100% SDS, leveraging standard servers and next-generation storage & networking components
Pools NVMe storage across a network at local speeds and latencies
Enables 100% converged infrastructure
Deployed as a virtual, distributed non-volatile array and supports both converged and disaggregated architectures
Client-Side, Distributed Architecture
Enables NVMe sharing that scales performance linearly at near 100% efficiency
Local Performance at Data Center Scale
Remote IOPS = Local IOPS
Remote BW = Local BW
Remote Lat. = Local Lat. + ~5us
Capacity-optimized platform with 24 NVMe drives delivers:
> 4.9M 4KB IOPS
< 200us of latency
20 servers, shared data, > 99% efficiency
128 servers @ NASA > 140 GB/s write throughput (bound by NVMe devices)
Using RDDA, 0% target CPU usage
Scales linearly to 100s of millions of IOPs per rack
Scale & Performance
- Leverage the full performance of your NVMe flash at scale, over the network.
- Predictable application performance – no noisy neighbors.
- Scale your performance and capacity linearly.
- Leverage high IOPS, high bandwidth or mixed.
- Maximize the utilization of your NVMe flash devices.
- No data localization.
- Easy to manage & monitor, reduces the maintenance TCO.
- Mix different storage media types to optimize for cost, scale or performance.
- Use native NVMesh and/or NVMf transports to future-proof your environment.
- Utilize products from any server, storage and network vendor, no vendor lock-in.
- Scale storage and compute separately.