Answer to Q re TLB entries in OpenSPARC T1

I received an offline question regarding TLB entries:
Are TLB entries tagged with the strand that inserted them?
The answer to this is that there is no "strand ID" in the TLB entries. That's because TTEs in the TLB contain Partition IDs, which in the end serve the same function.
Now, regarding the inevitable question about Partition IDs -- It turns out that some information regarding Partition IDs was unintentionally omitted from the MMU chapter of the posted UltraSPARC Architecture 2005 spec (darft 0.8.7, Hyperprivileged edition).
Look for a new rev to be posted (at http://opensparc-t1.sunsource.net/) in the next couple of weeks that contains the Partition ID information plus some other minor updates/corrections.

Still looking for an answer on this one....
Maybe the Q's worded poorly?
Basically I'm wanting to know how to export L8 DAE tracks for use in other apps. (Pro Tools in my case).
I've got a workaround (create DTDM audio tracks and physically move the audio files onto them, then export), but it doesn't allow for printing of the TDM plugs that were inserted on the tracks (which were a significant part of the sound of the track).
Any help is much appreciated!
I'll try back at the DUC as well...
Rick

Similar Messages

No# TLB entries of UltraSPARC III, UltraSPARC T1 and UltraSPARC T2

Hi,
I'm working with a large In-memory DB and I would want to experiment with ISM and Large page support (4M) to increase the TLB hit ratio. What puzzles me is that an old processor such as UltraSPARC III has got 2 x 512 entries in its data TLB whereas UltraSPARC T1 and T2 have got 64 and 128 respectively.
Am I misreadung the specs of the T1/T2 processors?
Is there something new in the SPARC architecture I fail to see, that allow T1/T2 to have fewer TLB entries but maintain the performance positives of a large, but more expensive HW TLB?
Is T1 and T2 not intended for applications with large data sets?
BRs
/Sune

Good, because that's about as much as I know about it!! :-)
I understand the concept of TLB entries and how thrashing them causes performance problems, but I've never had to really look at them in depth for tuning. I like questions like this that I know something about and can find someone else's explanations.
The main thing I got from the document was being able to have each entry point to independent sizes. Do you think that would make up for the reduced number of entries in most cases?
Darren

About OpenSPARC T2 TLB entries?

Hi,
I know that there is 64 itlbs and 128 dtlbs on OpenSPARC T2. I want to do some tests only using 8 itlbs and 8 dtlbs, so I modify tlb_size parameter to 8 in SAM. On this condition, configuration is 8 cores and 8 threads per core, hypervisor is ok, but some threads make some error when running OBP.
Could someone can tell me why? And I want to know how many tlbs(itlb and dtlb) there are at least on T2.
Thanks a lot

andapo wrote:
I know that OpenSPARC T1 supports fine-grained multithreading for the four thread per core. My problem is that I cannot find the same information about T2 processor. It suppots fine-grained multithreaded, coarse-grained multithreaded, simultaneous multithreaded?I think the parallel statement for T2 would be that it supports fine-grained multi-threading, with 8 thread per core.
But of course it also supports fine-grained multi-threading for all 64 threads on the 8-core processor, and it supports coarse-grained multi-threading!
= Dave

No free space in directory entries for NTAB buffer

Gurus ,
In ST02 -> for table definitions buffer there was a red alert with the SWAP count as 953275 .
hence i Double clicked on the swaps count to see the details , which is shown below ,
efficiency :
hit ratio 70
size :
allocated : 6799
available : 5703
used        :5703
free         : 0
Directory entries :
available : 20000
used      : 20000
free        : 0
swaps :
objects swapped : 953833
frames swapped : 0
what will happen if the 'directory entry ' and 'size' is zero we.
what process will get affected because of this ?
what i need to do next ..
guide me gurus

