AWS RDS Cost (2026): A Complete Breakdown of Every Charge
Verified August 2026
AWS RDS cost in 2026 ranges from roughly $15/month for a single-AZ db.t3.micro to $2,750+/month for production-grade Multi-AZ setups with read replicas, depending on instance type, storage, engine, and a half-dozen other choices most teams never think about until the bill arrives.
Pricing this accurately requires understanding that AWS RDS cost is calculated across a multi-meter billing model. Compute runs in per-second increments, storage in per-GB-months, IOPS in their own unit, and data transfer in per-GB chunks, each calculated on its own cycle and summed at the end of every billing period. Change the storage type, region, engine, or retention window and your monthly Amazon RDS pricing can shift 20-60% without any change to your application.
This guide breaks down every component that determines your AWS RDS cost in 2026: every instance family, every storage type, every hidden charge, with verified AWS pricing throughout. If you’ve been trying to understand why your AWS RDS bill grows so fast at scale, this is the breakdown you’ll bookmark.
All pricing reflects us-east-1 on-demand rates, verified August 2026. Verify against the AWS Pricing Calculator for your specific region and workload.
AWS RDS Pricing at a Glance: Every 2026 Rate on One Screen
AWS RDS bills across seven meters. A db.t4g.micro with 20 GB of storage runs $13.98/month, a production db.m5.large Multi-AZ with 100 GB runs about $291/month, and a 4 vCPU Multi-AZ setup with Provisioned IOPS and a read replica runs about $2,754/month. Every rate below produces those numbers. Verified August 2026, us-east-1.
| Charge | 2026 Rate (us-east-1) | Notes |
|---|---|---|
| db.t4g.micro (2 vCPU, 1 GiB) | $0.016/hr, about $11.68/month | Cheapest production-ready instance |
| db.t3.medium (2 vCPU, 4 GiB) | $0.072/hr PostgreSQL, $0.068/hr MySQL | Burstable, runs on a CPU credit model |
| db.m5.large (2 vCPU, 8 GiB) | $0.178/hr PostgreSQL, $0.171/hr MySQL | About $129.94/month, the production default |
| db.r5.large (2 vCPU, 16 GiB) | $0.25/hr PostgreSQL, $0.24/hr MySQL | Memory-optimized, double the RAM per vCPU |
| gp3 storage | $0.115/GB-month | 3,000 IOPS and 125 MB/s included |
| gp3 extra IOPS and throughput | $0.02/IOPS-month, $0.080/MB/s-month | Charged only above the included baseline |
| io1 and io2 storage | $0.125/GB-month plus $0.10/IOPS-month | IOPS rate tiers down above 32,000 IOPS |
| Magnetic storage | Retired April 2026 | Force-migrated to gp3 |
| Multi-AZ | 2x compute and 2x storage | Standby cannot serve read traffic |
| Read replica | 1x the source instance rate, each | Serves reads, but gives you no failover |
| Automated backups | Free to 100% of provisioned size, then $0.095/GiB-month | Tied to a 1 to 35 day retention window |
| Manual snapshots | $0.023/GB-month | Never expire, and survive instance deletion |
| Snapshot export to S3 | $0.013/GB of snapshot size | One-time charge on the source snapshot |
| Data transfer in, and same-AZ | $0.00 | Free regardless of volume |
| Cross-AZ and cross-region transfer | $0.01/GB and $0.02/GB | Cross-AZ is charged in both directions |
| Internet egress | First 100 GB/month free, then $0.09/GB | Tiers down above 10 TB/month |
| RDS Proxy | $0.015/vCPU-hour, 2 vCPU minimum | About $22/month floor per proxy |
| Extended Support | $0.100/vCPU-hour years 1-2, $0.200/vCPU-hour year 3 onward | Not discounted by Reserved Instances |
| Enhanced Monitoring | $0.14 to $8.04 per instance-month | CloudWatch Logs rates, set by granularity |
| CloudWatch Database Insights | Standard $0, Advanced $0.0125/vCPU-hour | Replaced Performance Insights in 2026 |
| Reserved Instances | 29-34% off on 1 year, 52-72% off on 3 years | Discounts compute only, nothing else |
Four of those lines changed during 2026 and most published RDS cost guides still carry the old numbers. The next section covers exactly what moved.
Table of Contents
How AWS RDS Charges You: Pricing Units and Billing Cycles Explained
Amazon RDS bills your AWS account in real time across seven independent meters: compute per second with a 10-minute minimum, storage per GB-month, Provisioned IOPS per IOPS-month, data transfer per GB, and add-on services either per vCPU-hour or, in Enhanced Monitoring’s case, per GB streamed into CloudWatch Logs. Each accrues continuously through the month and totals on your AWS invoice.
AWS RDS Billing Units and Cycles (2026)
Each of these meters accrues on its own cycle and none of them substitutes for another, which is why the total is always more than the instance rate.
| Meter | Pricing Unit | Billing Cycle | Minimum Charge |
|---|---|---|---|
| Compute (instance hours) | Per second | Continuous | 10-minute floor on status changes |
| Storage | Per GB-month | Pro-rated daily | None |
| Provisioned IOPS | Per IOPS-month | Pro-rated daily | None |
| gp3 throughput (beyond baseline) | Per MB/s-month | Pro-rated daily | None |
| Backup (beyond free tier) | Per GB-month | Pro-rated daily | None |
| Manual snapshots | Per GB-month | Pro-rated daily | None |
| Data transfer | Per GB | Per usage | None |
| RDS Proxy | Per vCPU-hour | Continuous | 2 vCPU per proxy minimum |
| Extended Support | Per vCPU-hour | Continuous | None |
| Enhanced Monitoring | Per GB ingested into CloudWatch Logs | Continuous | CloudWatch Logs free tier |
| CloudWatch Database Insights | Per vCPU-hour (Advanced mode only) | Continuous | Standard mode is free |
Cost Explorer breaks these out by usage type, RunInstance, Storage, IOPS, Backup Usage, Data Transfer, Enhanced Monitoring and Extended Support, so if a bill surprises you, that is where the meter is named.
For teams comparing this multi-meter structure against alternatives, the Complete Cost Comparison (2026) shows how these same meters behave on raw EC2 or Hetzner Cloud.
AWS RDS Compute Pricing: Instance Costs by Family and Engine (2026)
AWS RDS compute pricing varies sharply by instance family (t-class, m-class, r-class), database engine, and licensing model, with verified 2026 hourly rates ranging from $0.016 for a db.t4g.micro on MySQL or PostgreSQL to $0.750/hr for SQL Server Standard Edition on equivalent 2-vCPU instances in us-east-1.
Compute is typically 35-50% of the monthly bill before Multi-AZ, replicas, or storage IOPS are factored in.
