Q: Your application writes millions of small files to S3. Performance is slow on listing and retrieval. How do you optimize?
S3 automatically partitions by prefix — requests are distributed across S3 partitions. More unique prefixes = better parallelism. Add a hash/timestamp prefix to distribute keys: abc123/2024/01/filename instead of logs/filename.
Want to master this scenario in a live sandbox? Stephane Maarek's AWS Certified DevOps Engineer Professional Masterclass on Udemy covers this exact problem with hands-on terminal drills.
🛠️ Production Runbook & Step-by-Step Resolution
Direct Answer: "AWS reliability requires differentiating between AWS control plane limits and host-level resource exhaustion. When addressing this question, I walk the interviewer through our production incident runbook: isolating the blast radius, checking diagnostic logs and metrics, and applying a safe fix."
Initial Diagnostics & Root Cause Analysis
- S3 automatically partitions by prefix — requests are distributed across S3 partitions. More unique prefixes = better parallelism. Add a hash/timestamp prefix to distribute keys:
abc123/2024/01/filenameinstead oflogs/filename. - Avoid sequential keys — old S3 had hot partition issues with sequential keys (dates). Modern S3 handles this better but prefixing is still good practice.
- S3 Select or Athena — for querying/filtering data, use S3 Select to retrieve only needed data instead of downloading entire files.
- Aggregate small files — if files are <1MB, aggregate into larger files. S3 works best with larger objects. Use multipart for >100MB.
Remediation & Permanent Safeguards
- Request parallelism — use multipart download with parallel part fetching for large files.
- Batch operations — for bulk operations on millions of objects, use S3 Batch Operations instead of single API calls.
- S3 automatically partitions by prefix — requests are distributed across S3 partitions. More uniqu...
- Avoid sequential keys — old S3 had hot partition issues with sequential keys (dates). Modern S3 h...
- S3 Select or Athena — for querying/filtering data, use S3 Select to retrieve only needed data ins...
Senior hiring managers and bar raisers evaluate how you prioritize during outages. Avoid these common anti-patterns:
- Relying on legacy bucket ACLs instead of modern S3 Block Public Access and IAM bucket policies when configuring application writes millions AWS recommends disabling ACLs (Bucket Owner Enforced) in favor of centralized declarative bucket policies and account-level Block Public Access controls.
-
Confusing SSE-S3 encryption with KMS customer-managed key (SSE-KMS) authorization policies when configuring application writes millions
Accessing SSE-KMS encrypted objects requires both
s3:GetObjectandkms:Decryptpermissions; neglecting KMS key policies causes silent 403 Forbidden errors. - Using sequential timestamps or predictable prefixes for high-throughput S3 read and write keys when configuring application writes millions Sequential key prefixes limit S3 horizontal partitioning; randomized hash prefixes distribute workloads across partitions to achieve 3,500 PUT and 5,500 GET requests/sec.
- Failing to audit S3 object delete markers when troubleshooting missing versioned objects when configuring application writes millions Issuing a DELETE on a versioned bucket only adds a delete marker; engineers must retrieve previous versions or purge markers to restore data.