Hi Susin,
To Answer your question, directory free entries are '0' --> which means table definition buffer is full. That is the reason we can see "objects swapped : 953833"
Try to increase the size of table definition buffer which is low '6799' now.
I feel you can even double the value and observe the performance.
For General information regarding Hit ratio and Swapping:
The quality of a buffer and how often it is accessed is measured by the
'%Hit Ratio'. This value will indicate if the information stored in the
buffers, such as table entries, programs and screens, is being hit
directly from the buffer itself or, on the other hand, if the system
needs to bring that data from the database since it was not found in the
buffer.
To find out the buffers with poor quality, first check the system
startup time. When the system is started, all buffers (except the
program buffer which has a pre-load) are initially empty. Therefore,
all objects that are accessed for the first time have to be read from
the database and then loaded into the buffers.
If objects are not yet in the buffer, the hit ratio for the buffer will
be low. The hit ratio increases from the time objects are loaded into
the buffers. The rate of the increase depends on the workload in the
system and is different for each buffer.So it is to be noted that how often you restart the system which resuls in the loading of object again and causes hit rate to be low.
Poor buffer quality is not always due to a real problem. For example,
transports into a system can reduce buffer quality. Keep in mind though
that a value lower does not always shows that you have a problem.
Swapping occurs when the buffer is full, and the SAP System has to
load additional objects into the buffer. Objects in the buffer that
were used the least recently are removed. In this context, the term
"swap" means the objects removed from the buffer are lost and cannot
be replaced until a new database access is performed (replacing what
was lost).
There are two possible reasons for swapping
1 There is no space left in the buffer data area
The buffer is too small. You should increase the buffer size.
2 There are no directory entries left.
Hope this helps.
For any other information, feel free to reply me.
Thanks and Regards,
Pavan Kumar Gali.

OpenSparc used in a multi processor platform?

Hi,
I'd like to know if it is possible to implement two OpenSparcT1 into two different FPGAs so as to have a multiprocessor platform. Is there a pre-built platform that handles memories, communication between processors and between memories and processors and all software partitionning for the applications to be run on the processors?
Thanks for answering,
Regards.

The current design assumes the OpenSPARC board (ML505-XC5VLX110T). So the pinout and everything is set assuming that platform. If you go to another platform, you'll need to make changes to work with that platform. Among other things, you may need to use a different memory controller, or at least compile the memory controller with different parameters. Then you'll need to figure out how to communicate between the two OpenSPARC cores. If the FPGA is large enough to fit two OpenSPARC cores, then the communication between the cores gets easier. Otherwise, you'll have to evaluate what options you have on the target platform for communicating between two FPGAs on a board, or between two Boards. It may be that the Xilinx Aurora block may still work, even though it is running over a different physical interface. Of course, if your platform is an FPGA from a different manufacturer, you'll have to start from the beginning with all-new IP.
The current two core system is implemented with one core on each FPGA board. On each board, the bit file includes the OpenSPARC core, the Xilinx MicroBlaze controller, memory controller, serial port, and the Aurora interface, which connects to the other board. Different software is run on the board, however. On the slave board, the firmware simply forwards memory requests to the master board. Memory on this board is mostly unused, with only a small part being used for the MicroBlaze code and data. On the Master board, the firmware processes all memory requests, from the local core, as well as the remote core. On this board, memory is partitioned with a small part being used for the MicroBlaze code, about half being used for the OpenSPARC memory space, and the other half being used for the OpenSPARC RAM disk image.
The next question was about runtime scheduling of an executable. If the executable has only a single thread of execution, it will run on one core, one hardware thread, and cannot be partitioned between two cores, or even two hardware threads. If some mechanism has been used to create multi-threaded execution in the program (like explicit programming using pthreads, or maybe OpenMP), then at run-time, the operating system (OpenSolaris) will schedule the different threads of the program onto different hardware threads or different cores. If you try to run a program directly under Hypervisor, without an operating system, then all the partitioning would have to be done explicitly in the program itself.
Hope this helps,
formalGuy

Advanced action: jump to next slide if all text entry boxes are filled out. If not, pause project and show error message

Hi there
Working with captivate 4 and advanced actions, which is a great new feature, I encounter the following problem:
I have a sign-in slide, where the user has to insert his name and several other data. For user-friendly handling, I have the text entry boxes appearing one after another, so when the user confirms his entry, the next text entry box appears and has the focus.
The button for jumping to the next slide works with an advanced action:
Only if the user has filled out all the text entry boxes, the action "jump to the next slide" is performed.
If not, an error message is shown.
It works well so far.
Now, I want the project to pause there, and not to jump back to the beginning of the slide (as suggested in another Thread in this forum), because then the text entry boxes dissappear again.
I tried to set the movie control variable rdcmndPause to 1 in order to make the movie pause, but nothing happens.
I tried it with an external playback control bar and without, it didn't work in either case.
I found the hint:
To control the project through the skin file, add the prefix cpSkinLoader_mc to the variables. This might be the solution.(?)
But how can I change the name of a MovieControl Variable?
If anyone knows, how this is supposed to work or has another suggestion to solve the problem, please let me now.
Thanks a lot.
Petra

