Scribd
Upload
169 views5 pages

Exadata Monitoring and Management Best Practices

Oracle Enterprise Manager 13c provides unified monitoring of Oracle Exadata hardware and software. It allows monitoring of individual Exadata storage cells and the entire storage grid. Key capabilities include monitoring performance metrics across databases, identifying top resource consumers, and configuring I/O resource management. Enterprise Manager 13c also enables automated patching of the entire Exadata stack and provisioning of Exadata in virtual environments.

Uploaded by

Osman
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
169 views5 pages

Exadata Monitoring and Management Best Practices

Oracle Enterprise Manager 13c provides unified monitoring of Oracle Exadata hardware and software. It allows monitoring of individual Exadata storage cells and the entire storage grid. Key capabilities include monitoring performance metrics across databases, identifying top resource consumers, and configuring I/O resource management. Enterprise Manager 13c also enables automated patching of the entire Exadata stack and provisioning of Exadata in virtual environments.

Uploaded by

Osman
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Exadata Monitoring and Management Best Practices

Mughees A. Minhas
Oracle
Redwood Shores, CA, USA

Keywords:

Oracle, Exadata, monitoring, management, database, performance, monitor, Enterprise Manager, EM

Exadata Monitoring and Management


Oracle Enterprise Manager 13c provides a unified view of Oracle Exadata hardware and
software with deep insight into the health and performance of all components such as the
database servers, InfiniBand switches, Oracle Exadata storage cells, Oracle Databases,
Automatic Storage Management (ASM), etc.

For hardware targets, you now have a photorealistic view that displays all component details,
including any open incidents. Infrastructure components are now available as new target types
which enable drill down to alerts and notifications for specific components like IB switch,
power distribution unit (PDU), Switches etc.

Critical incidents in the hardware layer can now be published to a 3rd party ticketing system
using the connector framework, something that Oracle hardware customers have asked for a
long time.

As Oracle releases new hardware and software, the Oracle Exadata plug-in is updated, tested,
and certified to accommodate the new products and 13.2 currently would support the X6-2
configuration of Exadata

Key Enterprise Manager New Capabilities in 13c


EM 13c allows you to monitor and manage exadata storage both at the individual cell level
and at the grid level. The storage cells would consist of flash and hard disks.

The Grid and Cell home pages along with the performance pages can be used to get answers
to questions like

I. How is I/O distributed between flash and hard disk?


II. How much benefit are we getting from Flash Cache?
III. How much disk I/O is due to Flash Cache?
IV. How does I/O correlate to database single block reads?
Based on your analysis, it also allows you to do administrative tasks like setting up IORM or
run an exacli or cellcli command. The support for exacli is new in EM 13c

There are two main reasons to use ExaCLI.

 Many companies, especially government organizations, require root access or the SSH
service on the cells to be disabled as part the compliance requirements.

 Service providers that host Exadata machines provide access to a virtual machine
running on the compute node. Customers are not allowed to ssh into the cell to
manage the cell using CellCLI.
For the use cases above, the only way to manage the cell is by using ExaCLI running on the
compute node.
Enterprise Manager 13c allows you to monitor the extreme flash only configuration that was
announced with the X5-2 configuration of exadata. One of the typical questions that
customers have is how much of flash throughput would an OLTP consume? Benchmarks at
oracle say that OLTP peak loads would consume 20% of flash throughput which means that
we can use the rest of it for scans.

EM allows us to monitor and track the throughput usage, utilization percentage from a
throughput perspective and also to monitor the latency and Flash IORM waits. This gives us
a very good visual interpretation of the workload distribution and basis which we can chose to
change your IORM strategy.

New Charts on Cell and Grid Home Page


Starting with ESS 12.1.2.1.0 I/O Resource Management (IORM) now manages flash drive
I/Os in addition to disk drive I/Os to control I/O contention between databases, pluggable
databases, and consumer groups. Because it is very rare for Oracle Exadata environments to
be limited by OLTP I/Os, IORM automatically prioritizes OLTP flash I/Os over smart scan
flash I/Os, ensuring fast OLTP response times with little cost to smart scan throughput.

EM now allows us to monitor and track the throughput usage, utilization percentage from a
throughput perspective and also to monitor the flash latency and IORM waits in addition to
the Disk latency and IORM waits. You can also drill down to the performance page to see a
holistic view of I/O distribution across all the databases and identify the top consumers.

Cell and Grid Performance Page


You can further drill down to the performance page to see a holistic view of I/O distribution
across all the databases. To identify the top consumers, you can switch to navigate to the
workload distribution page across all the databases. This page allows you to configure IORM
plans and to allocate a fixed amount of space in the flash cache for a database. The value
specified in flashcachesize is a hard limit, which means that the database cannot use more
than the specified value. This is different from the flashcachelimit value, which is a "soft"
maximum: databases can exceed this value if the flash cache is not full. Scrolling down the
page gives you visual charts on the IORM distribution, IO Requests and throughput across the
databases and thereby to identify the rouge consumers.

