TimesTen performance evaluation

Hi all,
I'm in a PoC and I have to compare the performance of an Oracle DB with and without the TT Cache.
Is there an application which I can download that already does this? Generating workload and showing the results?
Thank you!

Typically TimesTen compliments the Oracle database. If the Oracle database can achieve the performance you require then there's no need for TT cache. If however Oracle database can't delver the performance needs of your application then you can look to use a caching technology such as TT cache to deliver microsecond response times and increased throughput by caching the data in-memory in the mid-tier. This is a long explanation to position why we don't have any specific code to compare the 2 products as they don't really compete with each other.
There is some benchmarking code provided as part of the TT install in the quickstart directory. In particular a benchmark call tptbm. There's a JDBC and OCI version, the source code is there and could be modified to run against the Oracle database.
Tim

Similar Messages

Database Performance Evaluation Benchmarking Tuning

Does anyone by chance have any articles or websites that deal with Oracle (or generic) database performance evaluation, benchmarking, and/or tuning for a STANDALONE PC database installation? I have Oracle 10g installed on my personal machine and I want to find some information that will help me place the performance evaluation into a more professinal context. Any links relating to this would be very appreciated.
regards,
John

Why do you expect there to be a difference in evaluating performance for a standalone PC database installation vs. a server database installation? Other than the fact that, presumably, you won't find network events in the top wait events, I'm hard-pressed to think of any differences.
You would evaluate performance and tune the system the same way on a standalone database as on a database on a server box-- you would figure out what operations are important, figure out how quickly those operations need to run, figure out which of the important operations are running slowly, figure out what those operations are spending time doing, and then figure out how to reduce the runtime. Of course, each of those steps can potentially be rather involved. There are plenty of articles and books on performance tuning-- Oracle has a few manuals, Jonathan Lewis's book on the cost based optimizer is excellent, Cary Millsap's optimization book is top flight, etc.
As for benchmarking, unless the intention is to run something like the TPC benchmark on your desktop, which would seem odd, your benchmark is generally closely tied to your application-- i.e. figuring out how quickly the system performs a particular business operation. Generic benchmarks like TPC tend not to be particularly useful in the real world because they are unlikely to mimic your real workload.
Justin

Supplier performance evaluation

Dear Guru's
Can we include supplier survey response and transaction data from SAPSRM/ECC to determine the KPI of suppliers? Can we also include the supplier status like blocked or black listed, expired in the report?
Can we include category of supplier (like small scale industry) in Supplier evaluation report?
You thoughts on it will be highly appreciated.
Thanks and regards,
Ranjan
Ranjan Sutradhar

Dear Experts,
As mentioned in the above memo, a supplier can not see its performance evaluation by the buyer in SUS. Could you please share your thoughts on how we can enable our suppliers to see the vendor evaluation report in the SRM system?
Any thoughts or idea will be highly appreciated.
Thanks and regards,
Ranjan

SAP Performance Evaluation

Hi: everybody
We have implementated SAP for one year. Our headquarter will check out SAP system. I have to write a report to show our SAP system performance evaluation. I wish him to satisfied with my team job. so this report will include hardware, business operation, user training, and so on. I think at least.
Can everybody provide me a template about this report?
Thank you!
Henry

Hi,
Very intersting query. I spent some time and got this link:
https://www.sdn.sap.com/irj/sdn/netweaver?rid=/webcontent/uuid/a06a3512-1a23-2a10-caa5-ee60d87ec953 [original link is broken],
Please keep us updated if you find something useful.
Regards,
Rakesh

Agent performance evaluation

Hi ,
My company need to deploy the uccx script to add new feature with old script the feature is agent performance evaluation quality. once the agent the hung off the head set the system transfer the call the evaluationperformancequality and the customer press 1 excellent 2 very good 3 good 4 bad . and send all data to SQL server the data incloud ( time duration , agent ext , caller ID , date and time ) ,
have we here uccx 8.5 the script attached. call manager 8.5, unity 8.5
please i need assistance to create this feature in this script and how to trunking with SQL server
thanks

This is called post call survey and in UCCX is typically invoked with CAD workflow.
Chris

Can large page improve timesten performance on aix??

hello, chris:
Can large page imporove timesten performance on aix? we havenot test it yet, I just want to know whether it is ok?? thank you.