Hi Petra,
I believe I can help you, but from a completely different angle than anything you've tried yet. I try to find simple solutions to complex problems. No advanced actions will be needed, just a little creative timing. First, I need to be sure I correctly understand what it is you're hoping to accomplish.
If I understand you correctly, you've got a single welcome slide. On this slide you've got several TEB's (text entry boxes), set up with variables to make your project more personal or gather some required information. These TEB's have no right or wrong answers, they simply require an entry: such as an address which begins with a numeric key. The TEB's are set to appear one after another, as one is completed the next one appears. Once all have been completed a next button, set to "jump to next slide" takes them to the next slide and on through your course. You want the project to pause until the user clicks this button, and only proceed if all TEB's have been entered. If not, you wish an error message to appear.
If the above description is correct, I'd do the following:
Leave your TEB's to show up one at a time as the previous is submitted, as you've described. (each one pauses the slide till info is submitted, then "continues" the slide and the next TEB displays)
Set each TEB to "display for rest of slide".
For the next button: In the properties dialog box, options tab, name the button "Next_button" and uncheck the visibility box.
Open the properties for the final TEB that students will fill in and set the on success property to "show" then choose the Next_button by clicking the drop down arrow beside the "show" field. This will hide the Next button until all TEB's have been filled in.
There is no error message, because there is no possibility of error. Why tell a student they've messed up, just eliminate that possibility! All TEB's only display after the previous is entered, and the Next button is hidden until the last TEB is entered.
I've tested this in a simple 3 slide project and it all worked fine, including some variables. If you need more detailed info, or would like to see the sample proj. I created let me know.
Hope this helps.
randy

What are the mapfiles for and when should I use them?

Hi,
what are the map files for and why/when should I use them?
/usr/lib/ld/map.noexstk
usr/lib/ld/map.noexbss
/usr/lib/ld/map.noexdata
/usr/lib/ld/map.pagealign

Hi, I'm sure you've read the comments, but for the benefit of those who haven't: /usr/lib/ld/map.noexstk # Linker mapfile to create a non-executable stack definition within an # executable. /usr/lib/ld/map.noexbss # Link-editor mapfile to create a non-executable bss segment definition # within an executable. /usr/lib/ld/map.noexdata # Link-editor mapfile to create a non-executable data segment definition # within an executable. These stop the various segments from being executable. Which could reduce the risk surface for exploiting security holes in an application. /usr/lib/ld/map.pagealign # Linker mapfile to create page-aligned data within an executable. Aligning the start of the data segment can reduce the number of TLB entries needed to map the data. So the smallest TLB page size is 8KB, then we have 64KB, then 4MB etc. If the data starts on a 4MB boundary then it can be mapped with 4MB pages, if not then it has to be mapped with 8KB pages until a 64KB boundary, then 64KB pages until we get to a 4MB boundary. You sometimes get a bit of performance from this, but you may also get performance stability from it. If the mapping is random sometimes it works well, sometimes it works less well. Regards, Darryl.

Cisco 6500 with sup720 booting to rommon

Please how can I solve this problem. This cisco 6500 boots to rommon mode, I can't see any error showing why its booting to rommon.

