Database High Availability Simplified

Why buy a storage array? Get FlashGrid software.

Turn standard NVMe PCIe SSDs inside database servers into scalable shared storage
0.4 to 153 TB per node,
2 to 16 nodes
Leverage proven Oracle ASM for high availability and data mirroring
Maximize database performance with FlashGrid Read‑Local™ Technology

12x performance at 1/6 of the cost of a flash array

Open Storage Designed for Oracle Clusters

FlashGrid Storage Fabric leverages proven Oracle ASM capabilities for volume management, data mirroring, and high availability. With FlashGrid managing storage is an easy task for DBAs and server administrators.

Reduce Database Licensing Costs

Accomplish more with fewer CPU cores. FlashGrid Read-Local™ Technology reduces CPU consumption and the unproductive I/O wait time by accessing flash storage through the highly optimized NVMe driver stack. The Remote Direct Memory Access technology eliminates the CPU overhead of the network stack and further reduces the I/O latency.

Standardize High Bandwidth and Low Latencies with NVMe PCIe SSDs

NVMe, the industry standard for PCIe SSDs, delivers outstanding performance of up to 3 GB/s and up to 750,000 random IOPS per SSD. Performance of a single NVMe SSD is similar to the performance of a typical all-flash array. Each server can have multiple NVMe SSDs, up to 48 SSDs in some server models.

Prevent Network Congestion with FlashGrid Read‑Local™ Technology

A congested storage network can limit the performance advantages of using flash storage. The FlashGrid Read‑Local™ Technology reduces network traffic by up to 90% or more by serving read requests from local SSDs instead of moving data over the network. Unlike caching, the FlashGrid Read-Local™ Technology accelerates even large table scans and backup operations.

Scale Large Clusters with 100 Gb/s InfiniBand Fabric

InfiniBand fabric provides 20 GB/s of network bandwidth per server. The storage bandwidth scales linearly as more nodes are added. The RDMA technology minimizes CPU cycles consumed for I/O by directly moving data between the nodes without using CPU.

Minimize Costs with Standard Hardware of Your Choice

Reduce storage costs by up to 90% compared to proprietary storage hardware. Pay only for the flash capacity that you actually need for a particular cluster, even if required capacity is as small as 1TB. Fully eliminate or cut the amount of the high-cost Fibre Channel infrastructure.