AWS RDS PostgreSQL Pricing 2026: Official Per-Hour Rates by vCPU and RAM (us-east-1)
These are AWS’s official on-demand PostgreSQL rates for us-east-1, per hour and per month at 730 hours, Single-AZ. Multi-AZ doubles every figure in the table. Storage is billed separately at $0.115 per GB-month for gp3. Verified August 2026.
| Instance | vCPU | RAM | Price Per Hour | Per Month, Single-AZ | Per Month, Multi-AZ |
|---|---|---|---|---|---|
| db.t4g.micro | 2 vCPU | 1 GiB | $0.016/hr | $11.68/mo | $23.36/mo |
| db.t4g.small | 2 vCPU | 2 GiB | $0.032/hr | $23.36/mo | $46.72/mo |
| db.t4g.medium | 2 vCPU | 4 GiB | $0.065/hr | $47.45/mo | $94.90/mo |
| db.t3.medium | 2 vCPU | 4 GiB | $0.072/hr | $52.56/mo | $105.12/mo |
| db.m5.large | 2 vCPU | 8 GiB | $0.178/hr | $129.94/mo | $259.88/mo |
| db.m5.xlarge | 4 vCPU | 16 GiB | $0.356/hr | $259.88/mo | $519.76/mo |
| db.r5.large | 2 vCPU | 16 GiB | $0.25/hr | $182.50/mo | $365.00/mo |
| db.r5.xlarge | 4 vCPU | 32 GiB | $0.50/hr | $365.00/mo | $730.00/mo |
| db.r5.4xlarge | 16 vCPU | 128 GiB | $2.00/hr | $1,460.00/mo | $2,920.00/mo |
| db.r6in.xlarge | 4 vCPU | 32 GiB | $0.692/hr | $505.16/mo | $1,010.32/mo |
Rates scale linearly inside a family, so a 4xlarge is exactly 8x the large. MySQL and MariaDB run 4-6% below PostgreSQL on identical hardware, and a 100 GB gp3 volume adds $11.50/month to every row above.
How database engine choice affects AWS RDS compute pricing
On identical hardware, engine choice changes your AWS RDS cost by more than 4x. Open source is cheapest, proprietary engines carry licensing premiums, and if you already hold Oracle enterprise licences, BYOL brings Oracle close to open-source rates.
AWS RDS Engine Cost Comparison on Equivalent 2 vCPU, 8 GiB Hardware (2026)
On identical hardware, engine licensing alone swings the hourly rate by more than 4x.
| Engine | Hourly Rate | Monthly Cost (730 hrs) | Premium vs MySQL |
|---|---|---|---|
| MySQL (db.m5.large) | $0.171/hr | $124.83/mo | baseline |
| MariaDB (db.m5.large) | $0.171/hr | $124.83/mo | 0% above MySQL |
| PostgreSQL (db.m5.large) | $0.178/hr | $129.94/mo | 4.1% above MySQL |
| Aurora MySQL Standard¹ | $0.29/hr | $211.70/mo | 69.6% above MySQL |
| Aurora PostgreSQL Standard¹ | $0.29/hr | $211.70/mo | 69.6% above MySQL |
| Oracle LI SE2 (db.m5.large) | $0.438/hr | $319.74/mo | 156.1% above MySQL |
| SQL Server LI Standard² | $0.750/hr | $547.50/mo | 338.6% above MySQL |
¹ Aurora rate shown for db.r5.large (Aurora doesn’t list db.m5.large in current pricing).
² Rate shown for db.m8i.large (identical specs to db.m5.large). Includes instance, Windows OS license, and SQL Server license.
DynamoDB vs RDS Pricing: Provisioned Instances Against Per-Request Billing
RDS and DynamoDB are priced on incompatible models, which is why a like-for-like comparison usually misleads. RDS bills an instance by the hour whether or not a query arrives, so a db.m5.large costs $129.94/month at zero traffic and the same at full load. DynamoDB On-Demand bills per request with no instance at all, so it costs nothing idle and scales linearly with traffic.
The crossover is workload shape, not size. A steady application with predictable load is almost always cheaper on RDS, because you are buying compute wholesale. A spiky or intermittent workload is usually cheaper on DynamoDB, because you stop paying between spikes. Where RDS gets expensive is the middle case: an always-on database that is mostly idle, paying full instance rate for utilisation it never reaches.
For teams comparing AWS RDS cost against managed PostgreSQL alternatives, the Managed PostgreSQL Comparison (2026) lays out side-by-side pricing.
AWS RDS Free Tier Limits in 2026: What You Actually Get
What the RDS free tier gives you now depends entirely on when your AWS account was created. There are two different offers running side by side, and most guides describe only the retired one.
Accounts created before 15 July 2025 keep the legacy offer: 750 hours a month of a db.t2.micro, db.t3.micro or db.t4g.micro Single-AZ instance on MySQL, PostgreSQL or MariaDB, plus 20 GB of General Purpose SSD and 20 GB of backup storage, for 12 months from account creation.
Accounts created on or after 15 July 2025 get Free Tier credits instead. You pick a Free plan or a Paid plan, and the Free plan runs up to 6 months or until the credits are spent. RDS hours draw down that shared balance rather than a separate RDS allowance, so a single always-on db.m5.large can consume it in under three weeks.
Either way, the free tier covers a dev database and nothing more. The moment you need Multi-AZ, a replica, or more than 20 GB, you are on standard rates. If a permanently free tier is what you are after, managed Postgres providers outside AWS still offer one, and the Managed PostgreSQL Comparison (2026) prices them side by side.
AWS RDS Storage Pricing in 2026: gp3, io1, io2, and Magnetic Storage Costs
AWS RDS storage pricing in 2026 has two live options: General Purpose SSD (gp2 and gp3) at $0.115 per GB-month, and Provisioned IOPS SSD (io1 and io2) at $0.125 per GB-month plus $0.10 per provisioned IOPS-month. Magnetic storage was retired in April 2026. On the same 100 GiB volume, io2 provisioned to 5,000 IOPS costs 7.8x what gp3 costs, which makes storage type the cheapest decision on this page to get wrong.
gp3 lets you scale capacity, IOPS and throughput independently, which is why it is the default for almost everything. gp2 costs the same per GB with tighter performance, and AWS migrates gp2 to gp3 in place with no downtime.

AWS RDS Provisioned IOPS pricing (io1 and io2): for I/O-heavy workloads
| Type | Storage | IOPS | Notes |
|---|---|---|---|
| io2 | $0.125/GB-month | $0.10/IOPS-month to 32,000 IOPS, tiering down above | Current generation (Block Express) |
| io1 | $0.125/GB-month | $0.10/IOPS-month to 32,000 IOPS, tiering down above | Previous generation |
io2 caveats, both of which distort quotes: provisioned IOPS pricing is tiered, so the flat $0.10 per IOPS-month holds only to 32,000 IOPS and drops above it. And Block Express delivers full performance only on Nitro-based instances (m5, r5, t3 and later), running degraded on older hardware. A quote built from a flat rate overstates cost at high IOPS while overstating performance on old instances.
AWS RDS Magnetic Storage Pricing: Retired in 2026, Migrate to gp3
Magnetic storage is no longer a pricing option. AWS began force-migrating RDS magnetic volumes to gp3 on 30 April 2026, changed the snapshot-restore default to gp3 on 1 June 2026, and since 1 July 2026 you cannot restore a snapshot onto magnetic at all. Any cost model still listing magnetic at $0.10/GB-month predates the deadline.
If an old volume is still on your bill, it has already been converted or is queued for conversion, and it now prices at gp3’s $0.115 per GB-month with a 3,000 IOPS baseline included.
AWS RDS Storage Type Cost Comparison and Use Cases (2026)
| Storage Type | Cost per GB-month | IOPS Pricing | Baseline IOPS | Best For |
|---|---|---|---|---|
| gp3 (recommended) | $0.115/GB-month | $0.02/IOPS-month above 3,000 IOPS | 3,000 IOPS + 125 MB/s included | Most production workloads |
| gp2 (legacy SSD) | $0.115/GB-month | Bundled, tied to volume size | 3 IOPS/GiB, 100 IOPS minimum | Existing volumes, migrate to gp3 |
| io2 (current IOPS) | $0.125/GB-month | $0.10/IOPS-month to 32,000 IOPS, tiering down above | Provisioned, 1,000 to 256,000 IOPS | New high-IOPS production workloads |
| io1 (legacy IOPS) | $0.125/GB-month | $0.10/IOPS-month to 32,000 IOPS, tiering down above | Provisioned, 1,000 to 256,000 IOPS | Existing high-IOPS workloads |
| Magnetic (retired 2026) | No longer offered | n/a | n/a | Force-migrated to gp3 from 30 April 2026 |
Verify rates on the official AWS RDS pricing page.
