1z0-902 Oracle Exadata Database Machine X9M Implementation Essentials Questions and Answers

High Capacity (HC) Storage Server: The Exadata X9M-2 HC storage server is configured with a combination of hard disks and flash memory, providing a balance of storage capacity and performance.

It includes 12 x 18 TB hard disk drives, which totals 216 TB of raw disk capacity​.

For flash storage, it has 4 x 6.4 TB PCIe flash cards totaling 25.6 TB of flash storage. This configuration uses NVMe Flash Accelerator F680 cards optimized for database caching​​.

The persistent memory in the HC server totals 1.5 TB, enhancing performance for OLTP (Online Transaction Processing) workloads by providing ultra-fast access speeds compared to traditional flash storage​.

Extreme Flash (EF) Storage Server: The EF storage server in the Exadata X9M-2 model is an all-flash storage server designed for high IOPS and low latency, making it ideal for performance-intensive workloads.

This configuration contains 8 x 6.4 TB PCIe flash cards, providing a raw capacity of 51.2 TB of high-performance flash storage​.

In terms of persistent memory, it includes 756 GB of Intel Optane Persistent Memory. This enhances flash storage performance by offloading data processing directly to persistent memory​​.

The Exadata X9M-2 storage configurations—both HC and EF—are optimized for different workload requirements. The HC configuration caters to larger datasets needing high storage capacity, while the EF configuration is designed for workloads requiring high throughput and minimal latency.

According to Oracle’s documentation12, the components that are monitored directly through the use of Exadata-specific Enterprise Manager Plug-Ins are:

The storage server ILOM2

The Power Distribution Units (PDUs)2

The RDMA over Converged Ethernet (RoCE) switches3

According to Oracle’s documentation1, the Cell Control Command-Line Interface (CellCLI) utility is the command-line administration tool for Oracle Exadata System Software. It can be used to perform various tasks on Exadata Storage Servers, such as creating and managing objects, monitoring performance, and configuring alerts.

To run CELLCLI commands on multiple cells simultaneously, you can use the Distributed Command-Line Interface (dcli) utility2. The dcli utility allows you to execute commands or scripts on a group of cells or hosts in parallel. You can use the dcli utility to copy files from one cell or host to multiple cells or hosts using the -f option3. You can also use the dcli utility to execute commands or scripts on multiple cells or hosts using the -l option to specify a user name and the -g option to specify a file containing a list of cells or hosts3.

Therefore, to achieve your requirement of running a script with several CELLCLI commands on each cell in your Exadata full rack simultaneously, you must:

Copy and execute the script on all storage servers using the dcli command.

Once the physical disk has been replaced, you will need to create a new celldisk and griddisks on the replaced physical disk. After the griddisks have been dropped from the associated ASM diskgroups, you can then add the griddisks back into the ASM diskgroup they used to be a member of and rebalance the data on the effected griddisks. Finally, you should verify that the griddisks located on the physical disk have been successfully dropped from the associated ASM diskgroups.

In order to monitor all components of this multirack configuration, Enterprise Manager agents must be deployed on at least two storage servers in both racks and all database servers in both racks. This is according to the Oracle Exadata Database Machine X9M Implementation Essentials official text book [1], in which it states "To monitor all components of the multirack configuration, you must deploy agents on all database servers and at least two storage servers in each rack" (page 6-15).

According to the Oracle documentation1, two monitoring tools that might be useful in examining I/O performance for the prod database are:

cellcli (or exacli/exadcli) to examine storage server metrics such as database, category, ceiidisk, and griddisk (D). This tool can help you monitor the I/O Resource Management (IORM) metrics and identify any bottlenecks or imbalances in the storage layer.

I/O-specific dynamic performance views such as v$iostat_fiie, v$iostat_function, and v$iostat_consumer_group from the prod database instances using SQL *p1us ©. These views can help you monitor the I/O activity and latency at the file, function, and consumer group level.

The Oracle Exadata Deployment Assistant Command Line Interface (OEDACU) supports a variety of configuration and lifecycle management tasks for Exadata systems, such as adding or removing nodes, resizing ASM disk groups, and managing Oracle Database home directories. However, it does not support lifecycle management for the Zero Data Loss Recovery Appliance (ZDLRA), as this device requires different management tools specific to its backup and recovery functions .

According to the Oracle Exadata Database Machine Technical Architecture1, the initial storage configuration after the standard (non-virtualized) deployment of a new Exadata Database Machine with High Capacity storage servers includes two ASM disk groups: DATA_ and RECO_. These disk groups are built on hard disks and flash disks inside the Exadata Storage Servers1.

The correct statements about this configuration are:

Option D: There is free space available on flashdisks inside the Exadata storage servers for possible use for storage indexes. Storage indexes are a feature of Exadata Storage Software that can improve query performance by avoiding unnecessary I/O operations. Storage indexes use a small amount of flash memory to store metadata about data blocks stored on disk1.

Option E: There is free space available on flashdisks inside the Exadata storage servers to configure Exadata Smart Flash Logs. Exadata Smart Flash Logs are another feature of Exadata Storage Software that can improve database performance by using flash memory as an extension of the database redo log buffer. This can reduce latency and increase throughput for redo log writes1.

According to Oracle’s documentation12, ASM failure groups provide redundancy by storing mirror copies of data on different disks or storage servers. When ASM allocates an extent for a normal redundancy file, ASM allocates a primary copy and a secondary copy. ASM chooses the disk on which to store the secondary copy so that it is in a different failure group than the primary copy1. This way, if one disk or storage server fails, ASM can still access the data from another failure group2.