Enabling large page support on AIX may help performance under some circumstances (typically with large datastores). TimesTen does support use of large pages on AIX and no special action is needed on the TimesTen side to utilise large pages. Here is some information on this which should eventually appear in the TimesTen documentation.
The TimesTen shared memory segment for AIX has been created with the flags
( SHM_PIN | SHM_LGPAGE) necessary for large page support for many releases.
It used to be the case that special kernel flags needed to be enabled so it
was never documented as being supported. Subsequent AIX releases have made
enabling large pages a matter of system administration.
1) Enable capabilities (chuser)
2) Configure page pool (vmo -r)
3) Enable memory pining (vmo -p)
4) [restart timesten daemons]
The documentation is quoted below.
AIX maintains separate 4 KB and 16 MB physical memory pools. You can
specify the amount of physical memory in the 16 MB memory pool using the vmo
command. Starting with AIX 5.3, the large page pool is dynamic, so the amount
of physical memory that you specify takes effect immediately and does not
require a system reboot. The remaining physical memory backs the 4 KB virtual
pages.
AIX treats large pages as pinned memory. AIX does not provide paging
support for large pages. The data of an application that is backed by large
pages remains in physical memory until the application completes. A security
access control mechanism prevents unauthorized applications from using large
pages or large page physical memory. The security access control mechanism
also prevents unauthorized users from using large pages for their
applications. For non-root user ids, you must enable the CAP_BYPASS_RAC_VMM
capability with the chuser command in order to use large pages. The following
example demonstrates how to grant the CAP_BYPASS_RAC_VMM capability as the
superuser:
# chuser capabilities=CAP_BYPASS_RAC_VMM,CAP_PROPAGATE <user id>
The system default is to not have any memory allocated to the large page
physical memory pool. You can use the vmo command to configure the size of
the large page physical memory pool. The following example allocates 4 GB to
the large page physical memory pool:
# vmo -r -o lgpg_regions=64 -o lgpg_size=16777216
To use large pages for shared memory, you must enable the SHM_PIN shmget()
system call with the following command, which persists across system reboots:
# vmo -p -o v_pinshm=1
To see how many large pages are in use on your system, use the vmstat -l
command as in the following example:
# vmstat -l
kthr memory page faults cpu large-page
r b avm fre re pi po fr sr cy in sy cs us sy id wa alp flp
2 1 52238 124523 0 0 0 0 0 0 142 41 73 0 3 97 0 16 16
From the above example, you can see that there are 16 active large pages, alp, and
16 free large pages, flp.
Documentation is at:
http://publib.boulder.ibm.com/infocenter/pseries/v5r3/index.jsp?topic=/com.ibmaix.prftungd/doc/prftungd/large_page_ovw.htm
Chris

Performance Evaluation to be done by HR Specialist

Hi All,
In Fusion Performance Management cycle, Line Manager will evaluate and provide the rate to Goals and Weightage in the performance document of employee.
Is there a possibility were HR Specialist can do the evaluation on behalf of the Line Manager.
Client HR wants to an evaluation and over write the ratings in Performance Document of employee.
Regards
Pradeep Badiganti

1. PL cannot be combined with CL, If employee is combining and applying for CL which has PL before then at the time of leave application, system to throw error message and should not allow him to apply for it
Ans:- For this you need to write a BADI/USER EXIT and Need to develop atable in se11 for Combination leaves this can be done by ABAPers
2. If employee is applying for sick leave in between PL then while applying for SL, should throw error message and should not him to apply for it
Ans:- For this you need to write a BADI/USER EXIT and Need to develop atable in se11 for Combination leaves this can be done by ABAPers
3. Holidays / Weekly offs falling betn PL treated as PL
Ans:-In this Table you can do this T556C or in Time Management -> Define Counting Rules
4. To chk for Saturday / sundays / holiday, if these holidays are falling between PL then PL to include / treat these holidays also
Leave taken during notice period, should throw error message and should not allow the user to apply
Can you Explain me in brief 4th Question
Warm Regards
Raja Sekhar

