Storage, compute, and nodes together
Nirvana’s Silicon Valley data center (us-sva-2) is the first facility where we operate storage, dedicated compute, and nodes together in the same Tier III data center.
This deployment brings Nirvana-owned storage racks, dedicated compute infrastructure, and bare-metal nodes into one physical environment. By keeping these components colocated, we shorten data paths, reduce contention, and improve latency consistency for performance-sensitive Web3 workloads.
Instead of relying on shared racks, shared IOPS pools, or public network hops, us-sva-2 gives us tighter control over the physical layer - which translates into more predictable performance in production.
What’s inside us-sva-2?
us-sva-2 is designed around physical proximity and isolation:
- Bare-metal nodes running alongside virtualized workloads
- Dedicated compute infrastructure, colocated in the same facility
- Private storage racks powering Accelerated Block Storage (ABS)
- Private Connect, a private fibre interlink in between Nirvana cloud and any 3rd party cloud with zero egress (coming soon)
Silicon Valley remains a major infrastructure hub, with close proximity to chain hubs, exchanges, and peering partners. Operating fully within this environment allows workloads to stay physically close to the broader Web3 ecosystem while retaining cloud-style flexibility.
Accelerated Block Storage (ABS) is live
Accelerated Block Storage runs on Nirvana-owned private storage racks and is built for sustained, always-on blockchain workloads.
Unlike traditional cloud block storage, ABS assumes that high IOPS, low latency, and continuous writes are the normal operating state - not short-lived bursts. Performance remains stable under continuous load, without throttling, burst credits, or performance cliffs.
Key characteristics include:
- Predictable pricing, without performance-based surcharges
- Elastic scaling, from tens of gigabytes to multiple petabytes
- High availability, with volumes decoupled from individual machines
- Sustained performance under long-running workloads
When paired with Nirvana’s bare-metal compute, ABS benefits from node-level colocation, keeping physical data paths short and latency low.
A performance-first foundation
By bringing storage, compute, and nodes together in a single, tightly controlled environment, SVA2 improves latency and predictability at the infrastructure layer. But proximity alone isn’t enough.
Deterministic performance requires infrastructure designed around real blockchain workloads — from hardware selection to virtualization and orchestration. us-sva-2 represents the blueprint for how we’re building the Performance Cloud for Web3, with the same architecture rolling out to additional regions over time.
ABS is live today.
Ready to stay fast under load? Get in touch and we will set you up.
Nirvana: Modern Cloud for Real-time Workloads
High Clock Speeds Compute. Low latency Storage. Radically Cheaper Bandwidth.
Powering Web3, AI, and real-time systems.
Learn more at Nirvana Labs
Nirvana Cloud| Pricing | Blog | Docs | Twitter | Telegram | LinkedIn
Related Posts

Altinity × Nirvana: The joint stack behind the hot data for ClickHouse®
Nirvana is now a BYOC option on Altinity.Cloud. This is the architecture behind it: Altinity manages everything from the database down to Kubernetes, Nirvana operates NKS and the infrastructure beneath it, and the stack was benchmarked before it ever went GA. Here's the layer-by-layer breakdown.

GPU Starvation Is a Data-Path Problem, Not Just a Chip Problem
Nirvana's layer is high-performance block storage. ABS ships 20,000 baseline IOPS against AWS gp3's 3,000 default, bursting to 600,000, and in our own benchmarks runs 10.5x faster than gp3 on cold-read analytical queries.

Nirvana Labs July 2026 Roundup
Nirvana Agent Sandboxes are live. Nirvana is a BYOC option on Altinity.Cloud. Self-serve billing is on. And we published the most comprehensive agent sandbox comparison in the market. July was a big shipping month.

