Public Cloud
ComputeContainersStorageDatabasesAIAnalyticsIaCQuantum

Public cloud — Databases

Managed databases, run on intSignal infrastructure.

MongoDB, PostgreSQL, MySQL, and Valkey — provisioned, patched, backed up, and tuned by intSignal. HA topology, point-in-time recovery, and encrypted everything. Other engines on request.

MongoDB

PostgreSQL

MySQL

Valkey

Schedule consultation  ⟶See the engines

Single operator

One team for all four engines and the platform underneath.

Encrypted by default

At rest and in flight on every database we operate.

Point-in-time recovery

Restore to any moment within retention window.

24/7 operations

On-call rotation with incident response and post-incident reviews.

Five managed offerings

One team. Four engines. Plus the platform that runs them.

The Managed Database Service is the platform — provisioning, HA, backups, monitoring, patching, performance tuning, on-call. Each engine inherits all of it. Other engines like Redis, Cassandra, MariaDB, or Elasticsearch are available on request.

DOCUMENT

MongoDB

Document store · NoSQL

Replica sets with automated failover. Sharding for horizontal scale. Atlas-compatible drivers and aggregation pipelines.

  • TopologyintSignal or hyperscaler
  • Versionsper cluster
  • BackupCIS Benchmark

RELATIONAL

PostgreSQL

Relational · ACID · SQL

Primary + standby with streaming replication. Extensions (PostGIS, pgvector, TimescaleDB) supported. Logical replication for change capture.

  • Topologyprimary + standby
  • Versions14 · 15 · 16 · 17
  • BackupWAL archive + base

RELATIONAL

MySQL

Relational · ACID · SQL

InnoDB cluster with group replication. Optional read replicas for analytics offload. Both 8.0 LTS and 8.4 LTS supported.

  • Topology3-node InnoDB cluster
  • Versions8.0 LTS · 8.4 LTS
  • Backupbinlog + snapshot

KEY-VALUE

Valkey

In-memory · Redis-compatible

Open-source Linux Foundation fork of Redis. Drop-in compatibility with existing Redis clients. Sentinel for HA, cluster mode for scale.

  • TopologySentinel or Cluster
  • Versions7.2 · 8.0
  • BackupRDB + AOF

Per-engine details

One team. Four engines. Plus the platform that runs them.

The Managed Database Service is the platform — provisioning, HA, backups, monitoring, patching, performance tuning, on-call. Each engine inherits all of it. Other engines like Redis, Cassandra, MariaDB, or Elasticsearch are available on request.

MongoDB · document store

VERSIONS 6.0 · 7.0 · 8.0

Deployed as a 3-node replica set across availability zones with automated primary election. Configurable read preferences let you split read load across secondaries. Aggregation pipelines, change streams, and time-series collections supported.

  • 3-node replica set with automatic failover
  • Continuous oplog backup for point-in-time recovery
  • Optional sharding for write-scale workloads
  • Index suggestions reviewed during quarterly tuning pass
  • TLS for client connections and intra-cluster traffic
javascript· connect.js

// Standard MongoDB driver. Use the SRV record we issue.
const {
  MongoClient
} = require('mongodb');

const uri = 'mongodb+srv://app:<pw>@db.intsignal.io' +
  '/orders?replicaSet=rs0&readPreference=primary' +
  '&tls=true';

const client = new MongoClient(uri);
await client.connect();

const db = client.db('orders');
const order = await db.collection('orders')
  .findOne({
    _id: orderId
  });

// → reads served by primary, writes acked w=majority

PostgreSQL · relational + extensions

VERSIONS 14 · 15 · 16 · 17

Primary + synchronous standby with streaming WAL replication. Read replicas added on demand. Major version upgrades drilled against your staging cluster first. Extensions enabled per-database: PostGIS for geospatial, pgvector for AI embeddings, TimescaleDB for time-series.

  • Primary + standby with synchronous replication
  • WAL archive + base backup for point-in-time recovery
  • Logical replication available for CDC pipelines
  • Connection pooling via PgBouncer in front
  • Quarterly EXPLAIN ANALYZE review for top queries
sql · explain.sql

-- Typical access. TLS required.
psql "postgres://app:<pw>@db.intsignal.io:5432/orders" \
     "sslmode=verify-full"

EXPLAIN (ANALYZE, BUFFERS)
SELECT o.id, c.name, SUM(li.qty * li.price) AS total
FROM orders o
JOIN customers c ON c.id = o.customer_id
JOIN line_items li ON li.order_id = o.id
WHERE o.created_at >= '2026-01-01'
GROUP BY o.id, c.name
ORDER BY total DESC LIMIT 50;

-- → Index Scan + HashAggregate, ~12ms

MySQL · InnoDB cluster

VERSIONS 8.0 LTS · 8.4 LTS

Three-node InnoDB cluster with group replication and automatic primary election. Read replicas added for analytics or geographic distribution. ProxySQL or MySQL Router fronts the cluster for transparent failover from the application side.

  • 3-node InnoDB cluster with group replication
  • Binlog + xtrabackup snapshots for point-in-time recovery
  • Optional read replicas for offload
  • Routing layer absorbs primary changes invisibly
  • Schema-change drills before applying in production
sql · cluster-status

-- Cluster health check via shell
$ mysqlsh --uri admin@db.intsignal.io
> var cluster = dba.getCluster()
> cluster.status()

{
  "clusterName": "prod-orders",
  "defaultReplicaSet": {
    "primary": "db-01:3306",
    "status": "OK",
    "topology": {
      "db-01": { "mode": "R/W" },
      "db-02": { "mode": "R/O" },
      "db-03": { "mode": "R/O" }
    }
  }
}

Valkey · in-memory key-value

VERSIONS 7.2 · 8.0

Open-source Linux Foundation fork of Redis. Sentinel topology for HA cache workloads; cluster mode for sharded data sets. All existing Redis client libraries continue to work without modification — connection strings stay the same.

