Oracle Performance Forecasting / Capacity Planning

Hi,
I am being told to Analyse and Propose the Oracle Performance Forecasting / Capacity Planning, since our database are being hosted on co-hosting services. By planning we can help the cost of CPU, Memory and Disk projection.
Is there any way, we can get the current workload from the current production environment, and calculate the forecast based on Statistical values.
I know, we can get the current workload from AWR, but I dont know the calculations and proper judging method.
Thanks in advance.

If your application is typical, it will consumer resources in rough proportion to the load in transactions/queries per second. If your business is stagnant, the resources consumed will stay about the same.
As load grows, though, you'll hit a point at which the response time starts to grow without bound. This is typical of growing businesses, or ones which have done acquisitions in order to improve their customer base.
Capacity planning is knowing where that point is, and planning to stay far away from it (:-))
If you can do a load test on a single instance of a cloud (or time-sharing) machine and replay previously recorded transactions, you can measure how many replayed users it will take to slow down the machine, and how many additional user it takes to bring the transaction time from a typical 2-3 seconds to 30.
If you plot that line, you'll find you have a hyperbola or "hockey-stick" curve, and that will allow you and your financial mangers to plan on exactly how many users to cap a given machine at, to keep the performance below 3 seconds or so. After that many users, add another machine.
The usual rule of thumb is to keep the response time from doubling from it's average at very low load, as after it's once doubled, it keeps doubling and redoubling with tiny increases in load.
I personally like to model first and stress-test only to confirm I've not messed up. The tool of choice is Teamquest Model, and have used it to accurately measure and predict the behaviour of an Oracle database (and lots of other stuff). If you're interested in the theory behind this, read anything by Neil Gunther...
--dave c-b

Similar Messages

Capacity Planer Group --- Maintenance work center

Hi,
I have a query
1. like do we have to create capacity planner groups same as the planner groups in equipment master data....Do they have any link??
2. What would be the start time and end time in maintenance work center, for 24 hrs open manufacturing industry with 3 shifts???
thanks
DM

Hello Deepak,
There is no relation between capacity planner grp & Maintenance planner group...These are 2 different things..
Now the work centers can be defined for the technicians if HR module is not implemented...
To perform the capacity planning for the technicians, there daily working hrs must be maintained in the system..Though the maintenance is working for 24 hrs,Single technician cannot work for entire 24 hrs..
Hope this will help..
Thanks

Capacity planning of database objects

Oracle 10.2.0.4
Windows platform
I was reading Oracle Admin Guide :- Capacity planning of database objects section, but these and most of other sections are very theoretical .
Kindly suggest how to get these topics in better way.

I think Reddy has it right: watch the trend in space usage primarily at the tablespace level but you would also want to take a look at the individual tables and indexes within a tablespace to spot the fast growers, to pick out tables that might need parameter adjustment (pct_free) to avoid row migration, to identify indexes that may not reach a reasonable stready state size, etc....
The dba_outstanding_alerts and related views might be of interest.
On occassion just taking a count(*) of how many application tables, indexes, and users exist on your system and storing this with space usage figures can also be useful information when it comes time to request more disk, cpu, or memory since you can show past growth.
HTH -- Mark D Powell --

Oracle VM documents (perf tuning, monitoring, metrics, capacity planning)

Hi,
Does anyone has any document on Oracle VM performance tuning, capacity planing and monitoring.
I have Oracle VM environment with two virtual domains, each has one Oracle database running. I have been asked to monitor performance, identify performance metrics at OS/VM level (for memory, cpu, IO). If required, estimate additional hardware (Capacity Planning).
Thanks,
Neeraj

Without sar utility (sysstat RPM, i'm pretty sure it is not included in OracleVM) you are pretty out of luck. As OracleVM is based on OEL5 you could try to install sysstat RPM from OEL5.

