OCI Cache vs Dragonfly Cloud: choose the engine, then size sustained traffic

Choose OCI Cache if you want managed Valkey within an existing OCI platform. Dragonfly Cloud runs a separate, company-controlled engine whose future licensing is a dependency to weigh before adopting it. For traffic sizing, OCI ties bandwidth to memory, while Dragonfly’s sustained bandwidth can be much lower than its burst figure.

At a glance

About OCI Cache

OCI Cache is Oracle’s managed caching service in Oracle Cloud Infrastructure. It supports Valkey and Redis, with sharded and non-sharded deployments. Customers configure memory, nodes, and sharding, and applications connect through private endpoints using supported network routing.

About Dragonfly Cloud

Dragonfly Cloud is a managed service built on the Dragonfly engine. Dragonfly is a separate implementation with a Redis-compatible API. The service offers standalone datastores and sharded Swarm deployments, with client requirements that depend on the selected mode.

Recent updates

September 9, 2026 - OCI added cross-region replication. Regional recovery is now an option for eligible non-sharded clusters in selected region pairs, but your team still promotes the secondary and redirects applications.

September 2, 2026 - Momento introduced Cluster and Flex capacity for Momento Cache. Both provide dedicated, managed Valkey resources. Cluster exposes topology choices, while Flex selects topology within capacity bounds, giving teams a choice in how they configure capacity while Momento operates the cluster.

Choose who controls the engine’s future

OCI Cache runs Valkey in Oracle’s managed service. Prefer Valkey over Redis for its broad industry backing and continued advancement of the state of the art. Dragonfly is a separate project maintained by one company, which can change licensing for future releases.

Redis Inc.’s March 2024 licensing change illustrates the risk of depending on one company’s future terms. Momento’s RIP Redis article explores that history.

Size for sustained traffic and client connections

OCI allocates bandwidth as provisioned memory changes, so throughput can influence sizing even when the dataset fits. Oracle also documents up to 40,000 client connections per node, with excessive concurrency potentially increasing latency.

OCI’s private endpoints require suitable virtual cloud network routing. Clients connect through service hostnames, and sharded deployments require cluster-mode clients with hostname support. (OCI connection guidance)

Dragonfly’s AWS Standard 12.5 GB configuration lists a 0.5 Gbps baseline and 10 Gbps burst. The burst is temporary and not guaranteed. Size sustained traffic against the baseline for your chosen cloud, memory, and compute tier.

Plan the scaling path before deployment

OCI lets your team change memory per node, node counts, and sharding as capacity needs grow.

Dragonfly Swarm added managed horizontal scaling on February 19, 2025. Swarm requires cluster-aware clients, and cluster mode cannot change on an existing datastore.

Try Momento Cache

If you want to keep standalone clients as the cluster grows, evaluate Momento alongside the engine and bandwidth decision.

Momento Cache puts dedicated, high-performance Valkey behind a managed gateway that keeps client integration stable as your application grows. Choose Cluster for direct control over instance types, shards, and replicas, or Flex to set capacity and replication bounds while Momento selects the topology. Both include built-in autoscaling and fully managed cluster operations.

Talk to a Momento solution engineer to evaluate the capacity model and configuration that fit your workload.

Frequently asked questions

Does 99.95% mean the same thing for both services?

No. Oracle’s commitment measures qualifying external-connectivity availability. Dragonfly’s uptime objective averages monthly shard availability.

Can I resize a Dragonfly datastore after creating it?

Yes. You can resize memory and compute on an existing datastore. Cluster mode is the choice that stays fixed.

How does OCI recover across regions?

OCI’s asynchronous active-passive replication supports non-sharded clusters in selected region pairs, with at least 8 GB per node. One primary accepts writes, and the secondary is read-only. Regional recovery requires converting the secondary to standalone and redirecting applications. Participating clusters do not support backups, and recent unreplicated writes may be lost.