{"slug": "aistore", "name": "AIStore", "category": "files", "category_name": "Files, sync and backup", "description": "", "recommenders": 3, "commenters": 5, "warners": 0, "score": 2.795, "rising": -0.98, "first": "2025-12-03", "latest": "2026-02-13", "why": "we used it to great success as a local on-prem s3 read/write cache with AWS as the backing S3 store", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 0, "compare": 2, "warn": 0}, "spark": [0, 2, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "we used it to great success as a local on-prem s3 read/write cache with AWS as the backing S3 store", "kind": "", "icon": false, "domain": null, "github_repo": "github.com/nvidia/aistore", "official_url": "https://github.com/NVIDIA/aistore", "pricing": "", "aliases": ["AIStore", "NVIDIA AIStore", "NVIDIA AIstore"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 752079073, "item_id": 47004942, "url": "https://news.ycombinator.com/item?id=47004942", "by": "gunapologist99", "date": "2026-02-13", "stance": "recommend", "quote": "NVIDIA has a phenomenal S3 compatible system", "excerpt": "Garage installation is easy. 1. Download or build the single binary into your system (install like `/usr/local/sbin/garage`) 2. Create a file `/etc/garage.toml`: metadata_dir = \"/data/garage/meta\" data_dir = \"/data/garage/data\" db_engine = \"sqlite\" replication_factor = 1 rpc_bind_addr = \"[::]:3901\" rpc_public_addr = \"127.0.0.1:3901\" rpc_secret = \"[your rpc secret]\" [s3_api] s3_region = \"garage\" api_bind_addr = \"[::]:", "context": "", "use_case": "", "switched_from": "", "story": "MinIO repository is no longer maintained"}, {"id": 738269475, "item_id": 46141842, "url": "https://news.ycombinator.com/item?id=46141842", "by": "56kbr", "date": "2025-12-03", "stance": "recommend", "quote": "There's a development/playground deployment for local K8s", "excerpt": "There's a development/playground deployment for local K8s (e.g. Minikube, KinD): https://github.com/NVIDIA/aistore/tree/main/deploy/dev/k8s. For production you'd need a proper cluster deployed via Helm, but for trying it out locally that setup is easy to get running.", "context": "", "use_case": "object storage", "switched_from": "", "story": "MinIO is now in maintenance-mode"}, {"id": 738186320, "item_id": 46136645, "url": "https://news.ycombinator.com/item?id=46136645", "by": "frellus", "date": "2025-12-03", "stance": "recommend", "quote": "we used it to great success as a local on-prem s3 read/write cache with AWS as the backing S3 store", "excerpt": "Check out from nvidia, aistore: https://github.com/NVIDIA/aistore It's not a fully featured s3 compatible service, like MinIO, but we used it to great success as a local on-prem s3 read/write cache with AWS as the backing S3 store. This avoided expensive network egress charges as we wanted to process data in both AWS as well as in a non-AWS GPU cluster (i.e. a neocloud)", "context": "", "use_case": "on-prem S3 read/write cache", "switched_from": "", "story": "MinIO is now in maintenance-mode"}]}], "caveats": [], "who": [], "tags": [], "use_cases": ["object storage", "on-prem s3 read/write cache"], "compared": [{"slug": "minio", "name": "MinIO", "co": 5, "switched_from": 0, "switched_to": 0}, {"slug": "garage", "name": "Garage", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [{"id": 740263824, "item_id": 46266489, "url": "https://news.ycombinator.com/item?id=46266489", "by": "alex-aizman", "date": "2025-12-14", "stance": "recommend", "quote": "AIStore is a reliable storage cluster that can natively operate on both in-cluster and remote data", "excerpt": "Appreciate the mention. Just to clarify: AIStore is a reliable storage cluster that can natively operate on both in-cluster and remote data, without treating either as a cache. Yes, it _can_ be configured to act as a cache in front of any of the four supported Clouds but (a) we've never done so at NVIDIA, and (b) its primary function always was - and remains - reliable, resilient storage for AI/ML workloads. For deta", "context": "", "use_case": "AI/ML object storage", "switched_from": "", "story": "We built another object