Capacity planning (performance tuning of WLP 8.1 SP6 application)

Hi Guys,
I have one question regarding the performance tuning of the weblogic portal application (or may be any J2EE application for that matter) :
I understand that there could be lot of performance bottleneck areas e.g
(a) network issues (not enough bandwidth),
(b) bottleneck because of the DBs,
(c) application design/coding
(d) JVM might not have been tuned as per the business requirement of the application
(e) lack of clustering
(f) hardware problems (CPUs, RAMs etc)
Inorder to do the capacity planning for such an enterprise applications, what would be the best starting point?
Is it a good idea to start and profile the application using some profiling tools like OptimizeIt or JProbe? and then analyze the profiling the data and conclusion can be drawn interms of what and where the major bottleneck is into the application and accordingly it can be worked upon.
Or if there is any better and elegant way doing capacity planning?
Thanks for your time.
Regards
Jameel

Hi
id say start with a load tool that simulates average/ peak behavior with varied inputs and exercise your site . then profile if necessary.
Also start with the critical/most used portions of your site.
Its difficult doing the above because unles your migrating noone really knows the expected usage of your site. And its hard for someone to specify acceptable response times.
regards
deepak

Capacity Plan with Oracle 10g

Good Day to all,
I've little time working in environments with Oracle databases, I have requested to carry out a capacity plan with Oracle Database 10g for a data warehouse project that is leading the company in which they work. I request to make a plan specifying, among other things: size and number of tablespace and datafiles, projection growth taking into account the initial charge and the charge per week (incremental). The truth is a bit complicated for my inexperience in this kind of sizing requirements we will ask for your valuable cooperation. There are mathematical formulas that allow me to take those projections into account the type of data and their lengths? , There is a standard for creating the tablespace and datafiles?.
In advance thank you for your contributions.

The first thing you need to get management to do is give you two things.
1. The cost to the organization for downtime, rated in dollars/hour.
2. The service level agreement for the system's customers.
3. The amount of data to be loaded into the system and the retention time.
4. What version of RAID or what ASM redundancy is planned.
With that you can start at the grossest level which is planning for database + archived redo logs + online back files.
I generally figure the database, itself, at about 25% of required storage because I like to have at least two full backups, a bunch of incremental backups, plus all of the archived redo logs that support them. And all on separate shelves.
The number of tablespaces and data files is really just a question of maintenance. Ease of transport. Ease of movement. Ease of backing up.
If you want to get down to the actual sizes of tables and indexes the best place to go is the DBMS_SPACE built-in package. Play with these two procedures:
CREATE_TABLE_COST and CREATE_INDEX_COST which do the job far more efficiently and accurately than any formulas you will receive. You can find demos
of this capability here: http://www.morganslibrary.org/reference/dbms_space.html.

Oracle identity Fedration 10g capacity Planning and sizing

Hi,
Looking for a capacity Planning and sizing document for Oracle identity Fedration 10g.
Thanks in advance.
Thanks,
Vikalp Sareen
<telephone information removed by moderator>
Requests for off-forum communication defeats the entire premise of a public discussion forum. Additionally, do you actually want your telephone number to be harvested by people without scruples or principles and possibly used to steal your identity?
Edited by: rukbat on Jun 27, 2011 7:40 AM

Hi Ajay,
I presume you are talking about this section http://download.oracle.com/docs/cd/E10773_01/doc/oim.1014/b25355/deployinstall.htm#BABHIJGJ . This section does specify the sizing for some deployment scenarios. It is not brief. We cannot propose the hardware configuration for any customer just referencing this.
Please let me know your thoughts.
-Mahendra.

Oracle Database capacity planning

Hello Team:
Does any one have a Capacity planning spread sheet for sizing the server Requirements? ( CPU,Memory etc )
Regards,
Bala

>
Thanks Justin. How do I size the Database server Hardware? Third party apps most cases come with a
recommendation for hardware. I am talking about a situation where we have to build a database based
on certain basic inputs such as Concurrent users, Load , response time etc etc.Craig Shallahamer's orapub site would be a good place to have a look.
HTH,
Paul...