Here is the boot process:
System Bootstrap, Version 8.5(4)
Copyright (c) 1994-2009 by cisco Systems, Inc.
Testing lower main memory - data equals address
Testing lower main memory - checkerboard
Testing lower main memory - inverse checkerboard
Clearing lower memory for cache initialization
Clearing bss
Clearing autoboot state machine
melody_present_reg: 1st read w/ 0x5555
melody_present_reg: 2nd read w/ 0xaaaa, reversed: 0x5555
Bootdisk adapter is detected, enabling bootdisk access...
Reprogramming CS1 w/ Melody value...
Reading monitor variables from NVRAM
Reset reason for CPU board 0xffff , BaseBoard 0x281ffff, display 0x20000System Reset by Software.
Enabling interrupts
Initializing TLB
Initializing cache
Initializing required TLB entries
MOBILE dimm - rev 0.15
Initializing main memory
Sizing NVRAM
Initializing PCMCIA controller
Initializing USB2.0 controller
Exiting init
Cat6k-Sup720/SP processor with 1048576 Kbytes of main memory
Autoboot: failed, BOOT string is empty
Autoboot executing command: "boot "
Initializing ATA monitor library...
string is bootdisk:s72033-ipbasek9-mz.122-33.SXJ2.bin
Loading image, please wait ...
Initializing ATA monitor library...
Stack pointer       : 0x8FFFFF80
monstack            : 0x800FFFC0
monra               : 0xBFC26EC4
edata : 0x80108580
magic : 0xFEEDFACE
memsize             : 0x10000000
uncomp_size         : 0x05090C00
comp_size           : 0x05090C00
comp_checksum       : 0x98ECBE18
uncomp_checksum     : 0x98ECBE18
CZIP_MEM_BASE       : 0x80000000
_end       : 0x8010BA00
Self extracting the image...
IOS compressed src copy is     : 0x80108594
czip + IOS tar size is :0x00E01FA8
czip + IOS Tar image is now :0x80108794
cptr is now :0x8010BB00
IOS compressed dest copy is     : 0x8010BB00
[OK]
image_entry :0x80100000
image_entry :0x80100000
__start : 0x80100000
tar_avail_size is now :0x0428EA00
Tar image address is             : 0x80F0DB00
tar_size is           : 0x0428E600
cpu type                   : 0x00000019
uncomp_size                : 0x05090C00
monstack                   : 0x800FFFC0
image_info.entry_point   = 0x80100000
image_info.section_count = 0x00000005
image_info.monstack      = 0x800FFFC0
image_info.monra         = 0xBFC26EC4
image_info.param0        = 0x00000002
image_info.param1        = 0x00000000
image_info.param2        = 0x800066A8
image_info.param3        = 0x80100000
image_info.reg_k0        = 0x80F0DB00
image_info.reg_k1        = 0x0428E600
Section Index = 0x00000000
    source    = 0x8010BC1C
    dest      = 0x80100000
    bytes     = 0x00010000
Section Index = 0x00000001
    source    = 0x8011BC1C
    dest      = 0x80110000
    bytes     = 0x00000720
Section Index = 0x00000002
    source    = 0x8011C33C
    dest      = 0x80110720
    bytes     = 0x00000020
Section Index = 0x00000003
    source    = 0x8011C35C
    dest      = 0x80110740
    bytes     = 0x00DF1730
Section Index = 0x00000004
    source    = 0x80F0DA8C
    dest      = 0x80F01E70
    bytes     = 0x00000000
reg_v0: 0x00000000
reg_k0: 0x80F0DB00
reg_k1: 0x0428E600
tar_start: 0x00000000
tar_size: 0x00000000
Tar image address is             : 0x80F0DB00
tar size is :0x00B6130E
Tar magic : ustar Tar filename : C2LC memsize             : 0x10000000
Tar gid         : 035231Tar uncomp_size         : 0x00B6130E
Tar mtime         : 11672344432 Tar username          : ccaiTar comp_checksum       : 0x0000125F
Tar group name    : buildTar prefix    : tar_size in czip         : 0x0428E600
Stack pointer       : 0x8FFFFF80
monstack            : 0x800FFFC0
monra               : 0xBFC26EC4
edata : 0x80110740
magic : 0xFEEDFACE
memsize             : 0x10000000
uncomp_size         : 0x02B3E7F0
comp_size           : 0x00DF171C
comp_checksum       : 0xE917F280
uncomp_checksum     : 0xD094D890
Compressed IOS src copy is     : 0x80110754
tar_dest is :0x8BD69200
tar_size is :0x0428E600
Compressed IOS dest copy is     : 0x8AF77AE4
Tar src before IOS decompression is     : 0x80F0DB00
Tar dest before IOS decompression is     : 0x8BD69200
compressed IOS src is     : 0x8AF77AE4
IOS uncompressed dest copy is     : 0x8013C160
