Sold by: abyo software
Deployed on AWS
Free Trial
Drop-in S3-compatible gateway that transparently compresses every object (CPU zstd or GPU nvCOMP), cutting S3 storage bytes 50-80 percent for compressible data with zero application changes. Includes pre-deployment savings estimation and measured-savings reporting.
Overview
S4 sits between your applications (boto3, aws-sdk, Spark, Trino, DuckDB, or anything that speaks S3) and your real S3 bucket, and transparently compresses each object with a codec chosen per payload: CPU zstd for text and logs, GPU (NVIDIA nvCOMP Bitcomp or zstd) for integer and columnar data, and passthrough for already-compressed inputs.
No application changes are required. S4 keeps the same S3 wire protocol, the same SigV4 authentication, and the same SDK calls. You only change the endpoint URL. Range GET requests stay fast through a sidecar frame index that is compatible with Parquet and ORC readers.
The on-backend format is open, so there is no lock-in. If you stop the gateway, the objects remain readable with the Apache-2.0 s4-codec command-line tool, the s4-codec Python package, or the s4fs fsspec adapter (pandas, pyarrow, and DuckDB can read the data without the gateway).
Operate S4 with built-in day-2 tooling. The s4 estimate command projects savings on your existing bucket before you deploy. The s4 migrate command retro-compresses objects already in the bucket. The s4 recompact command re-compresses cold data at higher levels. The s4 maintain command runs policy-driven maintenance. The savings ledger and the s4 savings command report the storage bytes and dollars actually saved in production. Prometheus metrics and a Grafana dashboard are included.
This listing is billed per pod per hour. The same software is also available free under the Apache-2.0 license as a public container image; the Marketplace build adds metered billing through your AWS bill and is the supported procurement path for enterprises.
Highlights
- 50-80 percent fewer S3 storage bytes for compressible data such as logs, JSON, and columnar files. The reduction is measured, with per-bucket savings reporting built in.
- Zero application changes. S4 exposes an S3 wire-compatible endpoint (SigV4, multipart, Range GET, server-side encryption). Point your SDK at S4 and keep working.
- Open format, no lock-in. Objects stay readable without the gateway through Apache-2.0 command-line, Python, and fsspec tooling.
Details
Sold by
Categories
Delivery method
Supported services
Delivery option
S4 gateway via Helm chart
Latest version
Operating system
Linux
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Free trial
Try this product free for 5 days according to the free trial terms set by the vendor. Usage-based pricing is in effect for usage beyond the free trial terms. Your free trial gets automatically converted to a paid subscription when the trial ends, but may be canceled any time before that.
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time. Alternatively, you can pay upfront for a contract, which typically covers your anticipated usage for the contract duration. Any usage beyond contract will incur additional usage-based costs.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/unit/hour |
|---|---|---|
Hours | Container Hours | $0.10 |
Dimensions summary
You pay a single usage-based fee measured in container hours. Billing accrues per hour the compression gateway runs, so your cost scales with runtime rather than storage volume or object count. This software fee is separate from the GPU EC2 instance you launch (g4dn, g5, g6, or g6e), which appears on your regular AWS invoice. There are no license keys; metering flows through AWS Marketplace. Because the charge is hourly, running the gateway longer increases the fee, while stopping it ends the charge.
Top-of-mind questions for buyers
One container hour is one hour that the compression gateway runs on your GPU instance. The gateway is packaged as an EC2 AMI running as a systemd service. The software fee accrues for each hour it operates, regardless of how many objects it compresses or how much data it stores.
The software fee meters running hours only. When you stop the gateway, the hourly charge ends. A powered-off instance stops accruing both the software fee and EC2 compute. Underlying AWS storage for the compressed objects and the AMI volume may still apply separately on your AWS invoice.
They bill separately. The container-hour software fee appears through AWS Marketplace metering. The GPU EC2 instance you launch — g4dn, g5, g6, or g6e — appears on your regular AWS invoice. Both charges accrue at the same time while the gateway runs. There are no license keys.
abyo.net
Helpful?
Vendor refund policy
S4 is billed per pod per hour on metered usage; stop the pods anytime to stop software charges, and a free trial is available. Usage already incurred is generally non-refundable. If you were billed in error or a product defect prevented normal use, email aws-support@abyo.net (abyo software LLC) within 30 days with your AWS account ID, product name (S4 - Squished S3), and billing period. Valid refunds are processed through AWS Marketplace.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
S4 gateway via Helm chart
Supported services: Learn more
- Amazon EKS
Helm chart
Helm charts are Kubernetes YAML manifests combined into a single package that can be installed on Kubernetes clusters. The containerized application is deployed on a cluster by running a single Helm install command to install the seller-provided Helm chart.