Capacity planning for a 2 million hits / day site

Hi,
I am doing a feasibility study for a 2 million hits per day
e-commerce site.
We are looking at various vendors but would prefer an
Oracle solution.
One option I am looking
at is a three-tier server architecture with instances of
Oracle Application server running on one set
of machines and instances of Oracle 8/8i running on a second
set of machines.
Macromedia generated HTML would be auto-translated to PL/SQL
cartridges to generate the relevant WRBXs.
Has anyone any experience of the performance issues /
robustness of this type of large-scale
development.
In particular, I am looking for some hard numbers / capacity
planning model on:
Number of instances of Application Server;
Number / type of Application Server boxes - SpecInt / Flt and
memory;
Number of instances of RDBMS / parallel Oracle;
Number / type of RDBMS boxes - SpecInt / Flt and memory; and
External communications bandwidth.
Any advice, personal anecdotes, recommended sites, or
literature references much appreciated.
Regards,
Ajit Jaokar
null

Hi,
1. Do cross forest authentication Go through Global Catalog?
Yes, cross-forest authentication relies on Global Catalogs of both forests.
2. Do Global catalogue cache ad object info if they are in different Forest. If not, how does the authentication request flow across the forest?
GC doesn’t cache information for objects from another forest. However, once a user has been authenticated and authorized, its service ticket remains (on the local machine) for a while before it logs off. The authentication request
is first received by the local DC, then GC, when GC couldn’t find a match in its own forest, it checks its database for trust information, if there is a name suffix matched, the authentication request is passed to the corresponding forest.
3. Is this calculation still the same when considering cross forest trusts?
Yes, establishing forest trust doesn’t consume much more space, because GC doesn’t store information of another forest.
4. Do we need to consider any other memory requirements in a cross forest trust environment?
Not really, as I mentioned above.
More information for you:
Accessing resources across forests
http://technet.microsoft.com/en-us/library/cc772808(v=WS.10).aspx
Best Regards,
Amy

Seeking a capacity planning consultant

Xpede is an financial services application service provider based in
Oakland, CA, seeking an capacity planning consultant for an short term
(under one week) initial engagement, with possible on-going consulting work.
Please send resume, qualifications, hourly rate or a fixed price bid to
[email protected]
Capacity Requirements Plan (2-5 days)
The Plan concentrates on the responsiveness, scalability, and throughput of
our middleware infrastructure through investigation, analysis, and a formal
report. This service is conducted as a pre-deployment evaluation. The
consultant will work with our staff to identify or establish our capacity
requirements in quantitative terms, which express the requirements for an
application and architecture, and will encompass both current and future
timeframes. Performance metrics may take the form of expected transaction
rate (workload), response time, growth factor, user capacity, database size,
and others. The time invested in this phase can be reduced when such
information already exists and can be provided to BEA.
- Performance Assessment Management Reports
- Benchmarking
- Performance bottleneck identification and resolution
- Performance tools (such as workload generators) to be developed
- How to interpret metrics
- How to gather metrics
- What metrics to gather
- Reference architecture on Solaris platform (build-out)
Qualifications
3-4 experience on Weblogic platform, with over 20 projects. In-depth
understanding of BEA technology, able to plan and design the architecture of
middleware projects. Able to design architectures using associated
technologies (Java, Object-Oriented Design and Analysis, Solaris, Veritas,
Netscape/iPlanet, Oracle). Experience with large-scale, full life-cycle
projects. Able to share technical and business knowledge with others.

thnx for the reply, Alex.
in the application tier, i would like to have 2 boxes:
1) service applications only like MMS, UPSA in one box
2) can i include search service appln in second box which is also a search query server? or
i need to create/configure search serv.appln in the first box itself?
i have created a farm topology as shown above:
would like to know whether its correct or not:

