ZettaLane Blog - Cloud Storage Technical Insights & Deployment Guides

Unleashing 13 GB/s from GCS on a Single VM — The Power of ZFS on Object Storage

13 GB/s cold read from standard GCS object storage using ZFS on c4-highcpu-144 with 12 buckets and 150 Gbps Tier_1 networking. Verified by Google Cloud Monitoring.

10 min read • March 4, 2026

Read Article

Recent Articles

MayaNAS AWS

MayaNAS at OpenZFS Developer Summit 2025: Native Object Storage Integration with objbacker.io

We presented MayaNAS at OpenZFS Developer Summit 2025, introducing objbacker.io—a native ZFS VDEV for object storage. No FUSE, direct kernel-to-SDK integration achieving 3.7 GB/s throughput on S3/GCS/Azure.

10 min • December 27, 2025

MayaNAS GCP

MayaNAS on GCP: 8.14 GB/s NFS Read Throughput - Validated Performance

MayaNAS delivers 8.14 GB/s NFS read throughput on Google Cloud with Active-Active HA. Validated performance using n2-standard-48, 75 Gbps TIER_1 networking, and 20 GCS buckets with parallel I/O.

12 min • December 24, 2025

MayaScale GCP

MayaScale Validated Performance Tiers on Google Cloud

MayaScale delivers validated sub-millisecond performance across 5 performance tiers on GCP, from 75K to 2.3M IOPS. See how N2 instances with local SSDs enable ultra-low latency shared storage.

14 min • December 22, 2025

MayaScale

Simplicity Scales: MayaScale's Open Architecture Approach

MayaScale's open architecture uses proven technologies: Linux RAID-1, standard NVMe-oF, and infrastructure as code. No proprietary protocols, no inline features adding latency, no vendor lock-in.

7 min • December 20, 2025

MayaScale

Server-Side vs Client-Side Storage: Why Architecture Matters for Cloud Performance

MayaScale's server-side mirroring delivers true separation of concerns—zero client overhead, full CPU/memory for applications. Learn why disaggregated storage architecture outperforms client-side RAID approaches.

12 min • December 19, 2025

MayaScale AWS

MayaScale Delivers on AWS's Nitro-Powered i4i Instances

MayaScale delivers exceptional performance on AWS i4i instances with Nitro System v5—134μs read, 186μs write latency with shared storage. See how AWS's breakthrough platform enables ultra-low latency workloads.

8 min • December 19, 2025

MayaScale AWS

MayaScale Validated Performance Tiers on AWS

MayaScale delivers validated sub-millisecond performance across 4 performance tiers on AWS, from 204K to 1.35M IOPS. Choose the right price/performance balance for your workload.

12 min • December 19, 2025

MayaScale GCP

MayaScale Breaks the 2M IOPS Barrier on Google Cloud

We achieved 2.3 million IOPS with 192μs latency on Google Cloud Platform. Validated performance that transforms what's possible with cloud storage.

7 min • December 19, 2025

MayaScale GCP

MayaScale Achieves 2.3M IOPS on Google Cloud N2 Instances

MayaScale Ultra tier achieves 2.3M read IOPS and 192μs latency on Google Cloud's N2 instances. See how this performance breakthrough transforms local SSDs into ultra-low latency shared storage.

10 min • December 19, 2025

MayaScale Azure

Azure's NVMe Comeback: How Laosv4 Changed Everything

Azure's Laosv4 instances are a game-changer: even 4-core VMs get 2.88TB of NVMe. An opinionated look at how Azure's comeback is heating up cloud storage competition.

10 min • December 16, 2025

MayaScale Azure

MayaScale Delivers on Azure's Game-Changing Laosv4 Instances

MayaScale delivers exceptional performance on Azure's new Laosv4 instances—140μs read, 220μs write latency with shared storage. See how this game-changing platform enables ultra-low latency workloads.

8 min • December 16, 2025

MayaNAS

Beyond the FUSE Debate: Why Native ZFS on Object Storage Changes Everything

MinIO says filesystem on object store is a bad idea. JuiceFS disagrees. They're both missing the point. MayaNAS with objbacker.io isn't FUSE—it's native ZFS with kernel-level object storage integration.

12 min • December 26, 2024