The systems the business runs on, operated 24/7.
What it is
Managed platform and database services are the operational half of Cloudtech. They cover the systems a business runs on: the databases and related applications, the clusters that keep them available, and the platforms underneath. Cloudtech takes contractual responsibility for running those systems to an agreed service level, around the clock.
This is the oldest Cloudtech practice. Its engineers have run scalable and failover clusters for Oracle databases and the applications around them since the Sun Microsystems era of the mid-1990s.
Managed services combine the other five Cloudtech service lines in one form or another. The data platforms, applications, pipelines and AI that the other lines design and build are run here, under one service level, and the engineers who built them take part in running them.
What we do
- 24/7 operations. Monitoring, on-call engineers and a P1 reaction time of one hour or less for every client, every hour of the year.
- Databases. Oracle, PostgreSQL, MySQL and MariaDB, Microsoft SQL Server, and time-series databases. Installation, configuration, patching, tuning, capacity, user and access management.
- Clusters and platforms. Failover clusters on Solaris Cluster, AIX and Linux, Kubernetes, virtualisation.
- High availability, disaster recovery and backup. Solaris Cluster, Oracle RAC, Data Guard and SharePlex replication, CDC replication through Kafka, PostgreSQL replication and Veritas NetBackup. Recovery procedures are rehearsed on a regular basis, not assumed.
- Upgrades and migrations. Major-version upgrades, platform migrations and hardware generation changes, planned inside agreed windows and reversible.
- Capacity, hardware and licence optimisation. Sizing and consolidation that matches the estate to the load and to licence terms, measured before and after.
- On-prem private clouds. Environments designed, built and run end to end, with the automation and service levels of a cloud, on capacity the customer owns.
What you get
The technical scope is set per contract, from a single Oracle database to an entire estate. The service management around it is the same for every client and follows ITIL 4 practice:
- Single point of contact, 24/7. Incidents, service requests and change requests are logged by phone or ticket and tracked to closure. A P1 is reported by phone, at any hour, and reaches an engineer directly.
- Monitoring and event management. The systems in scope are monitored against a declared inventory of what should be running. Alerts are validated by an engineer and raised as incidents, so the service reacts to Cloudtech’s own monitoring as well as to your reports.
- Incident management. A qualified engineer is working on a P1 within the hour, every hour of the year. Not an automated acknowledgement. Work continues until service is restored.
- Versioned configuration management. Configuration of the systems in scope is kept in version control and applied by automation wherever the platform allows, GitOps-style. The record of a change is the change itself, and the running state can be compared with the declared one.
- Patch and lifecycle management. Security patches and updates assessed against your inventory and applied in agreed windows. Versions approaching end of vendor support are reported in advance, and major-version upgrades are planned before support runs out.
- Backup, recovery and service continuity. Backups monitored and restores rehearsed to a schedule. Recovery time and recovery point objectives agreed per system; failover and disaster recovery tested, not assumed.
- Capacity and performance management. Load, storage and response time trended and forecast, with recommendations on capacity, hardware and licence terms in time to act on them.
- Access and security management. Named accounts, no shared logins, access over agreed secured channels, read-only wherever the task allows, and every administrative action recorded for audit.
- Vendor coordination. Support cases with Oracle, the operating system and hardware vendors are opened and driven by Cloudtech on your behalf.
- Service level reporting and reviews. A monthly service report: service-level attainment, incidents and post-incident reviews, changes, availability, capacity trends, backup and restore results, patch status and security events. Discussed at a regular service review.
- Documentation and runbooks. Written or validated during onboarding, kept current with every change, and handed over in full at exit.
AI adoption
We also adopt AI in an indirect manner: the AI has no direct access to any system, only read-only access to the log collection hosts where log files, trace files, system statuses, monitoring data and terminal logs are gathered. We employ it for faster and more precise troubleshooting, fault analysis and resolution proposals, security hardening, capacity and performance trend analysis, and the triage of vendor patches and security advisories. AI support is also integrated into the incident and support request management process.
On premises, by conviction
Critical systems belong on premises, and our customers share the view. An on-prem private cloud keeps production data under its owner’s physical and legal control. Capacity is bought once and used for years, and the systems that use the data run next to it. Cloudtech makes that model as automated and as well run as any public cloud. Where a cloud service is the right target for part of the data, it is connected deliberately.