Seeking a Weblogic capacity planning consultant

Xpede is an financial services application service provider based in
          Oakland, CA, seeking an capacity planning consultant for an short term
          (under one week) initial engagement, with possible on-going consulting work.
          Please send resume, qualifications, hourly rate or a fixed price bid to
          [email protected]
          Capacity Requirements Plan (2-5 days)
          The Plan concentrates on the responsiveness, scalability, and throughput of
          our middleware infrastructure through investigation, analysis, and a formal
          report. This service is conducted as a pre-deployment evaluation. The
          consultant will work with our staff to identify or establish our capacity
          requirements in quantitative terms, which express the requirements for an
          application and architecture, and will encompass both current and future
          timeframes. Performance metrics may take the form of expected transaction
          rate (workload), response time, growth factor, user capacity, database size,
          and others. The time invested in this phase can be reduced when such
          information already exists and can be provided to BEA.
          - Performance Assessment Management Reports
          - Benchmarking
          - Performance bottleneck identification and resolution
          - Performance tools (such as workload generators) to be developed
          - How to interpret metrics
          - How to gather metrics
          - What metrics to gather
          - Reference architecture on Solaris platform (build-out)
          Qualifications
          3-4 experience on Weblogic platform, with over 20 projects. In-depth
          understanding of BEA technology, able to plan and design the architecture of
          middleware projects. Able to design architectures using associated
          technologies (Java, Object-Oriented Design and Analysis, Solaris, Veritas,
          Netscape/iPlanet, Oracle). Experience with large-scale, full life-cycle
          projects. Able to share technical and business knowledge with others.


Hi,
What should be the Heap size cannot be determined by the RAM size 32GB or 64GB. It depends on the Application...Application Functionality, How many Long living objects it creates , how many JNI calls it makes, How many Jars/Classes it loads, How many users requests your application at peak load, what kind of Caching you are using...etc
Above things are the points which can be used to determine the required Heap Size and the JVM tunning options.
There is no standard calculation or Formula available...The only option to determined the required Tunning parameters and JVM settings is...Load Testing and Performance Testing.
Thanks
Jay SenSharma
http://middlewaremagic.com/weblogic/?page_id=2261 (Middleware Magic Is Here)

How to consider production capacity plan on sales order's delivery date.

HI experts,
How to consider production capacity on sales order's delivery date.
Say for example :
Production workcenter capacity to produce the XYZ material 10qty in a Day.
today i have created the sales order XYZ materail qty of 100. lead days has give based of MMR 21 days 17-07-2010( 3 sundays)
and in same day i have created other sales order XYZ material qty 5000 ,the lead days again it will came 17-07-2010.
Now my client requiremnt is when the schedule days is confirming in the sale order it sholud also consider the production capacity planning also
Even though my work centers are overloaded it is confirming the sales order on 17-07-2010 (based on leads days in MMR)
Thanks in Advance..
Sudha

Sudha,
Tell your client he needs to install SCM. Here, you can use either CTP or MLATP to confirm orders based on production capacity.
There are a couple of standard workarounds available in R/3.
One is to create a constrained forecast for planning materials, consistant with capacity, convert the forecast to PIRs, and then perform ATP against 'Planning' rather than against product, using MRP Strategy 63.
http://help.sap.com/saphelp_erp60_sp/helpdata/en/a2/4e799a927d11d1800f00c04fadcb73/frameset.htm
Another method is to leave the Sales orders unconfirmed at original order entry (eg, no lead time). Run MRP, run capacity levelling, firm the levelled planned orders, and then run BOP (called 'rescheduling' in R/3) against the unconfirmed backlog, with an ATP that considers firmed planned orders as supply elements. The sales orders will confirm against the planned orders, which are scheduled according to available capacity.
Either of these two methods will work, but the CTP SCM solution is by far the better answer. Downside of SCM solution in general is the substantial implementation effort required.
Rgds,
DB49