AWS RDS storage cost example: gp3 vs io2 for the same 100 GiB volume
A 100 GiB gp3 volume provisioned to 5,000 IOPS and 300 MB/s throughput:
- 100 × $0.115 = $11.50 storage
- (5,000 − 3,000) × $0.02 = +$40 IOPS
- (300 − 125) × $0.080 = +$14 throughput
- Total: $65.50/month
The same data on 100 GiB io2 with 5,000 provisioned IOPS:
100 × $0.125 + 5,000 × $0.10 = $512.50/month
7.8x more on io2 than gp3.
What Is a Fair Price Per GB for a Production Managed Database?
On AWS in 2026, $0.10 to $0.13 per GB-month is the fair band for production-grade SSD storage with a real IOPS baseline. RDS gp3 at $0.115 per GB-month sits squarely inside it. Below roughly $0.09 per GB-month you are usually looking at network storage with no IOPS guarantee, and above $0.25 per GB-month you are paying a serverless premium for storage that separates compute from data. Provisioned IOPS changes the calculation entirely: io2 at $0.125 per GB-month plus $0.10 per IOPS-month reaches $1.03 per effective GB-month on a 500 GB volume at 5,000 IOPS.
For the operational side of how storage scaling plays out at production scale, Database Scaling: Managed vs Self-Hosted, What Breaks First, covers what changes architecturally as your database grows.
AWS RDS Multi-AZ Cost: How High Availability Doubles Your Bill (2026)
AWS RDS Multi-AZ deployment doubles your compute and storage costs because AWS provisions a full synchronous standby instance in a separate Availability Zone. A $129.94/month Single-AZ db.m5.large PostgreSQL deployment becomes $259.88/month with Multi-AZ enabled. This is the single largest one-decision cost multiplier in AWS RDS.

AWS RDS Multi-AZ DB Cluster pricing: the 3-instance configuration for HA plus read scaling
Multi-AZ DB Cluster is a different product from classic Multi-AZ: three instances across three AZs, one primary and two standbys, and both standbys serve read traffic. Compute runs at roughly 3x the base instance rate, about $389.82/month on db.m5.large, storage moves to a cluster-shared pricing tier, and replication across the three AZs is included. It is the right shape when you need HA and read scaling together rather than provisioning a separate replica.
Verify storage/IOPS specifics on official AWS RDS Multi-AZ documentation.
AWS RDS Multi-AZ vs Single-AZ vs Read Replica Cost Multiplier (db.m5.large PostgreSQL, 2026)
Multi-AZ and read replicas cost the same in compute and differ entirely in what you get, so the choice is about failover versus read capacity, not price.
| Deployment | Compute Multiplier | Storage Multiplier | Best Use Case | Monthly Compute |
|---|---|---|---|---|
| Single-AZ | 1.0x compute | 1.0x storage | Dev, staging, low-risk production | $129.94/mo |
| Multi-AZ (classic, 1 standby) | 2.0x compute | 2.0x storage | HA-required production | $259.88/mo |
| Multi-AZ DB Cluster (2 readable standbys) | ~3.0x compute | Cluster-specific | HA plus read scaling combined | ~$389.82/mo |
| Single-AZ plus 1 read replica | 2.0x compute | 1.0x storage | Read scaling without HA | $259.88/mo |
| Multi-AZ plus 1 read replica | 3.0x compute | 2.0x storage | HA plus read scaling | $389.82/mo |
Multi-AZ is where an RDS bill stops being a rounding error, because it is the one toggle that doubles two meters at once. If you want the same 2 vCPU / 8 GiB Postgres on the same AWS hardware without the doubling, price it on Selfhost.dev’s calculator and compare it against the number above. No signup, and it takes about 30 seconds.
Multi-AZ is often the first big scaling decision growing teams face. See Database Scaling: Managed vs Self-Hosted, What Breaks First for what else changes architecturally.
AWS RDS Backup and Snapshot Pricing in 2026: What’s Free and What Costs Extra
AWS RDS automated backup storage is free up to 100% of your provisioned database size. Beyond that, backup storage costs $0.095 per GiB-month and manual snapshots are billed at standard Amazon S3 rates of $0.023 per GB-month. The structural difference between automated backups (auto-expire) and manual snapshots (persist forever) is one of the most consistent hidden cost drivers in mature AWS RDS deployments.
AWS RDS snapshot export to S3: archival pricing for long-term retention
Verified 2026 rate: $0.013 per GB of snapshot size, per AWS’s own RDS pricing page. Charged per GB of source snapshot (not exported data). Uses Parquet format (2x faster unload, 6x less storage than raw text). Once exported, billed at standard S3 rates. Additional charges apply for KMS encryption and S3 PUT requests.
AWS RDS Backup, Snapshot, and Export Pricing Summary (2026)
| Backup Type | Free Tier | Rate Beyond Free | Key Notes |
|---|---|---|---|
| Automated backups | 100% of provisioned database size | $0.095/GiB-month | Tied to your 1 to 35 day retention window, auto-deleted when the instance is terminated |
| Manual snapshots | None | $0.023/GB-month at standard S3 rates | Persist indefinitely and survive instance termination, so they accumulate silently |
| Snapshot export to S3 | None | $0.013/GB of snapshot size | One-time export charge, billed on the source snapshot size rather than the exported data |
| Terminated instance backups | None | $0.095/GiB-month | Retained backups keep billing after the instance is gone, until you delete them |
Verify rates on the official AWS RDS pricing page and Amazon S3 pricing page.
AWS RDS Data Transfer Cost in 2026: When Network Traffic Becomes a Hidden Bill
AWS RDS data transfer is free for inbound traffic and same-AZ traffic, but costs $0.01/GB for cross-AZ, $0.02/GB for inter-region, and $0.09/GB (tiered) for internet egress. Data transfer typically adds 5-15% to the monthly AWS RDS bill, and 30%+ for replicated or cross-region architectures.
AWS RDS Data Transfer Pricing (Internet Egress) Explained
| Monthly Volume | Rate per GB |
|---|---|
| First 100 GB | Free across all AWS regions combined |
| Next 10 TB | $0.09/GB |
| Next 40 TB | $0.085/GB |
| Next 100 TB | $0.07/GB |
| Above 150 TB | $0.05/GB |
The 100 GB free tier resets monthly and is shared across all AWS services, not just RDS.
AWS RDS Data Transfer Pricing by Source and Destination (2026)
Every free data transfer path on RDS is either inbound or same-AZ, so the cost of network traffic is really the cost of crossing a boundary.
| From | To | Cost per GB | Notes |
|---|---|---|---|
| Internet or external | RDS (inbound) | $0.00, free | All inbound traffic is free |
| EC2 (same AZ) | RDS (same AZ) | $0.00, free | Co-located workloads pay nothing |
| RDS primary | RDS standby (Multi-AZ) | $0.00, free | Replication is included |
| EC2 (different AZ) | RDS (cross-AZ) | $0.01/GB | Charged in both directions |
| RDS (region A) | RDS (region B) | $0.02/GB | Cross-region read replicas |
| RDS | Internet (egress) | From $0.09/GB after the first 100 GB/month | See the tiered egress table above for volumes past 10 TB. |
For teams running production workloads where data transfer is meaningful, AWS RDS vs Hetzner Cloud Cost: $3,150 vs $835 (2026) shows how Hetzner’s included monthly traffic (typically 20 TB) compares to AWS’s metered model.
AWS RDS Reserved Instance Pricing in 2026: When 1-Year and 3-Year Commitments Pay Off
AWS RDS Reserved Instances reduce compute costs by 29-34% for 1-year terms and up to 52-72% for 3-year terms, but the discount applies only to instance hours. Storage, IOPS, backups, data transfer, and add-on services continue at full on-demand rates. Even with a 3-year all-upfront RI, your total AWS RDS cost can still grow if those uncovered line items scale.
AWS RDS Reserved Instance Savings vs On-Demand (2026)
| Commitment Term | Payment Option | Compute Savings | Best For |
|---|---|---|---|
| 1 year | No Upfront | ~29% | Limited cash flow, predictable workload |
| 1 year | Partial Upfront | ~33% | Balance between upfront and savings |
| 1 year | All Upfront | ~34% | Stable workloads, 1-year horizon |
| 3 years | No Upfront | ~52% | Stable workloads, no cash for prepay |
| 3 years | Partial Upfront | ~60% | Long-term, moderate cash flow |
| 3 years | All Upfront | ~63-72% | Mission-critical, multi-year-stable |
Calculate exact savings on the official AWS Reserved Instance pricing page.
Even a 72%-discounted RI still pays AWS’s managed-service premium on storage, IOPS, and data transfer. For teams considering RI vs migrating to self-hosted, AWS RDS vs Self-Hosted PostgreSQL: Complete Cost Comparison shows the full math.
Hidden AWS RDS Costs in 2026: RDS Proxy, Extended Support, Performance Insights, and Enhanced Monitoring
Beyond compute and storage, AWS RDS adds optional charges that never appear in instance pricing: RDS Proxy at $0.015 per vCPU-hour with a 2-vCPU minimum, Extended Support at $0.100 to $0.200 per vCPU-hour for engine versions past EOL, Enhanced Monitoring billed through CloudWatch Logs at $0.14 to $8.04 per instance-month, and CloudWatch Database Insights Advanced at $0.0125 per vCPU-hour if you need more than 7 days of query history. Together they add $25 to $700 per instance per month, and Extended Support accounts for most of the top of that range.
AWS RDS Proxy pricing: per-vCPU-hour billing for connection pooling
- Rate: $0.015 per vCPU-hour
- Minimum: 2 vCPUs per proxy (~$22/month minimum)
- Included: default proxy endpoint
- NOT included: custom endpoints trigger AWS PrivateLink fees (~$0.01/hr + $0.01/GB)
For connection-heavy PostgreSQL workloads, the PostgreSQL MCP Server: A Complete Guide to AI-Driven Database Management covers how AI agents connect directly to managed Postgres without traditional pooling middleware.
AWS RDS Extended Support pricing: $0.10-$0.20 per vCPU-hour for retired engine versions
The rate doubles in year three, and Reserved Instances do not discount either tier.
| Time After Engine EOL | Rate per vCPU-Hour | Monthly Cost, 4-vCPU Instance |
|---|---|---|
| Years 1-2 | $0.100/vCPU-hour | $292/month |
| Year 3 onward | $0.200/vCPU-hour | $584/month |
Worked example
A db.m5.xlarge (4 vCPUs) running PostgreSQL 12 past EOL:
- Years 1-2: 4 × $0.100 × 730 = $292/month
- Year 3+: 4 × $0.200 × 730 = $584/month
For context, the underlying db.m5.xlarge PostgreSQL compute is ~$260/month. Extended Support in Year 3+ more than doubles your bill for the same database, just for continued security patches.
Avoid Extended Support charges: upgrade to a supported major version before EOL.
Reserved Instance discounts do not apply to Extended Support.
AWS RDS PostgreSQL End of Standard Support Dates: When Extended Support Starts Billing
Extended Support is not opt-in. On the end-of-standard-support date, RDS automatically enrols your instance and starts charging $0.100 per vCPU-hour, rising to $0.200 per vCPU-hour from year three. Here is when each major version crosses that line.
| PostgreSQL Major Version | Community End of Life | RDS End of Standard Support | Year 1 Pricing Starts | Year 3 Pricing Starts | Extended Support Ends |
|---|---|---|---|---|---|
| PostgreSQL 11 | 9 Nov 2023 | 29 Feb 2024 | 1 Apr 2024 | 1 Apr 2026 | 31 Mar 2027 |
| PostgreSQL 12 | 14 Nov 2024 | 28 Feb 2025 | 1 Mar 2025 | 1 Mar 2027 | 29 Feb 2028 |
| PostgreSQL 13 | Nov 2025 | 28 Feb 2026 | 1 Mar 2026 | 1 Mar 2028 | 28 Feb 2029 |
| PostgreSQL 14 | 12 Nov 2026 | 28 Feb 2027 | 1 Mar 2027 | 1 Mar 2029 | 28 Feb 2030 |
| PostgreSQL 15 | Nov 2027 | 29 Feb 2028 | 1 Mar 2028 | 1 Mar 2030 | 28 Feb 2031 |
| PostgreSQL 16 | Nov 2028 | 28 Feb 2029 | 1 Mar 2029 | 1 Mar 2031 | 29 Feb 2032 |
| PostgreSQL 17 | Nov 2029 | 28 Feb 2030 | 1 Mar 2030 | 1 Mar 2032 | 28 Feb 2033 |
| PostgreSQL 18 | Nov 2030 | 28 Feb 2031 | 1 Mar 2031 | 1 Mar 2033 | 28 Feb 2034 |
Two of these are already billing. Anyone still on PostgreSQL 13 has been paying Extended Support since 1 March 2026, and PostgreSQL 11 moved to the doubled year-three rate on 1 April 2026. On a 4-vCPU instance that is $292/month at the year-one rate and $584/month at the year-three rate, on top of normal compute, for security patches alone.
PostgreSQL 14 is the next deadline that matters. Community end of life is 12 November 2026 and RDS standard support ends 28 February 2027, so a 14.x fleet has roughly six months before the meter starts.
AWS RDS Enhanced Monitoring Pricing: Billed Through CloudWatch Logs, Not Per Instance-Hour
Enhanced Monitoring has no hourly rate of its own. It streams OS metrics into CloudWatch Logs and you pay CloudWatch’s rates, $0.50 per GB ingested plus $0.03 per GB-month stored. Your bill is set entirely by the granularity you choose, because the data volume at each setting is fixed:
- 1-second: 16.07 GB/month, about $8.04 per instance per month
- 5-second: 3.21 GB/month, about $1.61 per instance per month
- 10-second: 1.61 GB/month, about $0.81 per instance per month
- 60-second: 0.27 GB/month, about $0.14 per instance per month
Dropping from 1-second to 60-second cuts this line by 98% and still catches noisy-neighbour effects on burstable instances. Most teams enable 1-second by reflex and pay 57 times more than they need to.
AWS RDS Performance Insights Pricing in 2026: The Paid Tier Is Gone
Performance Insights no longer has a paid tier. AWS retired flexible retention on 30 June 2026 and the dashboard on 31 July 2026, and every RDS instance now defaults to CloudWatch Database Insights in Standard mode with 7 days of history at no charge. Budget zero for this line.
If you need history beyond 7 days, the replacement is CloudWatch Database Insights Advanced mode at $0.0125 per vCPU-hour, roughly $9.13 per vCPU per month, with 15-month retention. On a 4-vCPU instance that is about $36.50/month.
Any RDS cost model written before mid-2026 still bills Performance Insights at $0.02 per vCPU-hour. Delete that line from your estimate.
AWS RDS Add-On Service Pricing: RDS Proxy, Extended Support, Monitoring (2026)
None of these appear in instance pricing, and all of them bill continuously once enabled.
| Service | Pricing | Free Tier | When You Need It |
|---|---|---|---|
| RDS Proxy | $0.015/vCPU-hour (2 vCPU minimum) | None | Serverless/Lambda, high connection churn |
| Extended Support (Yr 1-2) | $0.100/vCPU-hour | None | Engine versions past end of standard support |
| Extended Support (Yr 3+) | $0.200/vCPU-hour | None | Year 3 onward of post-EOL, at 2x the Year 1-2 rate |
| Enhanced Monitoring | $0.14 to $8.04/instance-month, at CloudWatch Logs rates | CloudWatch Logs free tier | Performance troubleshooting, noisy-neighbour diagnosis on burstable instances |
| CloudWatch Database Insights | Standard $0. Advanced $0.0125/vCPU-hour | Standard mode, 7 days | Query analysis and historical baselining. Replaced Performance Insights in 2026 |
See official AWS RDS Extended Support pricing page for engine-version-specific rates.
A typical production setup with RDS Proxy, 60-second Enhanced Monitoring, and Database Insights Standard adds about $22 per instance per month, down from the $50 to $80 the same stack cost before Performance Insights was retired. If you are in Extended Support territory, add $200 to $700 per instance per month on top.
AWS RDS Cost by Region in 2026: How Pricing Varies Across us-east-1, eu-west-1, and ap-south-1
AWS RDS cost varies up to 42% by region: us-east-1 (N. Virginia) and us-west-2 (Oregon) are tied as the cheapest at $0.178/hr for db.m5.large PostgreSQL, while ap-south-1 (Mumbai) charges $0.253/hr, a 42% premium for the identical instance. Pricing differences reflect local infrastructure costs, energy pricing, regulatory overhead, and AWS’s competitive positioning per market.
Regional premiums track local power, real estate and utilisation, and they are worth paying only for latency to your users or for a data-residency rule you cannot break. A 40% premium on the database costs less than a 150ms round trip costs your conversion rate.
AWS RDS Compute Cost by Region: Same Instance, Different Locations (db.m5.large PostgreSQL, 2026)
The same instance costs 42% more in Mumbai than in Virginia for identical specs.
| Region | Hourly Rate | Monthly Cost (730 hrs) | Premium vs us-east-1 |
|---|---|---|---|
| us-east-1 (N. Virginia) | $0.178/hr | $129.94/mo | baseline |
| us-west-2 (Oregon) | $0.178/hr | $129.94/mo | 0%, tied with us-east-1 |
| eu-west-1 (Ireland) | $0.197/hr | $143.81/mo | 10.7% above us-east-1 |
| eu-west-2 (London) | $0.206/hr | $150.38/mo | 15.7% above us-east-1 |
| sa-east-1 (São Paulo) | $0.233/hr | $170.09/mo | 30.9% above us-east-1 |
| ap-southeast-1 (Singapore) | $0.247/hr | $180.31/mo | 38.8% above us-east-1 |
| ap-south-1 (Mumbai) | $0.253/hr | $184.69/mo | 42.1% above us-east-1 |
Verify rates on the official AWS RDS pricing page by selecting each region.
For teams comparing AWS regional pricing against alternatives like Hetzner Cloud (consistent global pricing, no regional premiums), AWS RDS vs Hetzner Cloud Cost: $3,150 vs $835 (2026) shows the structural difference.
Real AWS RDS Cost Examples for 2026: 3 Production Configurations Priced Line by Line
A realistic monthly AWS RDS cost ranges from approximately $14/month for a development setup to $2,754/month for an enterprise-grade. Multi-AZ Provisioned IOPS deployment with a read replica. Here are three real-world configurations priced line by line using verified 2026 us-east-1 rates.
AWS RDS development environment cost example: db.t4g.micro at ~$14/month
A dev AWS RDS setup on db.t4g.micro Single-AZ (PostgreSQL, 20 GiB gp3, 7-day backup) costs approximately $14/month, the smallest production-ready RDS configuration available.
AWS RDS Development Environment Cost Breakdown (db.t4g.micro Single-AZ, 2026)
At the smallest viable size, compute is 84% of the bill and every other meter rounds to nothing.
| Component | Calculation | Monthly Cost |
|---|---|---|
| Instance compute | $0.016/hr × 730 hrs | $11.68/mo |
| Storage (gp3) | 20 GiB × $0.115/GB-mo | $2.30/mo |
| Backup storage | Within the 100% free allowance | $0.00/mo |
| Data transfer | Free, inbound and same-AZ | $0.00/mo |
| Total monthly AWS RDS cost | $13.98/mo |
Best for side projects, internal tools, dev/staging, and small admin dashboards.
Annual cost: ~$168.
AWS RDS production SaaS cost example: db.m5.large Multi-AZ at ~$294/month
A production SaaS AWS RDS setup on db.m5.large Multi-AZ (PostgreSQL, 100 GiB gp3, 14-day backup, Enhanced Monitoring) costs approximately $291/month.
AWS RDS Production SaaS Cost Breakdown (db.m5.large Multi-AZ, 2026)
Multi-AZ doubles both compute and storage, taking this configuration to just over twice the Single-AZ equivalent of $141.44/month.
| Component | Calculation | Monthly Cost |
|---|---|---|
| Instance compute (Multi-AZ) | $0.178/hr × 2 × 730 hrs | $259.88/mo |
| Storage (gp3, replicated) | 100 GiB × $0.115/GB-mo × 2 | $23.00/mo |
| Backup storage | Within the 100% free allowance | $0.00/mo |
| Data transfer (moderate) | ~50 GB cross-AZ at $0.01/GB | ~$0.50/mo |
| Enhanced Monitoring (1-second, via CloudWatch Logs) | 16.07 GB × $0.50/GB | $8.04/mo |
| Total monthly AWS RDS cost | ~$291.42/mo |
Best for early-to-mid stage production SaaS with HA requirements.
Annual cost: ~$3,497.
AWS RDS enterprise cost example: db.r6in.xlarge with Provisioned IOPS at ~$2,760/month
An enterprise AWS RDS setup on db.r6in.xlarge Multi-AZ (PostgreSQL, 500 GiB io2 with 5,000 IOPS, 30-day backup, Performance Insights long retention, 1 read replica, RDS Proxy) costs approximately $2,754/month.
AWS RDS Enterprise Cost Breakdown (db.r6in.xlarge Multi-AZ, 2026)
Provisioned IOPS alone is $1,000 of this bill, more than the storage it sits on and twice what the read replica costs.
| Component | Calculation | Monthly Cost |
|---|---|---|
| Instance compute (Multi-AZ) | $0.692/hr × 2 × 730 hrs | $1,010.32/mo |
| Read replica (Single-AZ) | $0.692/hr × 730 hrs | $505.16/mo |
| Storage (io2, replicated) | 500 GiB × $0.125/GB-mo × 2 | $125.00/mo |
| Provisioned IOPS (replicated) | 5,000 IOPS × $0.10/IOPS-mo × 2 | $1,000.00/mo |
| Backup (200 GiB beyond free) | 200 GiB × $0.095/GiB-mo | $19.00/mo |
| Data transfer (500 GB egress) | (500 GB less 100 GB free) × $0.09/GB | $36.00/mo |
| CloudWatch Database Insights Advanced | 4 vCPU × $0.0125/vCPU-hr × 730 hrs | $36.50/mo |
| RDS Proxy | 2 vCPU × $0.015/vCPU-hr × 730 hrs | $21.90/mo |
| Total monthly AWS RDS cost | ~$2,753.88/mo |
Best for high-throughput transactional systems with HA, predictable IOPS, and read scaling.
Annual cost: ~$33,047.

AWS RDS Cost Comparison Across Development, Production, and Enterprise Configurations (2026)
The same seven meters produce a 197x spread from the smallest viable setup to a full enterprise deployment.
| Component | Dev (db.t4g.micro) | Production SaaS (db.m5.large) | Enterprise (db.r6in.xlarge) |
|---|---|---|---|
| Instance compute | $11.68/mo | $259.88/mo | $1,515.48/mo |
| Storage | $2.30/mo | $23.00/mo | $125.00/mo |
| Provisioned IOPS | $0.00/mo | $0.00/mo | $1,000.00/mo |
| Backup | $0.00/mo | $0.00/mo | $19.00/mo |
| Data transfer | $0.00/mo | ~$0.50/mo | $36.00/mo |
| Add-on services | $0.00/mo | $8.04/mo | $58.40/mo |
| Total Monthly | $13.98/mo | ~$291.42/mo | ~$2,753.88/mo |
| Annual | ~$168/yr | ~$3,497/yr | ~$33,047/yr |
Rates verified August 2026, us-east-1, on-demand. Verify your specific configuration on the AWS Pricing Calculator.
For comparison against managed Postgres providers (Neon, Supabase, DigitalOcean, Aiven), the Managed PostgreSQL Comparison (2026): $0 to $475/month lays out side-by-side pricing.
If it’s specifically the serverless route you’re weighing, Neon swaps RDS’s instance meters for usage-based ones (CU-hours, storage, history), and we break down exactly how Neon’s usage-based model actually bills in our dedicated Neon pricing guide.
How AWS RDS Cost Compares to Self-Hosted PostgreSQL and BYOC Alternatives in 2026
Self-hosting PostgreSQL on raw EC2 or third-party infrastructure typically cuts total database costs by 40-70% against AWS RDS, but you take on failover, upgrades, monitoring and backups yourself. A managed platform running on the same AWS hardware is the third path, and on matched configurations it lands 34% below RDS on a production workload and 69% below on an enterprise one, without the operational burden.
The Managed vs Self-Hosted Database: Which Is Better for Your Startup? walks through the trade-off in detail.
A typical production SaaS workload, about $291/month on RDS Multi-AZ, costs roughly $150 to $180/month self-hosted on EC2 with manual HA, and $93.38/month fully managed on Selfhost.dev’s equivalent 2 vCPU / 8 GiB instance with the same 100 GB of storage.
Selfhost.dev runs managed Postgres on the same AWS hardware across 38 regions, includes connection pooling, PITR, alerts and monitoring rather than metering them, carries a 99.9% SLA, is drivable from Claude or Cursor across 161 MCP tools, and will run inside your own AWS account if you want the data to stay there. The AWS RDS vs Self-Hosted PostgreSQL: Complete Cost Comparison (2026) shows the full math on both paths.
For non-AWS infrastructure, AWS RDS vs Hetzner Cloud Cost: $3,150 vs $835 (2026) compares the same production workload. Savings reach 73% at scale due to Hetzner’s flat-rate storage and included bandwidth.
BYOC as the third option: lower AWS RDS cost without operational burden
- You pay cloud-native rates directly to the cloud provider
- A platform handles operations (provisioning, failover, upgrades, monitoring, backups)
- No vendor lock-in, database lives in your account
- No DBA hire needed, operational layer is automated
On matched hardware the gap widens with size, which is the opposite of what most teams expect. Priced against the three configurations above on Selfhost.dev’s own calculator: a dev database runs $12.60/month against RDS’s $13.98, about 10% less. A production 2 vCPU / 8 GiB instance with 100 GB runs $93.38/month against RDS’s $141.44, about 34% less. A comparable 4 vCPU / 32 GiB three-node setup with 500 GB runs $842.26/month against RDS’s $2,753.88, about 69% less.
The reason the gap grows is that it comes almost entirely from compute and from what RDS charges separately. Storage is not where we win, we price close to RDS per GB. What changes the total is that RDS’s enterprise bill carries $1,000/month of provisioned IOPS that gp3 covers with a 3,000 IOPS baseline, plus pooling, monitoring, alerts and PITR that RDS bills as RDS Proxy, CloudWatch and Database Insights line items.
If BYOC is new, What Is BYOC? A Smarter Alternative to Expensive Managed Databases explains the model in detail.
AWS RDS vs Self-Hosted PostgreSQL vs BYOC Cost Comparison (Same Production Workload, 2026)
Baseline: 2 vCPU / 8 GiB, 100 GiB gp3, 60-second monitoring. Selfhost.dev figures are live calculator quotes for us-east-1, August 2026.
| Approach | Typical Monthly Cost | Operational Burden | Best For |
|---|---|---|---|
| Selfhost.dev, managed on AWS | $93.38/mo | Low, the platform handles ops | Teams wanting RDS simplicity for a third less |
| AWS RDS, Single-AZ | $141.44/mo | None, fully managed | Small teams, no DBA, AWS-centric |
| AWS RDS, Multi-AZ | ~$291/mo | None, fully managed | HA-required production on AWS |
| Self-hosted on EC2 | ~$150 to $180/mo | High, you own ops, upgrades and failover | Teams with DevOps capacity |
| Self-hosted on Hetzner | ~$80 to $120/mo | High, plus self-managed networking | Cost-optimized, outside AWS |
Comparing options? See how Selfhost.dev compares to AWS RDS directly for a side-by-side breakdown.

AWS RDS Cost in 2026: Key Takeaways and Your Next Step
Understanding your full AWS RDS cost in 2026 requires more than checking the instance hourly rate, the real bill emerges from how compute, storage, Multi-AZ, backups, data transfer, and add-on services stack together. The seven independent billing meters compound rather than substitute, which is why production setups regularly bill 3-5x what the headline instance price suggests.
Key takeaways
- Compute is just one of seven meters. Storage, IOPS, Multi-AZ, backups, transfer, and add-ons add 50-150% on top.
- Reserved Instances only discount compute. Net savings land at 25-35%, not the 72% headline.
- Multi-AZ doubles compute and storage costs. The largest one-decision cost multiplier in AWS RDS.
- Extended Support and RDS Proxy are common surprise charges. Extended Support can double a small instance’s bill in Year 3+.
- Managed alternatives on the same AWS hardware cut 34% off a production workload and 69% off an enterprise one, with the gap widening as the instance grows.
The AWS RDS cost ceiling is structural, not a problem you can optimize away. The real question is whether managed-service convenience still matches your team’s stage and budget.
What to do next
Model your own configuration. The AWS Pricing Calculator will price your exact instance, storage and region against the rates above.
Then price the same thing without the seven meters. Selfhost.dev runs managed Postgres on the same AWS hardware at a set hourly rate, with pooling, PITR, alerts and monitoring included rather than metered. On the production configuration in this guide that is $93.38/month against RDS’s $141.44. See your own number on the Selfhost.dev calculator, 30 seconds, no signup needed.
If you would rather see it running than read about it, verify your account and you can have a live Postgres instance in about two minutes, paid from credits, paused whenever you want.
Still comparing? The AWS RDS alternatives shortlist ranks every option on cost, bill predictability, migration effort and lock-in.
Related Reading: More on AWS RDS Cost Optimization and Self-Hosting Alternatives
- Why AWS RDS Is Expensive Once Your Product Starts Growing : the narrative side of why RDS bills compound at the growth inflection point
- AWS RDS vs Self-Hosted PostgreSQL: Complete Cost Comparison (2026): line-by-line cost breakdown of RDS vs running PostgreSQL on raw EC2
- Managed PostgreSQL Comparison (2026): $0 to $475/month: side-by-side pricing across Neon, Supabase, DigitalOcean, Aiven, Railway, and AWS RDS
- AWS RDS vs Hetzner Cloud Cost: $3,150 vs $835 (2026): production workload pricing showing 73% savings at scale
- What Is BYOC? A Smarter Alternative to Expensive Managed Databases: explanation of the BYOC model and how it compares to RDS and traditional self-hosting
- PaaS Pricing Comparison (2026): your database is just one line in a full app-hosting bill, this prices the whole app across Railway, Render, Vercel, Fly.io, Heroku, DigitalOcean and Selfhost.dev, and shows where a flat server bundles the database so you skip RDS’s seven meters.
- Best Managed ClickHouse Service: 6 Honest Picks (2026): the same metered-bill trap shows up in analytics databases, and this ranks, the managed ClickHouse options and shows where serverless pricing (ClickHouse Cloud) surprises you just like RDS’s meters do.
Frequently Asked Questions
Does AWS RDS cost money?
Yes. AWS RDS bills from the first hour on any account opened on or after 15 July 2025. AWS retired the old 12-month, 750-hour RDS allowance for new accounts on that date and replaced it with Free Tier credits: a Free plan that runs up to 6 months or until the credits are spent, whichever comes first. RDS usage draws that balance down rather than sitting alongside it. Accounts created before 15 July 2025 may still hold the legacy 12-month allowances.
How much does AWS RDS cost per month?
AWS RDS cost ranges from ~$14/month for a development setup on db.t4g.micro to about $2,754/month for enterprise Multi-AZ with Provisioned IOPS and a read replica. Most production SaaS workloads land at $250-$500/month. The AWS RDS vs Self-Hosted PostgreSQL Complete Cost Comparison shows alternative pricing on EC2 and Hetzner Cloud.
Is RDS free in AWS?
Not since 15 July 2025, for new accounts. RDS is now free only while your Free Tier credits last, on a Free plan capped at 6 months. The old 750-hour, 12-month allowance survives only on accounts opened before that date. For genuinely persistent free tiers, the Managed PostgreSQL Comparison (2026): $0 to $475/month lists the providers that still offer one.
Is AWS RDS free to use?
No. A new AWS account gets Free Tier credits on a Free plan lasting up to 6 months, and every RDS hour draws that balance down. Most production setups exhaust it in weeks, and anything above db.t4g.micro bills at standard rates from day one. Once the credits are gone, all RDS usage is billable regardless of size.
Why is AWS RDS so expensive?
AWS RDS gets expensive because billing stacks across seven independent meters simultaneously: compute, storage, IOPS, Multi-AZ replication, backups, data transfer, and add-on services like RDS Proxy or Extended Support. Each meter scales independently. The Why AWS RDS Is Expensive Once Your Product Starts Growing blog explains the compounding effect in detail.
How is AWS RDS Extended Support priced?
AWS RDS Extended Support bills per vCPU per hour for engine versions past their end-of-standard-support date. Years 1-2 charge $0.100 per vCPU-hour and Year 3 onward charges $0.200 per vCPU-hour. A 4-vCPU db.m5.xlarge running PostgreSQL 12 past EOL accrues $292-$584/month in Extended Support fees alone on top of normal compute.
How much do AWS RDS snapshots cost?
AWS RDS automated snapshots are free up to 100% of your provisioned database size. Manual snapshots bill at $0.023/GB-month (standard S3 rates) and persist indefinitely until manually deleted. Snapshots from deleted instances continue billing. Snapshot export to S3 adds $0.013 per GB of source snapshot size as a one-time charge.
What is the AWS RDS pricing free tier?
It depends on when your account was opened. Accounts created before 15 July 2025 get 750 hours/month of a db.t2.micro, db.t3.micro, or db.t4g.micro Single-AZ instance on MySQL, PostgreSQL, or MariaDB for 12 months, plus 20 GB of General Purpose SSD and 20 GB of backup storage. Accounts created on or after that date get Free Tier credits instead, on a Free plan lasting up to 6 months, with RDS drawing from the same balance as every other service.
Based on this page: ·
On now or still deciding?
What's the bill you're looking at?
Roughly how much a month, all in?
Matched to RDS, we bill 3.0x less in ap-south-1 (Mumbai) and 2.1x less in us-east-1.
2 vCPU / 4 GB with 20 GB: $24.54/mo in Mumbai, $35.27/mo in us-east-1, against an estimated $73.75/mo on RDS.
See a real RDS bill, before and after →Neon meters compute hours and sleeps. We bill one fixed hourly rate and never sleep.
{price} Same number in a quiet month and a launch month.
Price your size →Two ways off the Supabase bill and you can keep the Supabase stack.
Self-host Supabase on a project server from about $0.02/hr or move just the database to a dedicated Postgres. {price}
How self-hosted Supabase works here →An enterprise client is moving production ClickHouse to us at up to 60% less.
Worked example: off ClickHouse Cloud Scale to always-on managed ClickHouse, $891.86 to $364.03 a month, 59%.
See the ClickHouse bill, before and after →A PaaS bills per service. A project server is one bill for everything on it.
Worked example: a studio consolidated client hosting onto project servers, $390.00 to $82.66 a month, 78%.
See that bill, before and after →A managed database for as low as $5 a month. Dedicated Postgres or MySQL on AWS, with restore, pooling and Multi-AZ failover when you want it.
{price} Postgres and MySQL price the same.
Price your size →A managed database for as low as $5 a month. Dedicated Postgres or MySQL on AWS, with restore, pooling and Multi-AZ failover when you want it.
{price}
Price your size →A managed database for as low as $5 a month. Dedicated Postgres or MySQL on AWS, with restore, pooling and Multi-AZ failover when you want it.
{price} Pay-as-you-go credits, top up from $5, no plan tiers.
Price your size →Most setups map onto two shapes: dedicated databases on AWS and a project server for everything else.
{price} Project servers from about $0.02/hr.
Price your size →Under $200 a month, the calculator beats a human reply. Price it there first.
What's it for?
Where do your users mostly sit?
Dedicated managed Postgres or MySQL, with point-in-time restore, pooling and Multi-AZ failover when you want it.
{region} If you need scale-to-zero or instant branching, Neon fits better than we do.
What managed Postgres includes →Managed ClickHouse for events at volume or TimescaleDB on managed Postgres for time-series next to your app data.
Worked example: off ClickHouse Cloud Scale, $891.86 to $364.03 a month, 59%. {region}
See the ClickHouse bill, before and after →pgvector on managed Postgres: one database instead of a second vendor.
{region} TimescaleDB is one click away on the same instance if you also have time-series.
When a dedicated vector DB is worth it →A project server from about $0.02/hr: push to GitHub and it deploys, several apps and their databases on one box.
Auto-deploy on push, PR previews, custom domains. The $5 welcome credit covers the start of it, no card and you can pause it to $0.
How Projects work →Project servers: many sites per server, one bill.
Worked example: a studio consolidated client hosting, $390.00 to $82.66 a month, 78%.
See that bill, before and after →Most things fit dedicated databases on AWS plus a project server for the apps.
{region}
See it in the live demo →What are you trying to do?
Which editor?
We migrate you, free, on every plan size.
An engineer runs it with you: match the instance, replicate or dump and restore, test the restore, cut over.
See one migration, step by step →The Supabase stack, one click, on a server of your own.
One click from the template, on a project server from about $0.02/hr. Nothing to build or wire up.
The Supabase deploy guide →Paste one config and your editor can run the whole backend.
Run it once, sign in when the browser opens and all 189 tools register.
An hourly instance you pause when done.
Restore a snapshot into a second instance, point staging at it, pause it when idle.
Price a small instance →Push to GitHub and it's live: auto-deploy, PR previews, custom domains.
A project server from about $0.02/hr runs several apps and their databases.
Deploy guides by framework →The one that matters most:
A managed database for as low as $5 a month. Project servers from about $0.02 an hour.
The smallest dedicated Postgres, a t4g.nano with 1 GB, is about $5.34 a month in ap-south-1 (Mumbai) and about $6.41 in us-east-1 and the $5 welcome credit covers its first days. Pay-as-you-go credits, top up from $5, no tiers.
Price your size →When production needs help, you talk to the engineers who run it and a founder reads every request.
Outages, restores, performance issues and migrations reach the people operating the platform. No response-time promise or uptime percentage we cannot prove.
How support and recovery work →Failover, point-in-time restore and backups cover the common ways a database fails.
Multi-AZ failover is a toggle on every engine, Postgres and MySQL add point-in-time restore and all engines get automated backups, live metrics and alerts.
The reliability and recovery details →We migrate you, free. An engineer plans and runs the move with you.
Match the instance, replicate or dump and restore, test the restore, cut over. Any plan size, no extra cost.
See one migration, step by step →Run it in your own AWS account with BYOC, in the region you need.
Encryption in transit and at rest on every managed instance. We do not hold SOC 2 or HIPAA today and we will say so before you spend an hour.
What we can and cannot certify →Fair enough. The live demo is the fastest way to see it.
Or claim $5 to start, no card needed.
Watch the live demo →Our founder
You've already sent us your setup from this browser. A founder will reply to it; no need to send it twice.
Got it. I'll reply to within two working days, from my own inbox.
Noted. One note when it changes, nothing else.
Run managed Postgres, your way.
Dedicated PostgreSQL on AWS with PITR, connection pooling, Multi-AZ, pgvector and BYOC. From about $5/mo and it pauses at a zero balance so there are no surprise invoices. Limited-time offer: get $5 in free credit at signup, no card needed.
Based on this page: ·
On now or still deciding?
What's the bill you're looking at?
Roughly how much a month, all in?
Matched to RDS, we bill 3.0x less in ap-south-1 (Mumbai) and 2.1x less in us-east-1.
2 vCPU / 4 GB with 20 GB: $24.54/mo in Mumbai, $35.27/mo in us-east-1, against an estimated $73.75/mo on RDS.
See a real RDS bill, before and after →Neon meters compute hours and sleeps. We bill one fixed hourly rate and never sleep.
{price} Same number in a quiet month and a launch month.
Price your size →Two ways off the Supabase bill and you can keep the Supabase stack.
Self-host Supabase on a project server from about $0.02/hr or move just the database to a dedicated Postgres. {price}
How self-hosted Supabase works here →An enterprise client is moving production ClickHouse to us at up to 60% less.
Worked example: off ClickHouse Cloud Scale to always-on managed ClickHouse, $891.86 to $364.03 a month, 59%.
See the ClickHouse bill, before and after →A PaaS bills per service. A project server is one bill for everything on it.
Worked example: a studio consolidated client hosting onto project servers, $390.00 to $82.66 a month, 78%.
See that bill, before and after →A managed database for as low as $5 a month. Dedicated Postgres or MySQL on AWS, with restore, pooling and Multi-AZ failover when you want it.
{price} Postgres and MySQL price the same.
Price your size →A managed database for as low as $5 a month. Dedicated Postgres or MySQL on AWS, with restore, pooling and Multi-AZ failover when you want it.
{price}
Price your size →A managed database for as low as $5 a month. Dedicated Postgres or MySQL on AWS, with restore, pooling and Multi-AZ failover when you want it.
{price} Pay-as-you-go credits, top up from $5, no plan tiers.
Price your size →Most setups map onto two shapes: dedicated databases on AWS and a project server for everything else.
{price} Project servers from about $0.02/hr.
Price your size →Under $200 a month, the calculator beats a human reply. Price it there first.
What's it for?
Where do your users mostly sit?
Dedicated managed Postgres or MySQL, with point-in-time restore, pooling and Multi-AZ failover when you want it.
{region} If you need scale-to-zero or instant branching, Neon fits better than we do.
What managed Postgres includes →Managed ClickHouse for events at volume or TimescaleDB on managed Postgres for time-series next to your app data.
Worked example: off ClickHouse Cloud Scale, $891.86 to $364.03 a month, 59%. {region}
See the ClickHouse bill, before and after →pgvector on managed Postgres: one database instead of a second vendor.
{region} TimescaleDB is one click away on the same instance if you also have time-series.
When a dedicated vector DB is worth it →A project server from about $0.02/hr: push to GitHub and it deploys, several apps and their databases on one box.
Auto-deploy on push, PR previews, custom domains. The $5 welcome credit covers the start of it, no card and you can pause it to $0.
How Projects work →Project servers: many sites per server, one bill.
Worked example: a studio consolidated client hosting, $390.00 to $82.66 a month, 78%.
See that bill, before and after →Most things fit dedicated databases on AWS plus a project server for the apps.
{region}
See it in the live demo →What are you trying to do?
Which editor?
We migrate you, free, on every plan size.
An engineer runs it with you: match the instance, replicate or dump and restore, test the restore, cut over.
See one migration, step by step →The Supabase stack, one click, on a server of your own.
One click from the template, on a project server from about $0.02/hr. Nothing to build or wire up.
The Supabase deploy guide →Paste one config and your editor can run the whole backend.
Run it once, sign in when the browser opens and all 189 tools register.
An hourly instance you pause when done.
Restore a snapshot into a second instance, point staging at it, pause it when idle.
Price a small instance →Push to GitHub and it's live: auto-deploy, PR previews, custom domains.
A project server from about $0.02/hr runs several apps and their databases.
Deploy guides by framework →The one that matters most:
A managed database for as low as $5 a month. Project servers from about $0.02 an hour.
The smallest dedicated Postgres, a t4g.nano with 1 GB, is about $5.34 a month in ap-south-1 (Mumbai) and about $6.41 in us-east-1 and the $5 welcome credit covers its first days. Pay-as-you-go credits, top up from $5, no tiers.
Price your size →When production needs help, you talk to the engineers who run it and a founder reads every request.
Outages, restores, performance issues and migrations reach the people operating the platform. No response-time promise or uptime percentage we cannot prove.
How support and recovery work →Failover, point-in-time restore and backups cover the common ways a database fails.
Multi-AZ failover is a toggle on every engine, Postgres and MySQL add point-in-time restore and all engines get automated backups, live metrics and alerts.
The reliability and recovery details →We migrate you, free. An engineer plans and runs the move with you.
Match the instance, replicate or dump and restore, test the restore, cut over. Any plan size, no extra cost.
See one migration, step by step →Run it in your own AWS account with BYOC, in the region you need.
Encryption in transit and at rest on every managed instance. We do not hold SOC 2 or HIPAA today and we will say so before you spend an hour.
What we can and cannot certify →Fair enough. The live demo is the fastest way to see it.
Or claim $5 to start, no card needed.
Watch the live demo →Our founder
You've already sent us your setup from this browser. A founder will reply to it; no need to send it twice.
Got it. I'll reply to within two working days, from my own inbox.
Noted. One note when it changes, nothing else.