Monitoring SAP HANA

SAP HANA is an in-memory, relational database management system. It operates primarily as a columnoriented store, but row-oriented tables are also supported. Both transactional and analytical workloads are supported. SAP HANA is fully ACID compliant. Various features include a JSON store, spatial data processing, text analytics and search, time-series analytics, streaming data processing, and graph data processing.

An SAP HANA deployment requires a certified hardware appliance installed with one of several supported operating systems. Hardware requirements include specification of the CPU architecture, core count, minimum RAM, and minimum storage. The supported operating systems are Red Hat Enterprise Linux for SAP Solutions, Red Hat Enterprise Linux for SAP HANA, SUSE Linux Enterprise Server for SAP Applications, and SUSE Linux Enterprise Server. SAP HANA offers an optional, embedded web application server XS Advanced (XSA) based on Cloud Foundry. XSA offers native support for Node.js and JavaEE as well as extensibility through custom runtimes.

Transactional updates to the database are performed using multi-version concurrency control to enable read consistency and throughput. Writes are performed using row-level write locks with blocked transaction monitoring.

An SAP HANA system contains a single system database and zero or more isolated tenant databases. Multiple servers working together comprise the database system: name server, index server, preprocessor server, compile server, script server, XS Advanced runtime, and others. A single system can be distributed across multiple hosts to improve scalability and can utilize replication for high availability.

Key Features

The Foglight for SAP HANA Agent monitors the current and historical status of SAP HANA database systems. Database configuration and performance statistics are collected and presented in real-time dashboards and also available in packaged reports. Alerts are generated on respective components of the system when indicators of current or potential future performance or availability issues arise. Setup involves creating a monitoring agent for each target database system. Such agents are fully remote, meaning that they are typically configured on a separate host. Installing the agent on the Foglight Agent Manager ensures that nothing needs to be installed on the monitored system. A database user with appropriate permissions is used by the agent to collect data. An agent may also be configured to collect operating system metrics purpose of monitoring the database host and logs.

Data is collected from each configured SAP HANA system via the official JDBC driver. Collections take place at regular intervals, called periods, which are configurable by collection type in the Foglight Agent Properties. Critical metrics, such as system availability, are collected frequently, while data expected to change less frequently, such as configuration settings or individual table metrics, are collected at longer intervals. The Foglight Agent queries Hana system tables as well as the embedded statistics service for the latest values of key parameters and metrics. These metrics are collected and compared against thresholds and historical performance to identify system possible performance and availability issues.

Database hosts are monitored at the operating system level by the Infrastructure Agent which can also be used to monitor SAP HANA native logs.

Additional details will be provided shortly.