  • Sentinel for HA caches, Cluster for sharded data
  • RDB snapshots + AOF persistence options
  • Full Redis client compatibility (no driver swap)
  • TLS-protected client and replication links
  • Keyspace notifications and Streams supported
shell · valkey-cli

# Connects with any Redis client. TLS enabled.
$ valkey-cli --tls -h cache.intsignal.io -p 6379

cache.intsignal.io:6379> SET session:abc "<json>" EX 3600
OK
cache.intsignal.io:6379> GET session:abc
"<json>"

cache.intsignal.io:6379> INFO replication
role:master
connected_slaves:2
master_link_status:up

# Drop-in with existing redis-py / ioredis / Lettuce

High availability with documented recovery.

Every managed database runs with a primary and at least one standby across availability zones. The application talks to a routing layer that follows the current primary — failovers happen without changing connection strings.

Continuous backup ships WAL or oplog data to encrypted, cross-region storage. Point-in-time recovery is tested as part of monthly drills and verified by restoring to an isolated environment.

  • Monthly or annual lease payments
  • Long-term contracts (10-25 years)
  • No upfront investment required
  • Site development and permitting handled
  • Ongoing maintenance included
  • Property value enhancement

Day-two operations

What we operate, every day.

Database operations isn't just keeping things running. It's the boring, scheduled work that keeps the next incident from happening.

PATCHING

Scheduled patch cycles

Minor versions applied during agreed change windows. Major versions drilled against your staging environment first, with rollback path documented.

  • cadence · monthly minor / quarterly major

BACKUP & RESTORE

Tested recovery, not hope

Backups verified by restoring to an isolated environment on a monthly cadence. Restore time documented and reported in the monthly evidence pack.

  • drill · monthly · isolated env

MONITORING

Per-engine SLI dashboards

Query latency, replication lag, connection saturation, slow-query log, and disk pressure tracked against thresholds we tune with you during onboarding.

  • alerts · routed to on-call rotation

QUERY TUNING

Quarterly performance review

Top-N slow queries reviewed every quarter with index suggestions and query rewrites. We deliver findings; your team chooses what to apply.

  • deliverable · quarterly report

SECURITY

Encrypted everything

TLS 1.3 enforced on client connections and replication links. Encryption at rest with AES-256. Audit logging on by default with retention configurable by compliance need.

  • access · least-privilege roles · periodic reviews

INCIDENT RESPONSE

Named on-call engineer

24/7 on-call rotation. Every incident gets a written post-incident summary with root cause, timeline, and prevention measures. You see what we saw.

  • coverage · 24/7 · response by severity

What we operate, every day.

Hardening and operating practices aligned to the frameworks your assessors recognize. intSignal is not the certified entity for most of these — we deliver the controls and evidence that make your audit possible. Where required, we partner with FedRAMP-authorized providers for federal scoping.

HARDENING

CIS Benchmark

Engine hardening with documented exceptions.

    SOC 2

    Aligned to Type II

    Controls and evidence cadence ready for audit.

      ISO

      Aligned to 27001 / 27017

      Cloud-services control narratives.

        HIPAA

        HIPAA-compliant ops

        Encryption, access, audit; BAA via partner.

          FEDERAL

          FedRAMP via partner

          Authorized hyperscaler regions integrated.

            DATACENTER

            Compliant facility

            Hosting facility carries its own attestations.

              FAQ

              Questions database teams ask before signing.

              If yours isn't here, ask in the consultation — we'd rather flag the awkward bits early than discover them in production.

              On intSignal infrastructure, in our hosting facility. Same physical platform that runs our Kubernetes and storage services. The data center carries its own facility-level attestations; we add the operational controls on top.

              The four engines highlighted are the ones we operate as standard offerings. We also run Redis (in addition to Valkey), Cassandra, MariaDB, and Elasticsearch / OpenSearch on request, under the same managed-service model. If you need something we don't list, ask — we'll tell you honestly whether it fits.

              Yes. Typical migrations use logical replication (PostgreSQL, MySQL) or change streams (MongoDB) for online cutover with minimal downtime. We plan the migration with you, test cutover in staging, and run the production cutover in a defined change window.

              Backups are encrypted at rest with AES-256, replicated to a separate region, and stored in immutable form for the retention period you specify. Monthly restore drills verify backups actually work — the drill outcome is part of the monthly evidence pack.

              Major-version upgrades (e.g. PostgreSQL 15 → 17) are drilled against your staging environment first. Production upgrades have a written change record, rollback plan, and signed-off change window. We do not push major versions without your sign-off.

              We deliver a quarterly performance review covering the top-N slow queries with index suggestions and rewrite proposals. Your application team decides what to apply — we don't push schema or query changes without sign-off. For ongoing performance work, we can engage as part of a deeper consulting arrangement.

              Valkey is the Linux Foundation open-source fork of Redis that emerged after Redis Labs changed its license. It's command-compatible with Redis through 7.2 — every existing client library works unchanged. We operate both; pick Valkey if you want fully OSS, pick Redis if you have a specific reason to stay on the original.

              Stop running databases. Start querying them.

              Tell us which engines you need, your HA and recovery requirements, and any compliance constraints. We'll propose a managed setup and walk you through the day-two operating model.

              Schedule consultation  ⟶