Topic

Directory Buckets / Hot Object Storage

A purpose-built storage tier designed for single-digit millisecond latency, using a directory-based namespace within a single Availability Zone. Trades multi-AZ durability for consistently low access times.

6 connections 3 resources

Summary

What it is

Where it fits

Directory Buckets represent the high-performance end of S3 storage. They fill the gap between standard S3 (high durability, variable latency) and local disk (low latency, no durability), enabling latency-sensitive workloads like ML training and real-time analytics to use object storage.

Misconceptions / Traps

Directory Buckets are single-AZ only. Data is not replicated across AZs, so they are not suitable as a sole durable store for critical data.
The directory-based namespace is not the same as a filesystem. It enables faster listing within directory structures but does not provide POSIX semantics.

Key Connections

scoped_to S3, Object Storage — a specialized storage tier within the S3 ecosystem
S3 Express One Zone implements Directory Buckets / Hot Object Storage — the AWS implementation
solves Cold Scan Latency — single-digit ms access eliminates cold-start overhead
constrained_by Vendor Lock-In — currently an AWS-specific feature

Definition

What it is

Purpose-built storage tier for single-digit millisecond latency, using directory-based namespace within a single Availability Zone. Trades multi-AZ durability for dramatically reduced latency.

Why it exists

Standard S3 has ~50-100ms per-request latency. AI/ML inference, real-time analytics, and interactive workloads need lower latency. Directory buckets co-locate data with compute for speed.

Recent developments

Latest signals

2 million requests per second per directory bucket. Directory buckets support up to 2M GET + 200K PUT transactions/second per bucket — an order of magnitude above S3 Standard's per-prefix limits. Per AWS docs — Directory bucket high performance workloads.
Directory buckets are the substrate for Amazon S3 Files (April 2026 GA). S3 Files implements an intelligent hot/cold storage pattern on top of directory buckets — actively-used data cached for ~1ms latencies, larger sequential reads served from the underlying S3 bucket. Per InfoQ — AWS Introduces S3 Files.
Lowest-latency cloud object storage class today — 10× faster than S3 Standard. S3 Express One Zone (which lives in directory buckets) delivers single-digit-millisecond latency, 10× faster data access vs S3 Standard, and request costs 50% lower than S3 Standard. Per AWS — Object Storage Classes.
No S3 Lifecycle transitions — by design. Directory buckets only allow objects stored in the S3 Express One Zone storage class and do not support S3 Lifecycle transitions. The architectural commitment: this is a hot tier; cold data lives in different buckets. Per AWS — S3 Express One Zone storage class.
S3 Files: aggregate throughput in multi-TB/sec range. S3 Files caches actively-used data for low-latency access and provides up to multiple terabytes per second of aggregate read throughput — "storage never limits performance" for the hot-tier slice of the workload. Per Tutorials Dojo — Amazon S3 Files.