From adf5953695db7ba096415ad3d531f43b7b61707a Mon Sep 17 00:00:00 2001 From: "mintlify[bot]" <109931778+mintlify[bot]@users.noreply.github.com> Date: Tue, 26 May 2026 23:00:35 +0000 Subject: [PATCH 1/2] docs: add benchmark link and NVMe footnote to managed postgres overview --- products/managed-postgres/overview.mdx | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/products/managed-postgres/overview.mdx b/products/managed-postgres/overview.mdx index b200ce93..4b7267a6 100644 --- a/products/managed-postgres/overview.mdx +++ b/products/managed-postgres/overview.mdx @@ -25,11 +25,15 @@ Most managed Postgres services use network-attached storage like Amazon EBS, whi Managed Postgres uses NVMe storage that is physically attached to the same server as your database. This architectural difference delivers: - **Microsecond-level disk latency** instead of milliseconds -- **Unlimited local IOPS** without network bottlenecks +- **Unlimited local IOPS**[*](https://docs.aws.amazon.com/ec2/latest/instancetypes/gp.html#gp_instance-store) without network bottlenecks - **Up to 10x faster performance** for disk-bound workloads at the same cost For Postgres workloads that are primarily throttled by disk IOPS and latency, this translates to faster ingestion, quicker vacuums, lower tail latency, and more predictable performance under load. +For detailed benchmark results, see [Performance benchmarks](/products/managed-postgres/benchmarks). + +For see [AWS local NVMe limits for details](https://docs.aws.amazon.com/ec2/latest/instancetypes/gp.html#gp_instance-store). + ## Native ClickHouse integration {#clickhouse-integration} Managed Postgres integrates natively with ClickHouse to bring transactions and analytics together without complex ETL pipelines. From 9938170243517dd475503dd6412f83e7c9336fee Mon Sep 17 00:00:00 2001 From: "mintlify[bot]" <109931778+mintlify[bot]@users.noreply.github.com> Date: Wed, 27 May 2026 16:09:07 +0000 Subject: [PATCH 2/2] docs: sync managed postgres NVMe links with merged source PR --- products/managed-postgres/overview.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/products/managed-postgres/overview.mdx b/products/managed-postgres/overview.mdx index 4b7267a6..621da142 100644 --- a/products/managed-postgres/overview.mdx +++ b/products/managed-postgres/overview.mdx @@ -25,14 +25,14 @@ Most managed Postgres services use network-attached storage like Amazon EBS, whi Managed Postgres uses NVMe storage that is physically attached to the same server as your database. This architectural difference delivers: - **Microsecond-level disk latency** instead of milliseconds -- **Unlimited local IOPS**[*](https://docs.aws.amazon.com/ec2/latest/instancetypes/gp.html#gp_instance-store) without network bottlenecks +- **Unlimited local IOPS**\* without network bottlenecks - **Up to 10x faster performance** for disk-bound workloads at the same cost For Postgres workloads that are primarily throttled by disk IOPS and latency, this translates to faster ingestion, quicker vacuums, lower tail latency, and more predictable performance under load. For detailed benchmark results, see [Performance benchmarks](/products/managed-postgres/benchmarks). -For see [AWS local NVMe limits for details](https://docs.aws.amazon.com/ec2/latest/instancetypes/gp.html#gp_instance-store). +For local NVMe limits on AWS see [Memory optimized](https://docs.aws.amazon.com/ec2/latest/instancetypes/mo.html#mo_instance-store), [Storage optimized](https://docs.aws.amazon.com/ec2/latest/instancetypes/so.html#so_instance-store), [CPU optimized](https://docs.aws.amazon.com/ec2/latest/instancetypes/gp.html#gp_instance-store). ## Native ClickHouse integration {#clickhouse-integration}