Post

⚡ Introducing PlanetScale Metal ⚡ - Blazing fast NVMe drives - Unlimited IOPS - Drastic reductions in latency - And best price:performance ratio of any database available in AWS or GCP

Block, Intercom, Depot, and more have been running Metal in production for months and have recognized both performance gains and cost savings. See their stories here: Block - pscale.link/bl Depot - pscale.link/dep Intercom - pscale.link/int

Why Metal? Network-attached storage isn’t ideal for databases. Separating storage and compute creates inherent problems, most notably high and widely variable latency as well as lower throughput and higher costs.

NVMe changes everything. I/O latency is lower now because SSDs don’t need to seek and because the interconnects have gotten faster. Throughput is higher because the interconnects have higher bandwidth.

And that’s really the magic behind PlanetScale Metal: Hardware is really good now. The rest is PlanetScale doing everything it takes to let that hardware shine while upholding immaculate reliability. Read more about why we built it and how it works: pscale.link/why-metal

Durability through replication, backup, and restore is powerful. PlanetScale Metal databases utilize semi-synchronous, row-based, MySQL replication from a primary to two or more replicas distributed across three availability zones.