Self decompressing the image : ############################################################################################################################################################################################################################## [OK]
e_shoff :
0x02B3E660
e_flags : 0x10001001
e_phnum :
0x00000001
Source elf_hdr->e_shnum = 0x0000000A
Setting up to copy ELF section 0x00000001
to image_info section 0x00000000
sh_name = 0x0000000B
sh_type = 0x00000001
sh_flags = 0x00000007
sh_addr = 0x80100000
sh_offset = 0x00000060
sh_size = 0x02640000
sh_link = 0x00000000
sh_info = 0x00000000
sh_addralign = 0x00000020
sh_entsize = 0x00000000
Setting up to copy ELF section 0x00000002
to image_info section 0x00000001
sh_name = 0x00000011
sh_type = 0x00000001
sh_flags = 0x00000003
sh_addr = 0x82740000
sh_offset = 0x02640060
sh_size = 0x00373440
sh_link = 0x00000000
sh_info = 0x00000000
sh_addralign = 0x00000008
sh_entsize = 0x00000000
Setting up to copy ELF section 0x00000003
to image_info section 0x00000002
sh_name = 0x00000017
sh_type = 0x00000001
sh_flags = 0x00000003
sh_addr = 0x82AB3440
sh_offset = 0x029B34A0
sh_size = 0x0003EEE4
sh_link = 0x00000000
sh_info = 0x00000000
sh_addralign = 0x00000004
sh_entsize = 0x00000000
Setting up to copy ELF section 0x00000004
to image_info section 0x00000003
sh_name = 0x00000024
sh_type = 0x00000001
sh_flags = 0x10000003
sh_addr = 0x82AF2324
sh_offset = 0x029F2384
sh_size = 0x0000423C
sh_link = 0x00000000
sh_info = 0x00000000
sh_addralign = 0x00000010
sh_entsize = 0x00000000
sh_type = 0x00000008
sh_flags = 0x10000003
sh_addr = 0x82AF6560
sh_offset = 0x029F65C0
sh_size = 0x00000940
sh_type = 0x00000008
sh_flags = 0x00000003
sh_addr = 0x82AF6EA0
sh_offset = 0x029F65C0
sh_size = 0x021FE100
tar file start = 0x84D02F50
cpu type                   : 0x00000019
uncomp_size                : 0x02B3E7F0
monstack                   : 0x800FFFC0
image_info.entry_point   = 0x80100000
image_info.section_count = 0x00000005
image_info.monstack      = 0x800FFFC0
image_info.monra         = 0xBFC26EC4
image_info.param0        = 0x00000002
image_info.param1        = 0x00000000
image_info.param2        = 0x800066A8
image_info.param3        = 0x80100000
image_info.reg_k0        = 0x84D02F50
image_info.reg_k1        = 0x0428E600
Section Index = 0x00000000
    source    = 0x8013C1C0
    dest      = 0x80100000
    bytes     = 0x02640000
Section Index = 0x00000001
    source    = 0x8277C1C0
    dest      = 0x8274CFB0
    bytes     = 0x00373440
Section Index = 0x00000002
    source    = 0x82AEF600
    dest      = 0x82AC03F0
    bytes     = 0x0003EEE4
Section Index = 0x00000003
    source    = 0x82B2E4E4
    dest      = 0x82AFF2D4
    bytes     = 0x0000423C
Section Index = 0x00000004
    source    = 0x8BD69200
    dest      = 0x84D02F50
    bytes     = 0x0428E600
data_size in czip         : 0x00001000
bss end of IOS is         : 0x84D01F50
              Restricted Rights Legend
Use, duplication, or disclosure by the Government is
subject to restrictions as set forth in subparagraph
(c) of the Commercial Computer Software - Restricted
Rights clause at FAR sec. 52.227-19 and subparagraph
(c) (1) (ii) of the Rights in Technical Data and Computer
Software clause at DFARS sec. 252.227-7013.
           cisco Systems, Inc.
           170 West Tasman Drive
           San Jose, California 95134-1706
Cisco IOS Software, s72033_sp Software (s72033_sp-IPBASEK9-M), Version 12.2(33)SXJ2, RELEASE SOFTWARE (fc4)
Technical Support: http://www.cisco.com/techsupport
Copyright (c) 1986-2011 by Cisco Systems, Inc.
Compiled Thu 15-Dec-11 01:29 by prod_rel_team
Image text-base: 0x4010132C, data-base: 0x4274CFB0
Active crashed three times, disabling auto-boot and dropping to rommon
Firmware compiled 15-Nov-11 14:23 by integ Build [100]
*Jan 1 02:27:12.643: %PFREDUN-6-ACTIVE: Initializing as ACTIVE processor for this switch
*Jan 1 02:27:13.459: %PFREDUN-4-SUP_FORCE_TO_ROMMON: Supervisor forced to rommon with reason: Active crashed three times in a row
System Bootstrap, Version 8.5(4)
Copyright (c) 1994-2009 by cisco Systems, Inc.
Cat6k-Sup720/SP processor with 1048576 Kbytes of main memory
rommon 1 >