Survey: Performance Evaluation of ERP-Systems

Hello,
In course of my Bachelor thesis I´m conducting a survey about u201EPerformance evaluation of ERP-Systemsu201D. I would like to ask you to take part in this survey; it will take you only about 10 minutes. The link below will lead you to the survey, which is available in German language only.
http://erp-performance-umfrage.qpk.info
Thank you very much for your time!
Veronika Schreiber (Student at Technische Universität München)

you need to look at the landscape the current system is in.
How many systems are required for the full end to end processes?
User friendly?
What can SAP do that the other systems cant?
Will these things/ processes add value to the business i.e. MSS/ ESS, SRM, FSCM
How much manual work is required for month end?
SAP it can be automated, and the key month end closing should be able to be reduced.
Hope this gives you some ideas of what else to consider.
Please award points if this is useful.

Degraded TimesTen performance

Greetings,
I am on Solaris using TimesTen Release 7.0.5.0.0.
I fixed some threading bottlenecks in our application. Unfortunately this seems to put more pressure on TimesTen and degrades the overall throughput of our system. :-/
Someone suggested running "ttXactAdmin -latchstats show" so I did but I have a bit of trouble interpreting the results.
I ran the old software and the new software against the same performance test.
I discovered that many of the SpinCnt values are much bigger now (anywhere between 4x and 8x bigger) but what does it mean?
The number of collisions is also about 4x bigger (values range between 5 to 100) .
The important question is what does the ouput of ttXactAdmin mean and why is it worse now ? (well two questions.. ;-)
The second question is how can I improve this without touching the DB schema ? The DB schema does need some optimization work but if I could improve the performance a bit before making such changes it would be ideal.
Thanks,
Serge

Hi Serge,
Firstly, you need to be sure you are capturing accurate latchstats. Maybe you know this already but just inc case, what you need to do is, when you workload is running a reasonably steady state, run the commands:
ttXactAdmin -latchstats off dsnname
ttXactAdmin -latchstats clear dsnname
ttXactAdmin -latchstats on dsnname
sleep 60
ttXactAdmin -latchstats off dsnname
ttXactAdmin -latchstats show dsnname
This ensures you are capturing latchstats for just a 60 second period rather than any that may have been ther before. Do this for both cases (before app optimisation) and afer.
It is very likely that improving multi-thread / multi-CPu concurrency in the app will shift the concurrency issues into TimesTen. Basically high(er) values of colilisions, average spincnt and sleep count reflect increased contention within TimesTen. Depending on which objects these affect you may (or may not) be able to make some optimisations. There are several common contention points you may see depending on your workload:
1. Log buffer insertion latch. This is a common one if your workload has a lot of insert/update/delete operations. Not much you can do about this except to reduce the volume of write operations or use fewer concurrent application processes/threads.
2. Index latches (mainly for T-Tree indexes). Little to be done about these other than remove the indexes or split the table (and hence the indexes0 into multiple tables/indexes.
3. Command latches. Eliminate these by specifying PruvateCommands=1 in the ODBC attributes.
4. Heap latches. If these are on temp heaps not much can be done. if they are on Perm heaps you can try increase the number of perm heaps. the way to do this is undocumented but is sort of supported :-)
An important note; if you are using direct connection mode or local (intra machine) client/server then it is totally counterproductive to have more concurrent processes/threads accessing TimesTen than the number of CPU cores in your machine. Depending on the workload and what else is running on the machine you will see peak throughput and optimum response time with <= N concurrent processes/threads where N is the total number of CPU cores in the machine. if you are using remote client server then the optimal number of connections is obviously < N since each connection spends some time waiting for the network round trip.
Chris

ASM performance evaluation

Dear All:
We got the 8 Luns of each 100GB distributed across a diskgroup for our data/index storage. We ran massive update/insert statements on the tables built with indexes so there were numerous dbfile sequential reads.
We run the ADDM reports based on snapshots taken during these updates and Oracle alerted on I/O throughputs.
***The average response time for single block reads was 99 milliseconds.
*** Average datafile I/O throughput was 720K per second per reads and 1.2M per second per writes.
*** Even we already implemented ASM, oracle recommendation was to engage ASM, which is confusing.
*** On some index files, the average response time is >500 Milliseconds for single block.
Now what is meant by average response time per single block ? Is it database block or the stripe size for the diskgroup (1 MB I assume).
On the dbconsole, ASM performance TAB, we could see the response time, throughput in MB per second. Is there any script to get those info ? And what is an ideal benchmark for ASM throughput, single block average response time.
Any help, advise, suggestion will be highly appreciated.