Therefore, the statement that is true about how ASM failure groups provide redundancy is:

Failure groups contain all ASM disks in a single storage server preventing mirror copies being written to the same storage server.

According to Oracle’s documentation1, one of the utilities for diagnostics and repair of Oracle Exadata Storage Server is ExaWatcher, which collects operating system and network metrics on database servers and storage servers. The utility is located in the /opt/oracle.ExaWatcher directory1.

To investigate why an Exadata Database Server stopped communicating on the client network for 10 minutes over the past weekend, you can use the GetExaWatcherResults.sh script to collect and analyze ExaWatcher data for a specified time range1.

Therefore, the command that you should use to investigate this is:

/opt/oracle.ExaWatcher/GetExaWatcherResults.sh —from _17:00:00 —to _23:59:00

In order for an Exadata X9M-2 Elastic Rack to be installed, two power feeds are required for each of the three-phase 24kVA PDUs. A single cable from each PDU will not be enough to power the rack, and neither a splitter cable nor rearrangement of the power cables from the servers to the PDUs will be able to provide sufficient power for the system. Therefore, the installation cannot proceed until two power feeds are available per PDU.

For more information about power requirements for Exadata X9M-2 Elastic Racks, refer to the Oracle Exadata Database Machine X9M Implementation Essentials official text book and resources.

Search results: [1] Oracle Exadata Database Machine X9M-2 Full Rack Installation - Oracle Docs [2] Exadata Database Machine X8M-2 Mid-Size Rack Installation - Oracle Docs

https://docs.oracle.com/en/engineered-systems/exadata-database-machine/dbmin/exadata-site-requirements.html

For a non-virtual deployment of Oracle Exadata with a single cluster:

Single Client Access Name (SCAN) Listeners: By default, three SCAN listeners are configured on the RDMA over Converged Ethernet (RoCE) network. This setup allows for load balancing and high availability of incoming client connections across the cluster nodes​.

Oracle Cluster Registry (OCR) Multiplexing: The OCR is typically stored in multiple locations within the disk groups to enhance reliability and fault tolerance. On Exadata, the OCR is multiplexed across the DATA and RECO disk groups for enhanced resilience​.

ora.diskmon Resource: This is an essential component of Oracle Clusterware, responsible for monitoring and managing disk-related cluster resources. It is always enabled and kept online by Oracle Clusterware on Exadata systems​.

In an Exadata environment, the path of communication for SNMP traps involves:

Management Server (MS) Process to EM Agent: Alerts related to the storage server are generated by the MS process, which then forwards SNMP traps directly to the Enterprise Manager agent, allowing centralized monitoring and alert management​.

ILOM to EM Agent: The ILOM on each Exadata storage server can independently send SNMP traps to the EM agent, providing additional hardware-level monitoring and alerting capabilities .

These pathways ensure that alerts from both hardware and software layers reach Enterprise Manager efficiently, supporting proactive management and rapid response to system issues.

According to Oracle’s documentation1, the customer options for hosting the Platinum Services Advanced Support Gateway are:

Provide individual x86 64-Bit gateway hardware1

Purchase the recommended x86 64-Bit gateway hardware from Oracle1

Install on Oracle Database Appliance2

The dbmcli -e "LIST IBPORT DETAIL" command is not valid on Exadata X9M. According to the Oracle Exadata Database Machine X9M Implementation Essentials Official Text Book [1][2], this command is only valid on Exadata X3 and X4 models. The other three commands are valid on Exadata X9M.

https://forums.oracle.com/ords/apexds/map/tech/apps-infra/discussion/4342715/monitoring-roce-performance

Oracle Linux 7.9 is the minimum version required to support RoCE and Persistent Memory. This is because Oracle Linux 7.9 includes kernel enhancements that enable RDMA over Converged Ethernet (RoCE) networking and Intel Optane DC Persistent Memory Modules (DCPMM) for Exadata X8M systems3. These features allow faster access latency and higher throughput for database workloads2.

According to the Oracle Exadata Database Machine documentation12, two recommended configuration best practices for backup and recovery on Exadata are:

A. Placing the backup network on dedicated switches installed in the top of the rack (ToR) has the benefits of isolating the backup network from other workloads and providing a greater level of control.

D. When off-site long-term backup retention is needed, use Oracle Database Backup Cloud Service as a low-cost, offsite scalable storage for a disaster recovery solution.

According to Oracle’s documentation12, some of the statements that are true about applying updates on Exadata systems are:

Failed storage server updates are automatically rolled back to a previous release1.

When running a “yum update” for a new Exadata release, all other repositories should be disabled1.

According to Oracle.com documents or resources, the three options that are available for database consolidation on Exadata are:

A. Multiple pluggable databases in one or more container databases on a Virtual Machine or Bare Metal cluster123.

C. A single database with one or more schema per application12.

E. Multiple Databases on a cluster12.

https://blogs.oracle.com/exadata/post/database-consolidation-why-and-how

D. ASR can upload configuration metadata to support problem resolution1.

E. ASR Manager opens a service request (SR) automatically after sensors detect hardware faults1

https://docs.oracle.com/cd/E37710_01/doc.41/e23333/toc.htm

According to Oracle documentation1, Exadata Storage software has a complete set of automated operations for hard disk maintenance, when a hard disk has failed or has been flagged as a problematic disk. It will initiate a rebalance operation to redistribute data across other disks and mark the failed disk with an amber light.

According to the Oracle Auto Service Request (ASR) documentation1, ASR raises service requests for qualified Oracle products that are detected with specific faults. The qualified Exadata products include2:

Database servers

Storage servers

InfiniBand switches

Cisco switches (X8M and later systems)

Power distribution units (PDUs)