Field Cost Ctr is a required field for G/L account 5500

Hello Experts,
I need your help for one of the issue related to the cost center.
While posting the accounting document I am getting the below error -
Invocie Nr are: 95867831 and 95876503
Message is: Field Cost Ctr is a required field for G/L account 5500
52301203
Message no. F5808
Diagnosis
The value for field "Cost Ctr" in the interface to Financial Accounting
is an initial value but you are required to make an entry in the field
selection for G/L account "52301203" in company code "5500" linked to
the field selection for posting key "40".
Looking in the history of one of the two contracts I find the following:
Invoice 0095639591 from 16.11.2010, an accounting document was created
and in the accounting document CO object is cost center (and not COPA)
14901.
Account 52301203 has Field status group Z103 (cost - mad CostCenter)
Our user is saying that (eq 3529183592, invoice 95639591)
should have a cost center assigned in the equipment master. After
entering that cost center (14901) the accounting document was created
correctly.
However when we checked the "DefaultAcctAssgmnt" of the cost element 52301203, the filed Cost center is empty.
Can you please help with your suggessions?
Regards,
Reshma

from Note 518060 - FAQ: Physical inventory
12. Question:
Can I manually specify an account assignment (for example cost center) during the update of inventory differences?
Answer:
No. No. A direct entry of an account assignment for difference postings in the physical inventory (both MI07 and MI10...) is not possible.
A differentiation to different account assignments is only possible via a differentiation of the cost elements - refer to Note 13423.
In my words: FI has to change the field status

Partner Profile vanishes in BW system

Hi All,
We are facing a recurring issue which happens in BW QA system after we refresh is from the BW production system.
After about a week or so after the refresh, the data loads halt in the BW QA system and the BW team gets an error "EDI partner profile not available".
After i checked, i found that the logical system name entry for the EDW QA system (i.e. self system entry) is missing from the table EDPP1 and also from transaction WE20.
I re-created it in WE20 and the entry automatically got created in EDPP1 table.
My question is : I suspect that some job or application program is deleting this entry. We checked the background jobs but no luck.
How do i resolve this error. The client has come up with this question and he wants an answer as to why this
entry is being automatically.
Can you please help?
Regards,
Hari Kishan

Hi,
See the following SAP Notes:
SAP Note 886102 - System Landscape Copy for SAP NetWeaver BW
SAP Note 325470 - Activities after client copy in BW source systems
SAP Note 325525 - Copying and renaming systems in a BW environment
SAP Note 886102 - System Landscape Copy for SAP NetWeaver BW
SAP Note 1333302 - Special procedure for BW system copy
SAP Note 996238 - Dump 'RFC connection error' after BW system copy
Search for PDF in SDN
How Tou2026System Copy in SAP Business Intelligence System Landscapes
Thanks
Reddy

Any way to get more statistics and response-time info from cache processes?

I'm currently involved in an evaluation of Coherence for use as a global in-core cache/DB for data in the GB range. We'll try to run multiple cache process per cache in order to stay below 512MB heap-space per cache process.
In order to tune cache access times and understand the dynamic behaviour of the app we'd like to see some statistics info for cache access, e. g.
- requests processed per cache (process)
- requests answer time (per request, min, max, avg, median etc.)
- request answer size (number of cache entries returned, size of entries etc.)
- whatever else that helps ...
We are in position to get this kind of statistics info on the application side (when we call Coherence-APIs) but we'd need to see the statistics on a technical level per cache process.
We know about JMX/MBeans but that does not allow us to come up with log-files for statistical analysis and there is no way to correlate requests from the application to any statistics information (what request did cause what effect).
Any way to tell the cache processes to provide this kind of (logging) info (mesurements per request) and some summary.

