

 

Try FAST Object Storage

Table of contents

Try Gcore AI

Gcore all-in-one platform: cloud, AI, CDN, security, and other infrastructure services.

Related articles

Mission Space chooses European sovereignty: why the Luxembourg space startup moved to Gcore

An interview with Alexey Shirobokov, CEO & Founder of Mission Space with Dima Maslennikov, Head of Startups at Gcore, recorded at House of Startups, Luxembourg. At Gcore, we work closely with startups building at the edge of deep t

Introducing GPU VMs on NVIDIA AI infrastructure in Sines (EU): flexible, cost-efficient compute for AI workloads

Some AI jobs require the full power and predictability of dedicated bare metal clusters. Others need something more agile: compute that can be sized up or down quickly, used for a burst of experimentation, powered down when idle, and spun b

Gcore Everywhere AI evolves to full-lifecycle management with Slurm, Jupyter, and token-based inference integrations

AI adoption has a fragmentation problem. Organizations routinely stitch together separate tools for development, training, and serving, each with its own infrastructure, access controls, and operational overhead. The result is a patchwork t

Introducing faster, lower-cost LLM inference with NVIDIA Dynamo

Imagine if you could click a button and suddenly your GPUs increase their throughput by 6x. Or reduce latency by 2x. Or route inference requests seamlessly across different GPU types.That's the experience we're bringing to our inference cus

New AI inference models on Application Catalog: translation, agents, and flagship reasoning

We’ve expanded our AI inference Application Catalog with three new state-of-the-art models, covering massively multilingual translation, efficient agentic workflows, and high-end reasoning. All models are live today via Everywhere Inference

New AI inference models available now on Gcore

We’ve expanded our Application Catalog with a new set of high-performance models across embeddings, text-to-speech, multimodal LLMs, and safety. All models are live today via Everywhere Inference and Everywhere AI, and are ready to deploy i

Subscribe to our newsletter

Get the latest industry trends, exclusive insights, and Gcore updates delivered straight to your inbox.