Version release notes
S4 v1.5.2 (cumulative over v1.5.1, same-day releases). Fix wave for three defects found in a live EKS end-to-end verification of v1.5.0, each reproduced on chart-default deployments: (1) Multipart / framed GETs and CompleteMultipartUpload no longer buffer whole objects in gateway memory - they stream frame-by-frame, so a single 2 GiB GET no longer OOM-kills a 2Gi-limit pod (previously reproduced 3/3); Range requests decode only the covering frames. The post-Complete index scan now uses O(parts) small ranged reads instead of a full-object transfer. (2) --read-timeout-seconds is now a progress-based idle guard instead of a whole-connection wall clock: healthy large transfers are no longer killed at 30 seconds (plain aws-cli 2 GiB uploads went from 2/6 to 6/6 in live verification); the TLS handshake is covered too; slowloris protection is preserved. New S4_READ_TIMEOUT_SECONDS env override and a helm extraArgs value for operators. (3) An interrupted CompleteMultipartUpload is now idempotently retryable: the gateway keeps a durable completion record, so a client retry that previously received NoSuchUpload (leaving an unreadable phantom object at the key, potentially replacing live data) now finishes the upload and returns success. s4 maintain (mpu-state-gc) detects and reports committed-but-unindexed objects left by crashed uploads that were never retried. (4) s4 savings now shows an always-accurate saved-bytes column (the exact metered quantity) and documents the 5 MiB multipart padding floor: with the aws-cli default 8 MiB part size, at-rest savings are limited until s4 recompact - raise multipart_chunksize to 64 MiB or more for multipart-heavy workloads. All metering behavior is unchanged and was verified byte-accurate against live MeterUsage records. Verified end-to-end on EKS with this image: 6/6 plain 2 GiB uploads, 2 GiB GET at 145 MiB gateway RSS with zero restarts, and interrupted-Complete retry recovery on a real phantom.
Additional details
Usage instructions
S4 deploys on Amazon EKS via the provided Helm chart.
-
Subscribe, then follow the AWS Marketplace launch steps for this product to install the Helm chart on your EKS cluster.
-
Metered billing is automatic. The chart carries the Marketplace product code and runs under the service account AWS Marketplace provisions with the aws-marketplace:MeterUsage permission. S4 confirms entitlement at startup (the container exits if the pod is not entitled) and then reports one metered pod-hour every hour to the AWS Marketplace Metering Service.
-
Set your S3 backend endpoint at install time. Override backend.endpointUrl, for example https://s3.us-east-1.amazonaws.com - use the regional S3 endpoint for the bucket you put behind S4.
-
Point your S3 clients (boto3, aws-sdk, Spark, Trino, DuckDB, or anything S3) at the S4 service endpoint. No application code changes are required - the gateway accepts any SigV4-signed client (it re-signs to the real backend with its own IRSA credentials); the ETag presented is MD5(original payload); ListObjects(V2), HEAD, and GET are all client-transparent by default. v1.5.2: healthy large transfers are never killed by the connection guard, multi-GiB GETs stream at constant gateway memory, and interrupted multipart Completes recover idempotently on retry.
-
Billing is per pod per hour through your AWS bill. EKS infrastructure costs are separate and billed by AWS as usual.
Documentation: https://github.com/abyo-software/s4#readme Operations runbook: https://github.com/abyo-software/s4/blob/main/docs/ops/runbook.md
Resources
Vendor resources
Support
Vendor support
Community support is available through GitHub issues at https://github.com/abyo-software/s4/issues . Marketplace subscribers can email aws-support@abyo.net for best-effort response within 2 business days. Documentation, an operations runbook, and a threat model are included in the repository.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
Self-contained EC2 AMI of the S4 transparent S3 compression gateway with CPU codecs (zstd / gzip) preinstalled. Launch on any general-purpose or compute-optimized instance (t3 / m6i / m7i / c6i / c7i), point your S3 clients at it, and cut S3 storage bytes 50-80 percent for compressible data with zero application changes.
Self-contained EC2 AMI of the S4 transparent S3 compression gateway with NVIDIA nvCOMP GPU codecs preinstalled. Launch on a GPU instance (g4dn / g5 / g6), point your S3 clients at it, and cut S3 storage bytes 50-80 percent for compressible data with zero application changes.
Drop-in S3-compatible gateway that transparently compresses every object, cutting S3 storage bytes 50-80 percent for compressible data with zero application changes. This edition bills by measured savings: you pay per GB of backend storage avoided, per hour, at roughly one third of the avoided storage cost.
Cut your CloudWatch custom-metric bill: govern metric cardinality at ingest, then auto-baseline and roll up savings across your whole AWS Organization.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.