Exadata Host Interconnects Test

Infiniband Network is a high-performance, very low-latency network layer that is active-active in all directions at 40 Gb / sec, which enables communication between the Database Server and Exadata Storage Server. In a production environment where for e.g., an Exadata Database Machine Half Rack is deployed, the Database Machine uses a state of the art InfiniBand interconnect (through Infiniband swtiches) between the servers and storage. Each database server host and Exadata cell has dual port Quad Data Rate (QDR) InfiniBand connectivity for high availability.

The non-availability of the interconnect on any database server host can impair that host's communication with the Exadata storage server and that of other hosts. As a result, user requests to that host may not be served instantly delaying the I/O operations. In the aftermath of this, mission-critical business services using the resources of that host may experience prolonged outages or slowdowns, resulting in considerable loss of revenue and reputation. To avoid this, administrators need to continuously monitor the I/O transmissions for each database server host connecting to the Oracle Exadata Storage Server. For this purpose, you can use the Exadata Host Interconnects test.

This test periodically auto-discovers the database server hosts connected to the target Oracle Exadata Storage Server through the Infiniband interconnects and for each database host, reports the I/O transmission capability. Using this test, administrators can figure out the database server host to which data was dropped frequently and figure out the reason behind such data being dropped during transmission.

Target of the test : Oracle Exadata Storage Server

Agent deploying the test : A remote agent

Outputs of the test : One set of results for each host interconnect connecting the target Oracle Exadata Storage Server being monitored and the Oracle database server.

Configurable parameters for the test

Parameter	Description
Test period	How often should the test be executed
Host	The IP address of the host for which this test is to be configured.
Port	The port number at which the specified host listens. By default, this is NULL.
Username, Password and Confirm Password	By default, this test uses the Cell Control Command-Line Interface (CellCLI) to pull out the required metrics. To use the CLI, the test first needs to connect to the target storage server via SSH, and then run commands using CLI. For running the commands, this test requires the credentials of a cellmonitor user. Specify the login credentials of such a user in the Username and Password text boxes and confirm the Password by retyping it in the Confirm Password text box.
SSH Port	This test uses the Cell CLI to pull metrics from the target Oracle Exadata Storage Server. To run the CLI commands, this test first needs to establish an SSH connection with the target storage server. To enable the test to establish this connection, specify the SSH Port here.
Timeout	Specify the time duration for which this test should wait for a response from the storage system in the Timeout text box. By default, this is 120 seconds.

Measurements made by the test

Measurement	Description	Measurement Unit	Interpretation
Dropped data during transmission	Indicates the rate at which data was dropped during transmission through this host interconnect.	MB/sec	Compare the value of this measure across host interconnects to figure out the host interconnect that frequently dropped during transmission.
Dropped data during RDMA transmission	Indicates the rate at which data was dropped during RDMA transmission through this host interconnect.	MB/sec	Compare the value of this measure across host interconnects to figure out the host interconnect that frequently dropped during RDMA transmission.
Received rate	Indicates the rate at which data was received from this host interconnect.	MB/sec
Retransmitted rate	Indicates the rate at which data was retransmitted through this host interconnect.	MB/sec
Transmitted rate	Indicates the rate at which data was transmitted through this host interconnect.	MB/sec