High-Performance
NVMe Block Storage.

Our block storage architecture completely bypasses legacy spinning hard disk drives (HDDs) and older SATA SSDs. Instead, we exclusively utilize enterprise-grade NVMe (Non-Volatile Memory Express) flash storage arrays. These arrays are connected over a dedicated, highly optimized Storage Area Network (SAN) utilizing NVMe over Fabrics (NVMe-oF) with RDMA (Remote Direct Memory Access). This allows your Virtual Machine's CPU to read from and write to the network-attached storage array almost as quickly as if the NVMe drive was plugged directly into the local PCIe slot, resulting in sub-millisecond latency and massive IOPS throughput.

Yes, scaling is incredibly seamless. As your database grows or your application requires more space, you can dynamically expand the capacity of your Block Storage volume directly from the Neviri dashboard or via our API without any downtime. Once the underlying volume is expanded on our storage clusters, you simply run a standard operating system command (like `resize2fs` for ext4) within your Virtual Machine to expand the file system into the newly available space. Note: To ensure data integrity, volumes can only be scaled up, not scaled down.

Absolutely. Data durability is our highest priority. Unlike local ephemeral storage which is tied to the lifespan of a single physical host node, Neviri Block Storage is inherently highly available. Every single block of data you write is automatically and synchronously replicated three times across disparate physical storage nodes within the availability zone. If a physical drive, a storage server, or even an entire rack experiences a catastrophic hardware failure, our system instantly seamlessly routes your I/O requests to the healthy replicas with zero perceived downtime or data loss.

Yes. Block Storage volumes are decoupled from the compute lifecycle. This means if you need to perform an OS upgrade, troubleshoot a kernel panic, or migrate a database to a larger compute flavor, you can safely unmount the volume, detach it from the current Virtual Machine, and attach it to a brand new one. The new VM will instantly recognize the disk, and all of your data, file systems, and directory structures will be perfectly preserved. This makes it an invaluable tool for stateless, highly agile infrastructure.

Snapshots provide a point-in-time, crash-consistent backup of your entire volume. When you trigger a snapshot, Neviri creates an instantaneous delta of the volume's current state. Because our storage utilizes advanced copy-on-write technology, taking a snapshot incurs zero performance penalty and takes only seconds, regardless of whether the volume is 10 GB or 5 TB. These snapshots are then securely transferred and stored in our highly durable Object Storage clusters. You can use snapshots to create exact clones of production databases for staging environments, or to instantly restore your system in the event of accidental data deletion.

Standard Block Storage volumes are designed for a 1-to-1 relationship; they can only be attached to a single Virtual Machine at a time to prevent data corruption caused by uncoordinated simultaneous writes by different operating systems. If your architecture requires a shared file system accessible by multiple nodes simultaneously (such as a WordPress `wp-content/uploads` directory shared across a load-balanced cluster), you should utilize our upcoming Managed File Storage (NFS) or architect your application to utilize Object Storage via the S3 API.

We believe in radical pricing transparency. Block Storage is billed strictly on a per-gigabyte, per-month basis, prorated down to the hour. You pay exactly for the capacity you provision, regardless of how much of that capacity you actually fill with data. Furthermore, we do not charge any hidden fees for I/O requests or data transfer between your VM and the attached storage volume. The price you see is the absolute maximum you will pay.