user12018084 wrote:
On the dbconsole, ASM performance TAB, we could see the response time, throughput in MB per second. Is there any script to get those info ? And what is an ideal benchmark for ASM throughput, single block average response time. I do not understand how you can benchmark ASM using I/O.
I/O is directly dependent on your I/O subsystem. The driver, the hardware, the cable, the storage array, the speed of the disks. For example, if you use a PCI-X HCA card for your connectivity to a storage systen and not PCI-E, you have a very specific max thruput imposed on your I/O layer. Which would not have existed if you used PCI-E instead. Or your storage array could be implementing RAID5 without ASIC support, resulting in a hefty overhead for parity calculation on write I/O. Or your RAID10 storage array could have 5 striped sets across 2 disks per set, instead of a single set across 10 disks.
And now you want to measure I/O at ASM level (btw, who - which process- actually writes to disk?).. and then blame ASM for poor I/O performance?

Cisco nam 2204 performance evaluation

Hi,
We are planning export 500 Netflow of ISR G2 to a NAM2204-RJ45 appliance.
But I Cannot find information about NAM2204-RJ45 Netflow Data Export performance.
Can someone share user experience?

The 2204 is an appliance with 4 gigabit ports where the ISR is a module with two ports.
2204 data sheet:
http://www.cisco.com/en/US/prod/collateral/netmgtsw/ps5740/ps5688/ps10113/data_sheet_c78-655535.html
ISR data sheet:
http://www.cisco.com/en/US/prod/collateral/netmgtsw/ps5740/ps11658/data_sheet_c78-655516.html

Umfrage: Performance Evaluation von ERP-Systemen

Hallo,
Ich studiere an der Technischen Universität München Wirtschaftsinformatik. Im Zuge meiner Bachelorarbeit mache ich eine Umfrage über die u201EBeurteilung der Leistungsfähigkeit von ERP-Systemenu201C. Ich würde mich sehr freuen, wenn Sie mich unterstützen könnten, und den Fragebogen ausfüllen. Es beansprucht nicht mehr als 10 Minuten. Die Umfrage ist auf Deutsch und bezieht sich auf Unternehmen/Organisationen, die mit ERP-Systemen arbeiten. In dem Fragebogen geht es im Allgemeinen darum, ob Sie/Ihr Unternehmen die Leistung Ihrer ERP-Systeme messen und beurteilen.
Die Auswertung der Befragung erfolgt völlig anonym und lässt keine Rückschlüsse auf Ihre Person oder Ihr Unternehmen zu.
Vielen Dank für ihre Unterstützung,
Hier ist der Link zur Umfrage:
http://erp-performance-umfrage.qpk.info
Veronika Schreiber

Hallo,
Vielen Dank für Ihre zahlreiche Teilnahme an meiner Umfrage u201EBeurteilung der Leistungsfähigkeit von ERP-Systemenu201C.
Heute (05.01.2011) ist der letzte Tag der Umfrage. Wenn Sie heute noch 10 Minuten Zeit finden, würde ich mich sehr freuen, wenn sie an der Umfrage teilnehmen würden!
Über jede weitere Unterstützung würde ich mich noch sehr freuen.
Vielen Dank!
Veronika Schreiber