Infiniband Network Management


EM 13c provides enhanced IB switch management with enriched dashlets on the home page
covering important performance and configuration metrics. This allows you to detect and
notify and configuration changes or best practice violations. You also get to see a photo
realistic view the hardware component.

You could do a variety of administrative tasks like enable/disable port , clear error counters
and set up SNMP subscription which is new in 13c.

Monitoring: ILOM, Cisco Switch, PDU


You have similar capabilities for compute node ILOM, Switches, PDU etc. with enriched
dashlets on the home page covering a variety of performance and configuration metrics like
failures and speed control by looking at various views like Energy View, Network
connectivity ,Service processor (SP) configuration etc

Patching Automation Support of Exadata Stack


Oracle releases Quarterly Full Stack Patch Download (QFSDP) every quarter. This is part of a
new strategy for patching Exadata that reduces the number and frequency of patches released
to customers even more.
Primary components are:
 Database (Database, Clusterware).

 Infrastructure includes Exadata Storage Server, InfiniBand Switch, and PDU.

With EM 13c, there is full support for patching the entire Exadata stack, including:
 Compute nodes: Firmware and operating system (OS).

 Exadata Storage Server cells: Firmware and cell software.

 InfiniBand network: Switch firmware.

The patching application offers the facility to run the pre-flight checks and monitor the logs
from a single place (imagine having to manually monitor the patch execution logs for grid
infrastructure, operating system, storage for all the compute and storage cells in a rack)

Exadata Virtual Provisioning


Provisioning involves repeatable, reliable, automated, unattended, and scheduled mass
deployment of a RAC Cluster including virtual machines (VMs), Oracle Database (DB), Grid
Infrastructure, and ASM on Virtualized Exadata.
With the Exadata plug-in's virtualization provisioning functionality, you can:
 Define the size and shape of the Virtual Machines including compute, memory and
networking components.
 Create and delete a RAC Cluster including VMs, Oracle Database (DB), Grid
Infrastructure, and ASM.
 Scale up and scale down a RAC Cluster by adding or removing VMs including DB,
Grid Infrastructure, and ASM.

Provisioning and Deployment Challenges


Let us now focus on deployment challenges from a consolidation perspective. While
deploying databases on to exadata and also plan to consolidate into Oracle Public loud, you
will need to have answers to many questions like

I. Which apps to consolidate?

II. What are the risks?

III. How much cloud compute will I need?

IV. Will I have performance improvement?

V. How to validate performance improvements?

The new Consolidation Workbench provides an end-to-end workflow that provides three
distinct steps:

 What-if analysis on various consolidation scenarios: commodity to engineered


systems, non-multitenant to multitenant databases and on-prem to Oracle Cloud.

 The actual enactment of the consolidation by integrating with the provisioning


features

 Post-consolidation testing using the SQL Performance Analyzer

Benefits of the DB Consolidatiomn Workbench


The DB Consolidation Workbench helps you migrate to Oracle Exadata Cloud Service along
with multi-tenant for greater consolidation. It accurately estimate required Oracle cloud
compute size and shape for Exadata Cloud Service Instances. The workbench relies on AWR
data to give consolidation advice by identifying candidate databases for the designated
consolidation platform.
Health Check and Diagnostics Best Practices
Exachk – OraChk is a non-intrusive tool to conduct a comprehensive Exadata health check on
your Exadata Database Machine to validate your hardware, firmware, and configuration
which also checks for impactful reoccurring problems across Oracle customer base.

You can audit various important configuration settings within an Oracle Database Machine
Exadata System - Database Servers, Storage Servers and Infiniband Switches. It produces
two reports, summary and detailed. The detailed report will contain Benefit/Impact, Risk and
Action/Repair information.

With EM 13c, integration of Exacheck into the Compliance framework of Enterprise Manager
is complete; this would let you generate automated notifications and reports for any violation
the Exadata configurations. You can install setup and upgrade the exachk utility end to end
through EM.

Diagnose and repair Hardware Faults using ASR


Oracle Auto Service Request which is part of Oracle Premier Support can be used to get
notification of hardware failures and get rectified. ASR is now integrated with EM 13c

EM infrastructure is used to configure ASR, there is no need to identify a separate


infrastructure and separate set up steps for this as before. You can view ASR Incidents in
EM Incident Manager with minimal configurations

Contact address:

Mughees Minhas
Oracle
400 Oracle Parkway
Redwood City, CA 94065
USA

Phone: +1 650 506 8661


Email [email protected]
Internet: http://www.oracle.com

You might also like