Valkey 9.0 Delivers Atomic Slot Migration and Scales to 1 Billion Requests Per Second

Valkey 9.0 Introduces Multi-Database Clustering, Atomic Slot Migration, and Major Performance Gains

The Linux Foundation announced the general availability of Valkey 9.0, an open-source in-memory storage solution built as a successor to Redis, achieving a throughput of over 1 billion requests per second. This release introduces atomic slot migrations, hash field expiration, and full support for numbered databases in cluster mode, enabling scaling to 2,000 nodes.

Traditional cluster rebalancing often involves stepwise migrations, risking data inconsistency during transfer; Valkey 9.0’s atomic slot migrations ensure consistent key routing and predictable handoffs, addressing a common operational challenge in distributed caching systems. This reduces transient errors and simplifies live resharding.

Why This Matters

Ideal caching models assume perfect data distribution and zero downtime during scaling. In reality, rebalancing data in a distributed cache can be complex and error-prone, leading to service disruptions and data loss. The cost of these failures can range from degraded performance to complete application outages, especially for latency-sensitive applications.

Key Insights

• Atomic Slot Migration: Ensures consistent key routing during cluster rebalancing.
• Hash Field Expiration: Allows independent expiration of fields within Valkey hashes, improving memory management.
• Numbered Databases: Enables scalable, multi-database deployments within a cluster, providing logical data separation.

Working Example

# Example of setting and expiring a hash field in Valkey 9.0
# (Requires a Valkey client library - conceptual example)

import valkey

client = valkey.Client(host='localhost', port=6379)

client.hset("user:123", "name", "Alice")
client.hset("user:123", "email", "[email protected]")

# Expire the 'email' field after 60 seconds
client.hset_expire("user:123", "email", 60)

# Check if the field exists
print(client.hexists("user:123", "email")) # Output: True

# Wait for 61 seconds
time.sleep(61)

# Check again
print(client.hexists("user:123", "email")) # Output: False

Practical Applications

• E-commerce: Stripe could leverage atomic slot migrations to seamlessly scale their caching infrastructure during peak shopping seasons.
• Pitfall: Using whole-key expiration for individual fields in a hash can lead to inefficient memory usage and increased latency. Valkey 9.0’s field-level expiration mitigates this.

References:

• https://www.infoq.com/news/2025/11/valkey-9-atomic-migration/