From 032642e9ad72923d5f247dc564c88b8c7e9df8f4 Mon Sep 17 00:00:00 2001 From: Jim Allen Wallace Date: Tue, 29 Sep 2026 17:06:01 -0600 Subject: [PATCH] Remove hard upper memory limits from FAQ and ClickHouse docs Replace the 768GB and 1TB single-instance figures with "multi-terabyte" and drop the sentence qualifying 1TB as not "big data". Per the 2026-09-15 GTM/Prod/Eng sync, docs should not state a hard upper memory limit. Co-Authored-By: Claude Opus 5.5 --- docs/about/faq.md | 2 +- docs/integrations/clickhouse.md | 3 +-- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/docs/about/faq.md b/docs/about/faq.md index 69a39380..b9ed77f1 100644 --- a/docs/about/faq.md +++ b/docs/about/faq.md @@ -18,7 +18,7 @@ For teams that want out-of-the-box scaling, security, and expert support, we lau ## How does Dragonfly's vertical scaling compare to a Redis cluster? -Dragonfly utilizes the underlying hardware in an optimal way. This means that it can run on small 8GB instances and scale vertically to large 768GB machines with 64 cores. This versatility allows for far less complexity as well as lower infrastructure costs when compared to running cluster workloads. In addition, Redis cluster-mode imposes some limitations on multi-key and transactional operations while Dragonfly provides the same semantics as a single node Redis. +Dragonfly utilizes the underlying hardware in an optimal way. This means that it can run on small 8GB instances and scale vertically to the largest multi-terabyte machines. This versatility allows for far less complexity as well as lower infrastructure costs when compared to running cluster workloads. In addition, Redis cluster-mode imposes some limitations on multi-key and transactional operations while Dragonfly provides the same semantics as a single node Redis. ## When will you support X command? diff --git a/docs/integrations/clickhouse.md b/docs/integrations/clickhouse.md index 762f37e9..151126bd 100644 --- a/docs/integrations/clickhouse.md +++ b/docs/integrations/clickhouse.md @@ -94,11 +94,10 @@ However, if range queries are required, ClickHouse can still utilize the `SCAN` - Point query example -- `WHERE key = xx` or `WHERE key IN (xx, yy)` - Range query example -- `WHERE key > xx` -Dragonfly is able to support up to 1TB on a single instance. +Dragonfly supports multi-terabyte datasets on a single instance. Redis, on the other hand, is typically constrained to handling tens of GB on an individual instance. Further, the use of Redis Cluster, which might seem like a solution, is off the table as it doesn't support the `SCAN` command in its cluster mode. Hence, for applications requiring expansive datasets and range queries, Dragonfly stands as the most effective and reliable choice for the key-value table engine when paired with ClickHouse. -We do understand that 1TB is probably not "big data" in modern terms, but supporting 1TB on a single instance is still an advantage and should be sufficient for many ad-hoc analytics use cases. ## Useful Resources