Need Help - BPEL sizing & capacity planning

Greetings,
We are considering moving our EDI processing to Oracle Integration with BPEL. I have searched high and low for sizing and capacity planning guidelines for BPEL (and B2B) but have not been successful.
Could one of you help me out with this - point me to links, send me actual documents etc. My contact information is included below.
Thanks very much in advance.
Milind Gadre
Advanced Solutions Architect
adx - Advanced Data Exchange
t: 510.493.5119
e: [email protected]

Thare are some resources:
Oracle Appserver Performance Guide, chapter 7 Tuning BPEL:
http://download-west.oracle.com/docs/cd/B31017_01/core.1013/b28942/tuning_bpel.htm
BPEL Webinar, performance tips
http://www.oracle.com/technology/products/ias/bpel/htdocs/webinars.html
In general every customer has his own unique business processs implemented. Based on the implementation of the application architecture of the BPEL processes, the best way to size the system is to measure it. What is the expected / estimated load of message that BPEL PM will process? What type of adapters is used? etc...

OAG Capacity Planning

Hi,
Please help me to clarify on how to properly plan the capacity for an OAG deployment. I’m taking as base OAG documentation:
http://docs.oracle.com/cd/E55956_01/doc.11123/administrator_guide/content/admin_planning.html#p_admin_planning_ha
formula described there helps to get the number of Gateways
numberOfGateways = (ceiling (requiredThroughput x factorOfSafety / testedPerformance) x HA) x (1 + numDR) + staging + eng))
I’m assuming each gateway is deployed in machines with following prerequisites, as described in documentation
Intel Core or AMD Opteron at 2Ghz with Dual Core or faster—i386 or x86_64 (32-bit or 64-bit)
Minimum 2 GB free disk space, 50 GB recommended
Minimum 4 GB physical memory
In the case requiredThroughput has significant increments, I think we should deploy additional Gateways, please help us to clarify
Questions are:
In order to properly scale OAG deployment, its just required to deploy new Gateways (in additional machines with above characteristics)?
Increasing physical memory or adding more CPU threads has any positive impact in performance?
Please advise.
Regards
Juan

Hope you are asking regarding the Manufacturing Capacity Planning.
Oracle Capacity functionality is part of the Manufacturing module license.
Oracle Capacity lets you calculate your Capacity load ratio by resource or production line.
It ensures that you have sufficient capacity to meet your production requirements.
Oracle capacity provides you with two levels of capacity planning :
Rough Cut Capacity Planning (RCCP) to validate your master wschedules
Deetailed Capacity Requirements Planning to validate your MRP plans.
Check the Capacity User Guide available at Appsnet
(http://appsnet.oracle.com -> Manufacturing -> User and Implementation Guides)
for more details.
Warm Regards
Vinod Subramanian

Operations are not getting despatched in capacity planning table

Hi Experts,
I am using capacity planning table (Graphical) to level the capacities and sequence the process orders. The start date of my orders are very well in future and sufficient capacities also is available in resources. But when I go to CM25 and select one order and click on despatch, system is not despatching it. This is the case for almost all the overall profiles.
The surprising factor is, this function was working fine earlier without any issues. I havent done any changes in configs which will affect capacity planning.
What can be the reason for this? Any thoughts please?.
One more thing. Is there any option to avoid the capacity planning step if Iam using R/3 and manage it by some other way?
Appreciate your earlier reply
Thanks & Regards
Prathib

Most likely you didnt define the rowsource key properly. Please look under <install>/errors folder for any file there.
Also, please read the documentation. There is information there about how to troubleshoot a loading problem. Please always read the documentation and we would appreciate feedback on the documentation.
http://download.oracle.com/docs/cd/E17236_01/epm.1112/iop_user_guide/frameset.htm?launch.html

Oracle Performance Forecasting / Capacity Planning

Similar Messages

Maybe you are looking for