storage"}], "threads": [{"url": "https://news.ycombinator.com/item?id=46136023", "title": "MinIO is now in maintenance-mode", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=46254087", "title": "We built another object storage", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=47000041", "title": "MinIO repository is no longer maintained", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {"recommend": 2, "compare": 2}, "2026-01": {}, "2026-02": {"recommend": 1}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {}, "2026-07": {}, "2026-08": {}, "2026-09": {}, "2026-10": {}}, "quotes": [{"id": 752079073, "item_id": 47004942, "url": "https://news.ycombinator.com/item?id=47004942", "by": "gunapologist99", "date": "2026-02-13", "stance": "recommend", "quote": "NVIDIA has a phenomenal S3 compatible system", "excerpt": "Garage installation is easy. 1. Download or build the single binary into your system (install like `/usr/local/sbin/garage`) 2. Create a file `/etc/garage.toml`: metadata_dir = \"/data/garage/meta\" data_dir = \"/data/garage/data\" db_engine = \"sqlite\" replication_factor = 1 rpc_bind_addr = \"[::]:3901\" rpc_public_addr = \"127.0.0.1:3901\" rpc_secret = \"[your rpc secret]\" [s3_api] s3_region = \"garage\" api_bind_addr = \"[::]:", "context": "", "use_case": "", "switched_from": "", "story": "MinIO repository is no longer maintained"}, {"id": 740079056, "item_id": 46254941, "url": "https://news.ycombinator.com/item?id=46254941", "by": "jamiesonbecker", "date": "2025-12-13", "stance": "compare", "quote": "how does this compare to NVIDIA's AIstore", "excerpt": "These questions are meant to be constructively critical, but not hyper-critical: I'm genuinely interested and a big fan of open-source projects in this space: * In terms of a high-performance AI-focused S3 competitor, how does this compare to NVIDIA's AIstore? https://aistore.nvidia.com/ * What's the clustering story? Is it complex like ceph, requires K8s like AIstore for full functionality, or is it more flexible li", "context": "", "use_case": "", "switched_from": "", "story": "We built another object storage"}, {"id": 738269475, "item_id": 46141842, "url": "https://news.ycombinator.com/item?id=46141842", "by": "56kbr", "date": "2025-12-03", "stance": "recommend", "quote": "There's a development/playground deployment for local K8s", "excerpt": "There's a development/playground deployment for local K8s (e.g. Minikube, KinD): https://github.com/NVIDIA/aistore/tree/main/deploy/dev/k8s. For production you'd need a proper cluster deployed via Helm, but for trying it out locally that setup is easy to get running.", "context": "", "use_case": "object storage", "switched_from": "", "story": "MinIO is now in maintenance-mode"}, {"id": 738202433, "item_id": 46137652, "url": "https://news.ycombinator.com/item?id=46137652", "by": "bityard", "date": "2025-12-03", "stance": "compare", "quote": "The two products are extremely similar.", "excerpt": "And the naming conflicts with NVidia's AIStore (https://github.com/NVIDIA/aistore). The two products are extremely similar. I don't know which came first, but Minio is going to want to do another pivot very soon if they want to survive. I doubt they have the resources to stand up to NVidia's army of extremely well-paid IP lawyers.", "context": "", "use_case": "", "switched_from": "", "story": "MinIO is now in maintenance-mode"}, {"id": 738186320, "item_id": 46136645, "url": "https://news.ycombinator.com/item?id=46136645", "by": "frellus", "date": "2025-12-03", "stance": "recommend", "quote": "we used it to great success as a local on-prem s3 read/write cache with AWS as the backing S3 store", "excerpt": "Check out from nvidia, aistore: https://github.com/NVIDIA/aistore It's not a fully featured s3 compatible service, like MinIO, but we used it to great success as a local on-prem s3 read/write cache with AWS as the backing S3 store. This avoided expensive network egress charges as we wanted to process data in both AWS as well as in a non-AWS GPU cluster (i.e. a neocloud)", "context": "", "use_case": "on-prem S3 read/write cache", "switched_from": "", "story": "MinIO is now in maintenance-mode"}]}