From 91bf37b0c6880847a5c0b2e7d719cefc176f8aa1 Mon Sep 17 00:00:00 2001 From: TimLFletcher Date: Fri, 4 Sep 2026 13:57:31 +0100 Subject: [PATCH] [DOC-14444][AI] Add memory and XDCR criteria for choosing 1024 vBuckets Which Storage Engine Should You Use? did not mention two criteria that point at the 1024 vBucket configuration: having at least 1 GiB of memory per node available for the bucket, and needing XDCR replication from a cluster older than 8.0. Both are already documented elsewhere on the site. This surfaces them in the decision list where a reader making the choice will see them. Co-Authored-By: Claude Opus 5 --- .../pages/buckets-memory-and-storage/storage-engines.adoc | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/modules/learn/pages/buckets-memory-and-storage/storage-engines.adoc b/modules/learn/pages/buckets-memory-and-storage/storage-engines.adoc index 1589de28b1..04f5e78c1d 100644 --- a/modules/learn/pages/buckets-memory-and-storage/storage-engines.adoc +++ b/modules/learn/pages/buckets-memory-and-storage/storage-engines.adoc @@ -116,6 +116,11 @@ In this case, the 1024 vBucket configuration is less likely to have data imbalan Also, Magma with 1024 vBuckets uses less CPU for operations such as compaction verse the 128 vBucket configuration. * Your applications make heavy use of transactions with persistence-based durability. * Your dataset will grow beyond Couchstore's 3{nbsp}TB limit. +* You can allocate at least 1{nbsp}GiB of memory per node to the bucket. +The 1024 vBucket configuration gives better performance at scale. +* You need to replicate into the bucket by XDCR from a cluster running a version earlier than Couchbase Server 8.0. +Those versions require both buckets to have the same number of vBuckets, and do not support 128 vBuckets, so the target bucket must use 1024. +See xref:learn:clusters-and-availability/xdcr-overview.adoc[] for the full compatibility rules. When should you choose Couchstore?:: You should use the Couchstore backend if any of the following are true: