# Databases Kit: Apache projects for small business

October 30, 2026. Research checked October 29, 2026. [Read the article](https://jwatte.com/blog/apache-databases-small-business/).

## How to use this kit

Start with the business application and its supported database. All sixteen operational projects here need an IT provider; the remaining three are developer components, so there are no owner-run installation cards. Use the questions to obtain a complete support and recovery quote. Treat examples as planning illustrations and keep real business records out of an unreviewed trial. We have not installed these products for the series.

## Detailed provider cards

### CouchDB

JSON document storage with replication. Example uses: a nine person HVAC company’s offline service checklists; a two-location retailer’s product notes. Needs IT for application code, conflicts, access permissions, compaction and recovery. [Single-node setup](https://docs.couchdb.org/en/stable/setup/single-node.html), [Docker](https://docs.couchdb.org/en/stable/install/docker.html) and [Linux packages](https://docs.couchdb.org/en/stable/install/unix.html) are official starting points. No universal CPU/RAM minimum is stated there; our Small 4 GiB host is a pilot allowance with no host redundancy. Current [3.5.3 release](https://blog.couchdb.org/2026/10/12/3-5-3/) includes security fixes; source builds must follow current Erlang requirements.

Alternative: [Cloudant Standard](https://cloud.ibm.com/catalog/services/cloudant), starting $109.43/month, hourly provisioned billing; a floor rather than a full quote. [Cloudant docs](https://cloud.ibm.com/docs/Cloudant?topic=Cloudant-ibm-cloud-public) say replication is not an automatic backup. Ask: Which application resolves conflicts? Can we restore one mistaken deletion? Which API differences affect migration? What throughput, storage, support and backup charges sit above the floor?

### Cassandra

Distributed records for an application designed around its queries. Example uses: a twenty-driver delivery firm’s events; an eighteen person ecommerce brand’s custom activity service. [5.0 installation](https://cassandra.apache.org/doc/5.0/cassandra/installing/installing.html) offers packages, archives and Docker with Java 11 or 17. [Hardware guidance](https://cassandra.apache.org/doc/5.0/cassandra/managing/operating/hardware.html) gives a production floor of two cores and 8 GB RAM/node. We assume three Medium nodes and replication factor three; this is a chosen scenario, not a universal minimum. [Replication documentation](https://cassandra.apache.org/doc/latest/cassandra/architecture/dynamo.html) explains independent node copies. Needs IT for data design, repair, compaction, capacity and recovery.

Alternative: [IBM Cassandra (Astra DB)](https://www.ibm.com/products/watsonx-data/pricing) starts at $341/month for the published starter scenario: 20 GB, 80 reads/second at 4 KB, 80 writes/second at 1 KB, listed peaks of 100. Pricing is indicative and country-dependent. Ask: Why is a simpler database insufficient? What happens with one node unavailable? Who runs repairs and tests backups? Which managed capacity and support options match our requirements?

### Kvrocks

Redis-protocol key lookups with disk-backed RocksDB storage. Example uses: a six person print shop’s rebuildable material-price cache; a three-location car wash’s rebuildable account-attribute lookup. It does not automatically support every Redis feature. [Getting started](https://kvrocks.apache.org/docs/getting-started/) supplies an official Docker image and password setup, with no universal numeric hardware minimum. We budget one Small 4 GiB host for a tiny pilot. Use [backup](https://kvrocks.apache.org/docs/backup/) and [replication](https://kvrocks.apache.org/docs/replication/) instructions separately. Needs IT for storage, client compatibility and restore/rebuild behavior.

Alternative: [Redis Cloud Essentials](https://redis.io/pricing/) from $5/month for the entry size. Our comparison assumes a 100 to 250 MB dataset; it does not buy the advertised maximum service size for $5. Ask: Which commands must work? Is the cache rebuildable from an authoritative system? How long can the application operate without it? What is the correct price after the dataset grows?

## Other provider cards

Monthly staffing ranges below are assumptions, not vendor benchmarks. Every managed alternative needs a compatibility assessment and complete quote.

**Accumulo.** Its [installation documentation](https://accumulo.apache.org/docs/2.x/administration/in-depth-install) adds a database to HDFS and ZooKeeper. The single-machine development route is not a production cluster; the documented combined-stack sizing commonly uses 4 to 8 cores and 8 to 32 GB RAM. A parts distributor with restricted supplier datasets might justify its specialist access controls. Allow 6 to 16 hours/month for a small stack. If simpler permissions suffice, [DigitalOcean managed PostgreSQL](https://www.digitalocean.com/pricing/managed-databases) starts at $15.15/month for one 1 GiB node with 10 GiB base storage. It does not reproduce Accumulo’s model or imply high availability. Provider questions: Are cell-level permissions truly needed? Who maintains HDFS and coordination as well as the database?

**AGE.** This adds graph queries to PostgreSQL. The [setup guide](https://age.apache.org/age-manual/master/intro/setup.html) offers a container with the extension; start with one compatible PostgreSQL host, then size data and queries. Its documented version combinations must match the selected release, since older instructions do not cover every newer PostgreSQL artifact. A twelve person property manager tracing vendors and properties could evaluate it with 1 to 3 administration hours/month for a bounded pilot. [Neo4j AuraDB Professional](https://neo4j.com/pricing/) starts at $0.09/GB-hour with a 1 GB minimum: our assumed 730-hour month is $66 rounded. Query and application migration remain separate. Provider questions: Which exact PostgreSQL and AGE versions are supported together? What query migration would a managed graph service require?

**AsterixDB.** Its [0.9.9 service guide](https://asterixdb.apache.org/docs/0.9.9/ncservice.html) demonstrates a controller and two node services on one machine before multi-machine deployment. That is a learning setup; production needs an operator to size storage and recovery, with our 6 to 16 hours/month cluster allowance. An eight person parts seller combining nested supplier records might investigate it. [DigitalOcean managed MongoDB](https://www.digitalocean.com/pricing/managed-databases) starts at $15.23/month for one 1 GiB node with 15 GiB base storage. Its APIs differ, so an application migration requires separate work. Provider questions: Which controller and node roles are production resources? How are nested-record imports and recovery tested?

**Geode.** This provides distributed data services. Its [official quick start](https://raw.githubusercontent.com/apache/geode/develop/geode-docs/getting_started/15_minute_quickstart_gfsh.html.md.erb) starts a locator and server, then adds a server for replication. Memory and persistent-data needs follow the working set. Older published prerequisite pages describe a different release line, so the provider must match current runtime requirements. A two-location retailer with a custom lookup system might evaluate it; allow 6 to 16 hours/month for a replicated deployment. Redis Cloud’s $5 entry is a simpler small-cache alternative only if the application can dispense with Geode-specific features. Provider questions: Which current release requirements replace the older manual? What working set must stay in memory?

**HBase.** The [reference guide](https://hbase.apache.org/book.html) distinguishes a single-JVM learning setup from distributed operation with Hadoop storage and coordination. Follow the selected release’s guide; the public reference also contains legacy material. A regional courier with a large operational history might justify this stack and 6 to 16 administration hours/month. [Cloudera](https://www.cloudera.com/products/pricing.html) publishes Operational Database at $0.08 per CCU-hour, excluding infrastructure and networking. A CCU is its compute-capacity unit; that unit price is not a complete small-business monthly bill. Provider questions: Which guide matches our chosen branch? Who owns Hadoop, coordination and region-server recovery?

**HugeGraph.** Its [server guide](https://hugegraph.apache.org/docs/quickstart/hugegraph/hugegraph-server/) describes local RocksDB storage and a distributed path with separate services. The provider must match runtime requirements to the release, size graph data and protect query access. A ten person wholesaler identifying substitute parts could begin with the local evaluation and 1 to 3 administration hours/month; distributed operation needs the cluster range. Neo4j AuraDB Professional’s $0.09/GB-hour, minimum 1 GB, is a published graph-service comparison, with different APIs and migration work. Provider questions: Which release-specific runtime and storage backend are selected? How is query authentication enforced?

**Ignite.** The [quick start](https://ignite.apache.org/docs/latest/quick-start/java) begins with one node and then another; application integration and partition recovery remain engineering work. Its [download page](https://ignite.apache.org/download/) recommends the 2.x line while separately describing 3.x. A seven person print distributor with custom availability queries might investigate it, allowing 6 to 16 hours/month for a replicated deployment. Redis Cloud’s $5 small-cache entry is an alternative only when distributed SQL and computation are unnecessary. No fixed hosted-Ignite support price was verified. Provider questions: Why choose this branch and its distributed features? What is the recovery plan for a lost partition?

**Kudu.** Its [installation guide](https://kudu.apache.org/docs/installation.html) describes one master without fault tolerance or three masters for tolerating one failure, and at least three tablet servers for replicated data. Roles need not each mean a separate physical machine; the provider must design that placement. A thirty person manufacturer combining changing job records and analytical scans might justify 6 to 16 administration hours/month. Cloudera Data Hub lists $0.04/CCU-hour before infrastructure. Request a complete cluster quote rather than multiplying that rate by an invented node size. Provider questions: How are master and tablet roles placed? What complete cloud bill sits behind the CCU rate?

**Pegasus (Incubating).** The [deployment guide](https://pegasus.apache.org/administration/deployment) calls for two or three MetaServers and at least three ReplicaServers, with coordination and suitable storage. An experienced provider must operate this distributed database. A niche ecommerce company with a custom high-volume backend might evaluate it with 6 to 16 administration hours/month. The Astra DB $341 starter in the detailed card is an alternative distributed-record budget. It uses a different database with its own compatibility requirements. Ask why the particular Pegasus architecture is necessary before commissioning it. Provider questions: Which failure requires the proposed replica topology? Is an existing managed database sufficient?

**Phoenix.** The [quick start](https://phoenix.apache.org/Phoenix-in-15-minutes-or-less.html) adds server components to HBase region servers and a matching SQL client. It inherits HBase’s operational stack; it does not turn that stack into a standalone small SQL database. A wholesaler already running HBase could use SQL access and allow 6 to 16 hours/month for the combined system. Cloudera’s published Data Hub $0.04/CCU-hour or Operational Database $0.08/CCU-hour are relevant service rates, with cloud infrastructure and a real configuration quote still required. Provider questions: Which HBase and Phoenix versions match? Who patches every region server and the clients?

**ResilientDB.** The [official project instructions](https://raw.githubusercontent.com/apache/incubator-resilientdb/master/README.md) describe replicated consensus with 3f + 1 replicas, so tolerating one faulty participant requires four replicas. A group of independent farms sharing a transaction record might need that trust model; most single businesses do not. Allow 6 to 16 administration hours/month plus specialist application work. Managed PostgreSQL from $15.15/month uses a simpler central authority model and cannot reproduce Byzantine-fault-tolerant consensus. No hosted ResilientDB price was verified, so no equivalent-service savings claim is made. Provider questions: Do participants actually distrust one central administrator? Who operates all four replicas and application consensus logic?

**Rya.** Its [documentation](https://rya.apache.org/documentation/) and [official quick start](https://raw.githubusercontent.com/apache/rya/master/extras/rya.manual/src/site/markdown/quickstart.md) combine RDF/SPARQL handling with Accumulo or MongoDB and a web application. A nonprofit linking program and funding relationships might investigate it, but the database and web stack need specialists and our 6 to 16 hours/month allowance. [Amazon Neptune](https://aws.amazon.com/neptune/pricing/) gives a US East (N. Virginia) example at $0.348/hour for db.r5.large. Storage, I/O and backup costs are additional; the semantic-graph service has different interfaces and operating behavior from Rya. Provider questions: Which backing store and web runtime are supported? What graph queries justify the additional stack?

**ShardingSphere.** The [proxy quick start](https://shardingsphere.apache.org/document/current/en/quick-start/shardingsphere-proxy-quick-start/) adds a proxy in front of backend databases. One proxy is a pilot, while resilience covers this routing layer and the actual databases. A multi-store retailer whose measured SQL workload has outgrown its current design might justify 6 to 16 administration hours/month for the combined deployment. First compare a suitably sized managed PostgreSQL service, whose DigitalOcean entry is $15.15/month. Use that entry to start a sizing discussion; a sharded workload will usually need more capacity. Provider questions: What measured limit requires sharding? Who restores a consistent state across the backend databases?

## Full group table

Incubating means not yet fully endorsed by the ASF; see [its explanation](https://incubator.apache.org/policy/incubation.html). Archive dates are file timestamps; registry records can lag. Accumulo’s alpha is a preview. HBase’s catalog row is the latest dated 2.6 line; its download page labels 2.5.16 stable. Developer components are GraphAr, Jackrabbit and PouchDB; they do not become owner applications because a database appears in their names.

| Project | What it does | Who runs it | Latest release and date |
|---|---|---|---|
| [Apache Accumulo](https://accumulo.apache.org) | Stores large datasets with access controls on individual data cells. | Your IT department or provider | [4.0.0-alpha-1 (preview, registry); 2026-09-24](https://projects.apache.org/json/foundation/projects.json) |
| [Apache AGE](https://age.apache.org/) | Adds graph queries to PostgreSQL so applications can explore relationships in business data. | Your IT department or provider | [1.8.0; 2026-08-06 (archive date)](https://downloads.apache.org/age/PG18/1.8.0/) |
| [Apache Asterixdb](https://asterixdb.apache.org) | Stores and queries varied business data, including nested documents. | Your IT department or provider | [0.9.9; 2024-03-05 (archive date)](https://downloads.apache.org/asterixdb/asterixdb-0.9.9/) |
| [Apache Cassandra](http://cassandra.apache.org) | Stores operational data across servers for applications that need continued availability. | Your IT department or provider | [5.0.9; 2026-08-07](https://cassandra.apache.org/_/download.html) |
| [Apache CouchDB](https://couchdb.apache.org/) | Stores documents and synchronizes copies between servers and devices. | Your IT department or provider | [3.5.3; 2026-10-12](https://blog.couchdb.org/2026/10/12/3-5-3/) |
| [Apache Geode](https://geode.apache.org) | Shares frequently used application data across servers. | Your IT department or provider | [2.0.3; 2026-09-19](https://github.com/apache/geode/releases/tag/rel/v2.0.3) |
| [Apache GraphAr (Incubating)](https://graphar.incubator.apache.org/) | Defines files for storing graph data that different analysis tools can share. | Your developer | [0.13.0-incubating; 2025-08-29](https://github.com/apache/incubator-graphar/releases/tag/v0.13.0) |
| [Apache HBase](https://hbase.apache.org) | Stores very large tables for applications built on Hadoop. | Your IT department or provider | [2.6.7; 2026-10-01](https://hbase.apache.org/downloads/) |
| [Apache Hugegraph](https://hugegraph.apache.org) | Stores and queries connected data such as customers, products and their relationships. | Your IT department or provider | [1.7.0; 2025-11-28](https://hugegraph.apache.org/docs/download/download/) |
| [Apache Ignite](http://ignite.apache.org) | Combines distributed data storage and computation for demanding applications. | Your IT department or provider | [2.18.0; 2026-04-26](https://ignite.apache.org/download/) |
| [Apache Jackrabbit](http://jackrabbit.apache.org/) | Provides a content repository for applications that store documents and their metadata. | Your developer | [Core 2.22.5; Oak 2.8.0; 2026-10-07 core; 2026-10-20 Oak](https://jackrabbit.apache.org/jcr/downloads.html) |
| [Apache Kudu](https://kudu.apache.org) | Stores analytical tables that need both fast scans and ongoing updates. | Your IT department or provider | [1.18.1; 2026-01-09](https://kudu.apache.org/releases/) |
| [Apache Kvrocks](https://kvrocks.apache.org) | Stores large key and value datasets through a Redis compatible interface. | Your IT department or provider | [2.17.0; 2026-09-19](https://github.com/apache/kvrocks/releases/tag/v2.17.0) |
| [Apache Pegasus (Incubating)](https://pegasus.incubator.apache.org/) | Stores application data across servers as keys and values. | Your IT department or provider | [2.5.0-incubating; 2023-12-01](https://pegasus.incubator.apache.org/docs/downloads) |
| [Apache Phoenix](http://phoenix.apache.org) | Lets applications query HBase data using SQL. | Your IT department or provider | [5.3.2; 2026-07-04](https://phoenix.apache.org/downloads/) |
| [Apache PouchDB (Incubating)](https://pouchdb.apache.org) | Stores application data in a browser and synchronizes it with compatible databases. | Your developer | [9.0.0; 2024-05-24 (registry record)](https://projects.apache.org/json/foundation/projects.json) |
| [Apache Resilientdb](https://resilientdb.apache.org) | Provides a replicated ledger for applications that need agreement on recorded transactions. | Your IT department or provider | [1.13.0; 2026-09-30 (archive date)](https://downloads.apache.org/resilientdb/1.13.0/) |
| [Apache Rya](https://rya.apache.org) | Stores and queries linked facts for applications using semantic data. | Your IT department or provider | [4.0.1; 2020-12-22](https://rya.apache.org/download/) |
| [Apache Shardingsphere](https://shardingsphere.apache.org) | Coordinates database access across multiple databases and servers. | Your IT department or provider | [5.5.3; 2026-03-01](https://shardingsphere.apache.org/document/current/en/downloads/) |

## Cost worksheet

Application and required queries: ______. Authoritative records: ______. Maximum tolerable data loss/outage: ______. Low/high stored data and backup volume: ______ / ______. Nodes and replication design: ______. Setup hours: ______ / ______. Monthly administration: ______ / ______. Managed plan and region: ______. Price date and billing term: ______. Extra storage/transfer/support: ______. Development and migration quote for each option: ______.

The shared method uses $68.09 per administration hour, an employee compensation estimate from [BLS wages](https://data.bls.gov/oesprofile/?areas=INDUSTRY,STATE,MSA&major_group=150000&measure=01&occupation=151244&year=2025) divided by the [70.0% wages share](https://data.bls.gov/timeseries/CMU2020000000000P). An outside IT provider bills more than this proxy, so get a quote. [DigitalOcean](https://www.digitalocean.com/pricing/droplets) supplies our Small server at $24/month (2 vCPU, 4 GiB RAM, 80 GiB disk), Medium at $48 (4 vCPU, 8 GiB, 160 GiB), and Large at $252 (8 dedicated vCPU, 32 GiB, 100 GiB). These are monthly caps without an annual commitment. [Backblaze B2](https://www.backblaze.com/cloud-storage/pricing) costs $6.95 per decimal TB/month after the first 10 GB free. All were checked October 29, 2026.

Setup = hours × $68.09. Monthly self hosted = server charges + max(0, retained backup GB minus 10) × $0.00695 + administration hours × $68.09. The 36 month total adds setup to 36 monthly bills; savings subtract that total from the alternative. Break even is setup divided by positive monthly savings, rounded up; otherwise it is never. Calculations retain cents internally and display whole dollars.

Our assumptions are 4 to 12 setup hours and 1 to 3 administration hours/month for a bounded single server, or 24 to 60 and 6 to 16 for a small cluster. These exclude application development, migration, training, incidents and on call coverage on both sides. Alternative setup is assumed zero incremental infrastructure labor in these examples. Implementation still takes work. Backup volumes include retained copies. We assume no chargeable backup requests or excess transfer; add them if your usage exceeds the provider allowances. Low and high cases are planning ranges to replace with measured requirements.

### CouchDB filled example

One $24 Small server; 10 to 50 GB retained backups; 4 to 12 setup hours; 1 to 3 administration hours/month. Low monthly = $24 + $0 + $68.09 = $92.09; high = $24 + 40 × $0.00695 + 3 × $68.09 = $228.548. [Cloudant](https://cloud.ibm.com/catalog/services/cloudant) starting floor is $109.43/month, held fixed for 36 months. Unauthenticated unit prices were not verified; actual usage, backup and support can add charges. A single node is not equivalent to managed replication.

| CouchDB single server | Low labor/workload | High labor/workload |
|---|---:|---:|
| Setup labor | $272 | $817 |
| Monthly self hosted | $92 | $229 |
| Self hosted, 36 months | $3,588 | $9,045 |
| Cloudant starting floor, monthly | $109 | $109 |
| Cloudant starting floor, 36 months | $3,939 | $3,939 |
| Self hosting savings, 36 months | $352 | Negative $5,105 |
| Break even month | 16 | Never |

### Cassandra filled example

Three $48 Medium nodes; 100 to 300 GB retained backups; 24 to 60 setup hours; 6 to 16 administration hours/month. Low monthly = 3 × $48 + 90 × $0.00695 + 6 × $68.09. High = 3 × $48 + 290 × $0.00695 + 16 × $68.09. [Astra DB](https://www.ibm.com/products/watsonx-data/pricing) starter is $341/month for its published indicative scenario, held constant; a larger workload needs a new quote. The provider must validate its capacity and pricing for the actual country and usage.

| Cassandra three nodes | Low labor/workload | High labor/workload |
|---|---:|---:|
| Setup labor | $1,634 | $4,085 |
| Monthly self hosted | $553 | $1,235 |
| Self hosted, 36 months | $21,548 | $48,562 |
| Astra DB starter, monthly | $341 | $341 |
| Astra DB starter, 36 months | $12,276 | $12,276 |
| Self hosting savings, 36 months | Negative $9,272 | Negative $36,286 |
| Break even month | Never | Never |

### Kvrocks filled example

One $24 Small node; 1 to 5 GB retained backups, within B2 free storage; 4 to 12 setup hours; 1 to 3 administration hours/month. Monthly = $24 + $0 + hours × $68.09. [Redis Cloud](https://redis.io/pricing/) entry is $5/month, assuming a compatible 100 to 250 MB dataset and required commands. A larger disk-backed dataset would require a different comparison. Neither this one-node pilot nor the tiny managed price is a universal availability guarantee.

| Kvrocks small cache | Low labor/workload | High labor/workload |
|---|---:|---:|
| Setup labor | $272 | $817 |
| Monthly self hosted | $92 | $228 |
| Self hosted, 36 months | $3,588 | $9,035 |
| Redis Cloud entry, monthly | $5 | $5 |
| Redis Cloud entry, 36 months | $180 | $180 |
| Self hosting savings, 36 months | Negative $3,408 | Negative $8,855 |
| Break even month | Never | Never |

### Setup time when Claude or ChatGPT does the typing

October 30, 2026 update. See the [post’s AI setup comparison](https://jwatte.com/blog/apache-databases-small-business/#setup-time-when-claude-or-chatgpt-does-the-typing).

We keep the CouchDB example: one $24 Small server and 10 to 50 GB of backups. Its single-node failure limits remain. All hours are our planning assumptions, not measured installations. Human review is already included: 1 to 3 setup hours and 0.5 to 1.5 hours/month. Low and high are paired scenarios. We keep hosting and backups unchanged, with monthly H = $24 + max(0, retained GB minus 10) × $0.00695. Existing exclusions still apply.

| CouchDB: paired planning cases | Low | High |
|---|---:|---:|
| Manual setup hours | 4 | 12 |
| AI setup share; hours | 50%; 2 | 80%; 9.6 |
| Manual administration hours/month | 1 | 3 |
| AI administration share; hours/month | 75%; 0.75 | 100%; 3 |
| Manual total, 36 months | $3,588 | $9,045 |
| Claude Pro subscription total, 36 months | $3,039 | $9,081 |
| ChatGPT Plus subscription total, 36 months | $3,039 | $9,081 |
| API pay per token: Claude Opus 5.5 total, 36 months | $2,853 | $8,995 |
| API pay per token: GPT-6 Astra total, 36 months | $2,876 | $9,166 |
| Opus setup/monthly token charges | $0.764 / $0.382 | $6 / $3 |
| Astra setup/monthly token charges | $1.970 / $0.985 | $15 / $7.50 |

Let S/A be manual setup/monthly hours, q/r the AI shares, H monthly hosting plus backups, and T0/Tm setup/monthly tool charges. We use $68.09/hour and round totals only after calculation.

```text
Manual total = S × $68.09 + 36 × (H + A × $68.09)
AI total = S × q × $68.09 + T0 + 36 × (H + A × r × $68.09 + Tm)
Savings = Manual total minus AI total
```

AI setup = S × q × $68.09 + T0. AI monthly = H + A × r × $68.09 + Tm. Substitute your own measured hours before deciding.

Claude Pro and ChatGPT Plus each cost $20/month, billed monthly. We charge one full separate setup month, then 25% ($5) for each of 36 operating months only if other work uses the remaining allowance. Without sharing, charge the full price and add $540 to either subscription total. Never allocate more than 100% across projects. Equal prices do not imply equal usage limits or results; extra usage and repeated setup months cost more. API billing is separate, with no assumed subscription credits.

Our cumulative API setup budget is 200,000 to 1,000,000 input tokens and 20,000 to 100,000 output tokens; monthly administration is 100,000 to 500,000 input and 10,000 to 50,000 output. These are our assumptions including retries, repeated context and billed reasoning, not a single prompt. Keep each request in the applicable short context pricing band. Low input is 30% ordinary, 10% short cache writes and 60% cache reads; high input is 100% ordinary with caching disabled. Do not assume a cache survives between monthly sessions.

Standard API rates per million tokens, read October 29, 2026: Claude Opus 5.5 input $4, output $20, cache read $0.20, short cache write $5; GPT-6 Astra input $10, output $50, cache read $1, short cache write $12.50. Writes replace ordinary input charges. No batch discount, paid browser service, premium routing or storage charge is assumed.

```text
API tool cost = (ordinary input × input rate + cache writes × write rate
                + cache reads × read rate + output × output rate) / 1,000,000
Opus low setup = 0.06 × $4 + 0.02 × $5 + 0.12 × $0.20 + 0.02 × $20 = $0.764
Astra low setup = 0.06 × $10 + 0.02 × $12.50 + 0.12 × $1 + 0.02 × $50 = $1.970
```

The base uses free local Desktop Commander or Codex CLI. Optional Desktop Commander Remote Pro is $20/month; one setup month plus a justified 25% monthly share adds $200 over this period. With exclusive use, its full monthly fee applies. Model usage and hosting are separate.

The high case saves no monthly labor. Compared with manual work, shared subscriptions become more expensive in month 29, Opus API in month 53 and Astra API in month 20. For a slower-work sensitivity, use 119% of manual hours plus tool charges; that is a sensitivity inspired by a coding study, not a deployment forecast. [92] The conclusion stays: the low case beats Cloudant’s $3,939 starting floor over 36 months; the high case loses. Managed replication and single-node CouchDB still differ.

Use a staging or new server; never paste secrets into chat. Approve commands or use a restricted sandbox, and watch browser actions. Use supported subscriptions under human direction and API keys for custom automation. Review changes and test a database restore and conflicting offline edits; backups and updates remain our responsibility.

## Recovery and first-week evaluation

On day 1 identify the application, its supported version and one business query. On day 2 have the provider load representative sample data into an isolated trial. On day 3 test the actual query and a denied permission, then observe a lost connection. On day 4 restore an independent backup and reconcile record counts and business identifiers. On day 5 review the evidence and the full quote, including responsibilities retained on the managed service.

Keep backup retention, encryption, credential recovery and restore ownership in writing. For CouchDB, test conflicting edits. For Cassandra, test the chosen consistency settings during node loss. For Kvrocks used as a cache, rebuild it from the authoritative store and check that the application behaves correctly while it is empty. Do not copy a database directory and assume it is a consistent backup without following the selected project’s procedure.

Review [CouchDB security](https://docs.couchdb.org/en/stable/cve/index.html), [Cassandra community/security guidance](https://cassandra.apache.org/_/community.html), [Kvrocks security](https://kvrocks.apache.org/community/security/) and [ASF security](https://www.apache.org/security/). [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0) permits commercial use under conditions but supplies no warranty or trademark license. All pricing was checked October 29, 2026, in US dollars before tax; no trial credit is used.

## Fact-check notes and sources

Official pages and the series catalog evidence were read October 29, 2026. Prices are US dollars before tax; billing terms appear with each comparison. We have not installed these projects for this series.

1. [Single-node setup](https://docs.couchdb.org/en/stable/setup/single-node.html). Read October 29, 2026, 21:37:33 UTC.
2. [Docker](https://docs.couchdb.org/en/stable/install/docker.html). Read October 29, 2026, 21:38:24 UTC.
3. [Linux packages](https://docs.couchdb.org/en/stable/install/unix.html). Read October 29, 2026, 21:37:32 UTC.
4. [3.5.3 release](https://blog.couchdb.org/2026/10/12/3-5-3/). Read October 29, 2026, 21:41:22 UTC.
5. [Cloudant Standard](https://cloud.ibm.com/catalog/services/cloudant). Read October 29, 2026, approximately 21:55 UTC, supplied browser review.
6. [Cloudant docs](https://cloud.ibm.com/docs/Cloudant?topic=Cloudant-ibm-cloud-public). Read October 29, 2026, 21:39:33 UTC.
7. [5.0 installation](https://cassandra.apache.org/doc/5.0/cassandra/installing/installing.html). Read October 29, 2026, 21:39:59 UTC.
8. [Hardware guidance](https://cassandra.apache.org/doc/5.0/cassandra/managing/operating/hardware.html). Read October 29, 2026, 21:39:58 UTC.
9. [Replication documentation](https://cassandra.apache.org/doc/latest/cassandra/architecture/dynamo.html). Read October 29, 2026, 21:37:35 UTC.
10. [IBM Cassandra (Astra DB)](https://www.ibm.com/products/watsonx-data/pricing). Read October 29, 2026, 21:38:23 UTC.
11. [Getting started](https://kvrocks.apache.org/docs/getting-started/). Read October 29, 2026, 21:37:33 UTC.
12. [backup](https://kvrocks.apache.org/docs/backup/). Read October 29, 2026, 21:40:28 UTC.
13. [replication](https://kvrocks.apache.org/docs/replication/). Read October 29, 2026, 21:38:24 UTC.
14. [Redis Cloud Essentials](https://redis.io/pricing/). Read October 29, 2026, 21:38:35 UTC.
15. [installation documentation](https://accumulo.apache.org/docs/2.x/administration/in-depth-install). Read October 29, 2026, 21:40:30 UTC.
16. [DigitalOcean managed PostgreSQL](https://www.digitalocean.com/pricing/managed-databases). Read October 29, 2026, 21:38:25 UTC.
17. [setup guide](https://age.apache.org/age-manual/master/intro/setup.html). Read October 29, 2026, 21:37:43 UTC.
18. [Neo4j AuraDB Professional](https://neo4j.com/pricing/). Read October 29, 2026, 21:38:26 UTC.
19. [0.9.9 service guide](https://asterixdb.apache.org/docs/0.9.9/ncservice.html). Read October 29, 2026, 21:38:24 UTC.
20. [official quick start](https://raw.githubusercontent.com/apache/geode/develop/geode-docs/getting_started/15_minute_quickstart_gfsh.html.md.erb). Read October 29, 2026, 21:40:58 UTC.
21. [reference guide](https://hbase.apache.org/book.html). Read October 29, 2026, 21:37:43 UTC.
22. [Cloudera](https://www.cloudera.com/products/pricing.html). Read October 29, 2026, 21:40:28 UTC.
23. [server guide](https://hugegraph.apache.org/docs/quickstart/hugegraph/hugegraph-server/). Read October 29, 2026, 21:38:25 UTC.
24. [quick start](https://ignite.apache.org/docs/latest/quick-start/java). Read October 29, 2026, 21:37:44 UTC.
25. [download page](https://ignite.apache.org/download/). Read October 29, 2026, 21:54:31 UTC.
26. [installation guide](https://kudu.apache.org/docs/installation.html). Read October 29, 2026, 21:37:44 UTC.
27. [deployment guide](https://pegasus.apache.org/administration/deployment). Read October 29, 2026, 21:39:32 UTC.
28. [quick start](https://phoenix.apache.org/Phoenix-in-15-minutes-or-less.html). Read October 29, 2026, 21:37:44 UTC.
29. [official project instructions](https://raw.githubusercontent.com/apache/incubator-resilientdb/master/README.md). Read October 29, 2026, 21:39:58 UTC.
30. [documentation](https://rya.apache.org/documentation/). Read October 29, 2026, 21:37:45 UTC.
31. [official quick start](https://raw.githubusercontent.com/apache/rya/master/extras/rya.manual/src/site/markdown/quickstart.md). Read October 29, 2026, 21:40:59 UTC.
32. [Amazon Neptune](https://aws.amazon.com/neptune/pricing/). Read October 29, 2026, 21:41:22 UTC.
33. [proxy quick start](https://shardingsphere.apache.org/document/current/en/quick-start/shardingsphere-proxy-quick-start/). Read October 29, 2026, 21:37:45 UTC.
34. [its explanation](https://incubator.apache.org/policy/incubation.html). Read October 29, 2026, 21:41:51 UTC.
35. [Apache Accumulo](https://accumulo.apache.org). Read October 29, 2026, 21:12:49 UTC (catalog observation).
36. [4.0.0-alpha-1 (preview, registry); 2026-09-24](https://projects.apache.org/json/foundation/projects.json). Read October 29, 2026, 21:00:00 UTC (catalog snapshot).
37. [Apache AGE](https://age.apache.org/). Read October 29, 2026, 21:12:49 UTC (catalog observation).
38. [1.8.0; 2026-08-06 (archive date)](https://downloads.apache.org/age/PG18/1.8.0/). Read October 29, 2026, 21:26:01 UTC (catalog observation).
39. [Apache Asterixdb](https://asterixdb.apache.org). Read October 29, 2026, 21:12:52 UTC (catalog observation).
40. [0.9.9; 2024-03-05 (archive date)](https://downloads.apache.org/asterixdb/asterixdb-0.9.9/). Read October 29, 2026, 21:19:42 UTC (catalog observation).
41. [Apache Cassandra](http://cassandra.apache.org). Read October 29, 2026, 21:12:53 UTC (catalog observation).
42. [5.0.9; 2026-08-07](https://cassandra.apache.org/_/download.html). Read October 29, 2026, 21:37:32 UTC.
43. [Apache CouchDB](https://couchdb.apache.org/). Read October 29, 2026, 21:39:34 UTC.
44. [Apache Geode](https://geode.apache.org). Read October 29, 2026, 21:38:35 UTC.
45. [2.0.3; 2026-09-19](https://github.com/apache/geode/releases/tag/rel/v2.0.3). Read October 29, 2026, 21:14:09 UTC (catalog observation).
46. [Apache GraphAr (Incubating)](https://graphar.incubator.apache.org/). Read October 29, 2026, 21:13:52 UTC (catalog observation).
47. [0.13.0-incubating; 2025-08-29](https://github.com/apache/incubator-graphar/releases/tag/v0.13.0). Read October 29, 2026, 21:17:33 UTC (catalog observation).
48. [Apache HBase](https://hbase.apache.org). Read October 29, 2026, 21:13:38 UTC (catalog observation).
49. [2.6.7; 2026-10-01](https://hbase.apache.org/downloads/). Read October 29, 2026, 21:54:31 UTC.
50. [Apache Hugegraph](https://hugegraph.apache.org). Read October 29, 2026, 21:13:38 UTC (catalog observation).
51. [1.7.0; 2025-11-28](https://hugegraph.apache.org/docs/download/download/). Read October 29, 2026, 21:14:10 UTC (catalog observation).
52. [Apache Ignite](http://ignite.apache.org). Read October 29, 2026, 21:13:38 UTC (catalog observation).
53. [Apache Jackrabbit](http://jackrabbit.apache.org/). Read October 29, 2026, 21:54:43 UTC.
54. [Core 2.22.5; Oak 2.8.0; 2026-10-07 core; 2026-10-20 Oak](https://jackrabbit.apache.org/jcr/downloads.html). Read October 29, 2026, 21:54:44 UTC.
55. [Apache Kudu](https://kudu.apache.org). Read October 29, 2026, 21:13:39 UTC (catalog observation).
56. [1.18.1; 2026-01-09](https://kudu.apache.org/releases/). Read October 29, 2026, 21:14:33 UTC (catalog observation).
57. [Apache Kvrocks](https://kvrocks.apache.org). Read October 29, 2026, 21:13:39 UTC (catalog observation).
58. [2.17.0; 2026-09-19](https://github.com/apache/kvrocks/releases/tag/v2.17.0). Read October 29, 2026, 21:40:28 UTC.
59. [Apache Pegasus (Incubating)](https://pegasus.incubator.apache.org/). Read October 29, 2026, 21:13:53 UTC (catalog observation).
60. [2.5.0-incubating; 2023-12-01](https://pegasus.incubator.apache.org/docs/downloads). Read October 29, 2026, 21:17:54 UTC (catalog observation).
61. [Apache Phoenix](http://phoenix.apache.org). Read October 29, 2026, 21:13:44 UTC (catalog observation).
62. [5.3.2; 2026-07-04](https://phoenix.apache.org/downloads/). Read October 29, 2026, 21:22:21 UTC (catalog observation).
63. [Apache PouchDB (Incubating)](https://pouchdb.apache.org). Read October 29, 2026, 21:13:39 UTC (catalog observation).
64. [Apache Resilientdb](https://resilientdb.apache.org). Read October 29, 2026, 21:37:44 UTC.
65. [1.13.0; 2026-09-30 (archive date)](https://downloads.apache.org/resilientdb/1.13.0/). Read October 29, 2026, 21:16:01 UTC (catalog observation).
66. [Apache Rya](https://rya.apache.org). Read October 29, 2026, 21:13:45 UTC (catalog observation).
67. [4.0.1; 2020-12-22](https://rya.apache.org/download/). Read October 29, 2026, 21:15:42 UTC (catalog observation).
68. [Apache Shardingsphere](https://shardingsphere.apache.org). Read October 29, 2026, 21:13:45 UTC (catalog observation).
69. [5.5.3; 2026-03-01](https://shardingsphere.apache.org/document/current/en/downloads/). Read October 29, 2026, 21:15:57 UTC (catalog observation).
70. [BLS wages](https://data.bls.gov/oesprofile/?areas=INDUSTRY,STATE,MSA&major_group=150000&measure=01&occupation=151244&year=2025). Read October 29, 2026, 21:12:08 UTC.
71. [70.0% wages share](https://data.bls.gov/timeseries/CMU2020000000000P). Read October 29, 2026, 21:17:59 UTC.
72. [DigitalOcean](https://www.digitalocean.com/pricing/droplets). Read October 29, 2026, 21:10:42 UTC.
73. [Backblaze B2](https://www.backblaze.com/cloud-storage/pricing). Read October 29, 2026, 21:10:42 UTC.
74. [CouchDB security](https://docs.couchdb.org/en/stable/cve/index.html). Read October 29, 2026, 21:40:28 UTC.
75. [Cassandra community/security guidance](https://cassandra.apache.org/_/community.html). Read October 29, 2026, 21:41:51 UTC.
76. [Kvrocks security](https://kvrocks.apache.org/community/security/). Read October 29, 2026, 21:39:59 UTC.
77. [ASF security](https://www.apache.org/security/). Read October 29, 2026, 21:39:57 UTC.
78. [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0). Read October 29, 2026, 21:39:56 UTC.

79. [Claude plans](https://claude.com/pricing). Read October 29, 2026, 23:31:08 UTC; reused for this update.
80. [ChatGPT plans](https://chatgpt.com/pricing/). Read October 29, 2026, about 23:27 UTC in a regular browser.
81. [Codex plan access](https://help.openai.com/en/articles/11369540-using-codex-with-your-chatgpt-plan). Read October 29, 2026, 23:31:16 UTC; reused for this update.
82. [Codex CLI](https://developers.openai.com/codex/cli). Read October 29, 2026, 23:32:01 UTC; reused for this update.
83. [Desktop Commander pricing](https://desktopcommander.app/pricing/). Read October 29, 2026, 23:31:08 UTC; reused for this update.
84. [Desktop Commander setup and permissions](https://raw.githubusercontent.com/wonderwhy-er/DesktopCommanderMCP/main/README.md). Read October 29, 2026, 23:31:41 UTC; reused for this update.
85. [Claude in Chrome](https://support.claude.com/en/articles/12012173-get-started-with-claude-in-chrome). Read October 29, 2026, 23:31:11 UTC; reused for this update.
86. [Chrome permissions](https://support.claude.com/en/articles/12902446-claude-in-chrome-permissions-guide). Read October 29, 2026, 23:31:09 UTC; reused for this update.
87. [ChatGPT cloud browser](https://help.openai.com/en/articles/20001280-using-cloud-browser-in-chatgpt). Read October 29, 2026, 23:32:01 UTC; reused for this update.
88. [Claude API pricing](https://platform.claude.com/docs/en/about-claude/pricing). Read October 29, 2026, 23:31:14 UTC; reused for this update.
89. [OpenAI API pricing](https://developers.openai.com/api/docs/pricing). Read October 29, 2026, 23:31:13 UTC; reused for this update.
90. [OpenAI prompt caching](https://developers.openai.com/api/docs/guides/prompt-caching). Read October 29, 2026, 23:32:00 UTC; reused for this update.
91. [METR coding study limitations](https://metr.org/blog/2026-02-24-uplift-update/). Read October 29, 2026, 23:31:09 UTC; reused for this update.

92. [METR July 2025 developer experiment](https://metr.org/blog/2025-07-10-early-2025-ai-experienced-os-dev-study/). Read October 29, 2026, 23:30:58 UTC; reused for this update.

*Apache, the Apache feather logo and the Apache project names are trademarks of The Apache Software Foundation. This site is not affiliated with or endorsed by the ASF. Other product names are used for identification only.*