Managed Hosting
We administer SAP under RISE, SAP on AWS, SAP in your own data centre, and AWS on its own. Monitoring, backups, patching, on-call. With an agreed SLA.
The project ends and the environment stays. Someone has to be watching it at three in the morning on a Sunday, know when to apply a patch, and hold a backup that genuinely restores. Keeping that combination of skills on payroll rarely pays off — you need it a few times a year, but it has to be available every day. And we do not turn work away because the system happens to sit somewhere unusual.
Takeover
A review of what exists, filling in what is missing, writing down the procedures. We do not take over an environment we do not understand.
Monitoring
Signals from the application layer, not only the infrastructure — with SAP, straight from the system. What matters is whether the process runs, not whether the host answers a ping.
Backup and recovery
Copies written outside the production account. Recovery is tested, not assumed.
Patching
Security fixes, SAP notes, operating system releases and AWS service versions — scheduled into windows agreed against your financial close calendar.
On-call and response
A ticket reaches someone who knows this system. Response times are written into the contract.
Authorizations and audit
Access review — in SAP also duty conflicts, in AWS roles and privileged access. Reports in a form auditors accept without translation.
- SoD GuardSAP authorization conflict analysis and role matrix management. It finds where one person can create a vendor and approve their payment — before an auditor does.In production
- SkynetAIAn operations platform for the SAP systems we run. It collects signals straight from SAP over RFC, surfaces state and risk, and gives the customer the same view we have.In production
- VaultSAP backups written to S3 in a region independent of the one the system runs in. Neither a regional failure nor access to the production account reaches where the copy lives.In development
- MonitorExtends SAP Cloud ALM with signals it does not carry out of the box. Monitoring stays in the tool the team already knows — no extra console to learn.In the works