David,
In addition to statistics that you can collect at the client tier (measuring the latency of individual cache calls), you could use a number of attributes exposed by the ServiceMBean.
If you look at a ServiceMBean for your partitioned cache service, the RequestAverageDuration and RequestMaxDuration attributes describe the latency of the underlying communication tier as seen from the client and does not include the serialization/deseriation cost.
The TaskAverageDuration, TaskMaxBacklog and ThreadAverageActiveCount expose the latency and expenses as seen by the server side, excluding the network tier expenses.
Also, there are number of statistics exposed by the CacheMBean related to the underlying backing map efficiency such as AverageGetMillis, AverageHitMillis, AverageMissMillis and AveragePutMillis.
You can also create your own subclasses for backing map implementation and implement custom MBeans that would expose any addition information you wish to collect.
Regards,
Gene

Using ARIS process models as the foundation for Business Process Monitoring

Hello,
Do anyone know if you can use IDS Scheer ARIS process models and the integration between IDS Scheer ARIS and SAP Solution Manager as the starting point for end-to-end Business Process Monitoring (BPM) so that you get the process description/model automatically in Solution Manager BPM and just need to configure the monitoring (more or less that you monitor against the modelled process in ARIS)?
Is there an API for SAP Solution Manager BPM so that non-SAP solutions/systems can be part of the end-to-end process monitoring?
If not, how do you usually integrate non-SAP solutions that are part of an end-to-end process that you want to monitor into SAP Solution Manager BPM, or do you not use Solution Manager BPM for this?
What is the roadmap for SAP around SAP Solution Manager BPM when it comes to non-SAP systems/applications/solutions and the new SAP NetWeaver BPM functionality?
Regards,
Markus Johansson

Hello Markus,
I just paste my answer form the other forum entry Re: Business process management strategy in SAP?
in a nutshell BPM Netweaver is focussing on so called edge-processes (usually
SOA-based and often workflow related). These edge-processes a thought to enhance exisitng core business processes and should provide the fast competitive edge versus competitors. Examples would be also xApps. For this Netweaver provides process modelling tools, where you can get exectuable code through the integration between CE and PI. The monitoring part for these processes is mainly targeting business end-users and business manager.
Business Process Monitoring with Solution Manager focusses on core business processes which are realized with the "conventional", "old" SAP Architecture, e.g. Order-to-Cash or Procure-to-Pay. You get a combination of technical monitors (average response times of transactions, background job runtimes, monitors for IDOC or qRFC processing) and application related monitors (you get more than 200 pre-configured application monitors out-of-the-box like number of overdue customer orders or outbound deliveries or number of deliveries with goods issue posted but no invoice). The target group here is the support organization of the customer, from Basis-Support over application support on IT side to process owner on business department side.
Overview presentations can be found under
https://service.sap.com/bpm --> Media Library --> Customer Information -->
"Business Process Monitoring - Part 1 & 2".
Or look under https://www.sdn.sap.com/irj/sdn/nw-processmonitoring
In the end BPM Netweaver and BPMon SolMan are complementary for different purposes and different target groups.
Best Regards
Volker

How do I change font in a calendar?

I can change the font of any additional text I add to a date, but is it possible to change the font of the numbers in the calendar to match my chosen font?
Thank you

I'm having the same issue. Did you get an answer? Seems that calendar entries created within Outlook on my laptop are fine, but entries created on my iPhone have a very tiny font. changing default font in "Settings - Display and Brightness" does not help. Thanks.

Boxee box and airplay

I bought a boxee box so I could use airplay. But the videos no longer work, I have sound but no picture. Why?

their site had the answer http://support.boxee.tv/entries/20757851-Streaming-from-iOS-devices

EDPAR and idoc segment E1EDKA1

Hi
   I have read all threads related to EDPAR but didn't get any satisfactory answer.
   I have maintained entries in EDPAR table thru t-code VOE4. I understand that I need to pass 'customer' and 'external partner' number in idoc to automatically determine 'internal customer' number. In which segment-fields do I pass these two values?
     I passed 'customer' or 'external partner' store number in idoc segment field E1EDKA1-PARTN but sold-to-party was not determined based on entries in EDPAR table. I tried various permutations but was not successful. Please help.
thanks in advance

Hi Koustav,
you need to maintain partner profiles in WE20
for both vendor and customer
Regards,
Santosh

Answer to Q re TLB entries in OpenSPARC T1

Similar Messages

Maybe you are looking for