Maximize LR CC Performance - Evaluating system info

Hello,
I have a fairly powerful Windows 7 machine and really appreciate the performance gains in LR CC. After taking a look at my System Information, I'm wondering if I'm making full use of my hardware. Below are some excerpts from my LR System Information and a few questions:
Logical processor count: 12
Processor speed: 4.0 GHz
Built-in memory: 65477.5 MB
Real memory available to Lightroom: 65477.5 MB
Real memory used by Lightroom: 754.9 MB (1.1%)
Virtual memory used by Lightroom: 728.6 MB
Memory cache size: 0.0 MB
Maximum thread count used by Camera Raw: 6
You can see that I have 64 GB of physical system RAM. Can I have LR use more than 1.1% of this?
Should I care that the "memory cache size" is 0.0 MB? Would a memory cache size greater than zero improve performance?
Check OpenGL support: Passed
Vendor: NVIDIA Corporation
Version: 3.3.0 NVIDIA 350.12
Renderer: GeForce GTX 580/PCIe/SSE2
LanguageVersion: 3.30 NVIDIA via Cg compiler
Adapter #1: Vendor : 10de
Device : 1080
Subsystem : 15883842
Revision : a1
Video Memory : 3020
Build: Uninitialized
Direct2DEnabled: false
GPUDevice: not available
OGLEnabled: true
Can someone tell me if I should care about Direct2DEnabled being false? Do I want this to be true?
Thanks for any suggestions,
~Steve

AlanUniqueName wrote:
Just a thought, if anyone from Adobe is reading this, I am also a little surprised regarding what is not included from the system info that could help you or any of the many experts around here to solve problems - for example...
Disk Free space on the drives used by LR.
Details of Graphics cards and drivers.
More details of monitors and profiles in use.
Actually all HDs and the % amount of space used may be useful. If any drive is getting quite full, then the entire system can slow down in my experience. A few times when my OS seems to be grinding to a halt, I've realised I need to free up some space on my storage drives. Macs seem a bit more sensitive to this it seems than PCs. Even if I have 150G of space left on a drive, if it's a 2TB drive, that is only 7.5% free space left.

JVM performance evaluation

I'm currently working on a project to evaluate the performance of different JVMs on the Itanium 2 platform under Linux. The main idea is to evaluate the performance of WebLogic application server using different JVMs on the forementioned platform.
The problem I'm facing now is how to do this so that I get valid and accurate results. Does anyone know how to do this? Is it a good idea to use the benchmarks by SPEC? Which of them (SPECjbb2000, SPECjAppServer2002)?
Thanks for the help.

I'm currently working on a project to evaluate the
performance of different JVMs on the Itanium 2
platform under Linux. The main idea is to evaluate the
performance of WebLogic application server using
different JVMs on the forementioned platform.
The problem I'm facing now is how to do this so that I
get valid and accurate results. Does anyone know how
to do this? Is it a good idea to use the benchmarks by
SPEC? Which of them (SPECjbb2000,
SPECjAppServer2002)?
Thanks for the help.
Its a good idea to go by the 'benchmark', otherwise you need to do the following.
1. Set up a test bed consisting of various machines loaded with linux from diffrent vendors such as redhat, mandrake etc
2. Install JVM from diffrent vendors like Sun, IBM, x, y ...
3. Launch weblogic on those JVM using diffrent options in -mx360m -msxxxm, -mixed, -server as is required. (check for java -X options or jvm.cfg)
4. If you have a good chunk of web application source, you can try compiling using -g option, which adds debugging information to the class files and thus grows the size of class file. the idea is to get large class files to stress out the JVM.
5. Identify a substatially large web app/testsuite and do a load test on it. (*or* become a SUN licensee and get a j2ee JCK)...
This was something that we did couple of months(>12) ago, There might be better solution today.
SA

Performance evaluation for file storage.

If I were working in C/C++ I would know what should be an appropriate record size so it could be stored in one cache line, and processed therein; precisely how many pointer transitions will be required to fetch a record, and how microprocessor registers will be used.
This gives an idea about the level of optimization I want to achieve. Can you please point me towards literature that gives a detailed description of how Java Card handles its variable and object references at the lowest level? This becomes a more pertinent issue in case of Java Cards.
While I understand that individual card vendors have specific JVM and JCRE implementations, what is the lowest level I can go to while I'm working with Java card development kit? As of now I plan to use the SimpleFile and Record classes availalbe with kit samples.
Thank you for your help.

As of now I plan to use the SimpleFile and
Record classes availalbe with kit samples.I'll appreciate if you can suggest better approaches.

TimesTen performance evaluation

Similar Messages

Maybe you are looking for