Items cached and keep pool

Hello
Could someone tell me how to check below:

(1) what objects are cached?

(2) what are the objects in the pool to keep?

This is my environment:

DB: Oracle 10g 10.2.0.3.0
OPERATING SYSTEM: HP - UX B.11.31

See you soon
Steve

SELECT   ds.BUFFER_POOL,
         Substr(do.object_name,1,9) object_name,
         ds.blocks                  object_blocks,
         Count(* )                  cached_blocks
FROM     dba_objects do,
         dba_segments ds,
         v$bh v
WHERE    do.data_object_id = v.objd
         AND do.owner = ds.owner (+)
         AND do.object_name = ds.segment_name (+)
         AND do.object_type = ds.segment_type (+)
         AND ds.BUFFER_POOL IN ('KEEP','RECYCLE')
GROUP BY ds.BUFFER_POOL,
         do.object_name,
         ds.blocks
ORDER BY do.object_name,
         ds.BUFFER_POOL;

Tags: Database

Similar Questions

Acer R7 - 571 level disk cache to large SSD to the boot drive and keep the traditional HARD drive.

Is it possible to replace the drive 24 GB of cache in the R7-571 with a larger SSD (say 128 GB and 256Go) to use as the startup disk and keep the traditional disc for storage on the other?

I saw the references to individuals 'redevelopment' of SSD, but normally they keep the 24GB disk cache and upgrade their traditional 500 GB drive?
If it is possible to replace the disk cache of 24GO, it seems silly to keep it and throw a perfectly good car. Seems would not be much point in having the disk cache if your main player is an SSD.

Yes, I put a 128 GB Plextor M5M in my mSATA port. Make sure that you first create a USB recovery, then open the R7 and replace 24GO with your new mSATA disk drive.

Make sure that your new mSATA, not ordinary SATA SSD!

You can see a R7 being open and where everything is here: http://www.youtube.com/watch?v=XLWdiLxLoyI

After clearing the cache and cookies, my combo even keeps returning when I connect to Facebook.

Empty cache and delete cookies, even my combo guard returning whenever I connect to Facebook. I need to know how to delete the history of connection.

This data is called form history
Options/Preferences > privacy > clear recent history > everything + history

Difference between keep pool Vs recycle pool vs default

Good morning experts;
I have need of a few differences between keep pool recycle Vs vs default pool.
How it is differ from each other.
Thanks in advance...

8f953842-815B-4D8C-833d-f2a3dd51e602 wrote:

Thanks for your reply Marg.

If I pin an object in the shared pool, the whole object (all blocks) is in the buffer pool.

but you say depending on the query plan,

Oracle will place only parts of objects in the cache of the buffers at any time.

Example:

> To pin a table

SQL > alter table emp storage (USER_TABLES) keep;

This table with 1 million record and it contains "n" columns.

Consider that I need o/p of columns name, emp_id, salary only.

i.e. which receive salaries over $ 8000.

----------------------

Oracle will show required o/p. According to your explanation, I can't guess...

Question: How oracle will place only parts of objects in the buffer instead of the object in its entirety?

-------------------------------

Please explain a little more.

Oracle uses blocks. The lines are in blocks. When you request a column in a row, Oracle must get the block. When you ask a couple of columns from several lines, Oracle must obtain several blocks. Oracle makes copies of blocks. Oracle must manage maybe a lot of people to access the rows in the same or different blocks. Everyone must have the block appear as if it did at the start of the transaction.

Oracle has several ways to get the blocks. It can look in the SGA, if appropriate is not it be read from the disk, or it may decide to read the many blocks both a disc, or it could even decide to read just as much as he can in PGA a user, maybe also will cancel in any of these ways to make a consistent copy of reading for the user.

So when you look at the statistics for a session, you might see gets sequential or gets scattered. The first is often the index access, then a single block is obtained from anywhere and placed in the SGA. The latter is often the analysis, and the blocks are scattered, they don't necessarily have to be obtained in order. Remember, an Oracle block can be a number of operating system blocks and a multi [-oracle]-read block can be a lot of data.

So with all these blocks to enter the LMS, it must decide what remains and what is happening. It uses an algorithm (LRU) least recently used to eject blocks and can read blocks in the middle or the end of the list, according to. This is why the default buffer pool works so well, whatever it is continuously available in the grand scheme of things will stay warm and stay there. SGA was much smaller, it's much easier to have things isn't everything in fact so hot get ejected and written, only to be read in a short time after, for the pools of spare would allow these places to be kept, or recycled, as arbitrarily defined.

So do blocks like the passage of objects in the SGA. There are usually several copies of the blocks.

Find people to KEEP pool
Hello

recently, a DBA I know has asked to make a list of candidates to KEEP pool in the buffer cache. He is not familiar with the application. Is there a way to do automated, or is - something that only a person familiar with the business logic can do manually?

I googled a little on the subject, but haven't found much, with the exception of Mr. Burleson scripts (which has a history silly advice so I don't him trust) theoretical or general discussions (like the thread on AskTom). I am interested in someone who has made (or learned why this cannot be done) is sharing his experience.

Thanks in advance!

Best regards
Nikolai
Nikolay Savvinov wrote:

recently, a DBA I know has asked to make a list of candidates to KEEP pool in the buffer cache. He is not familiar with the application. Is there a way to do automated, or is - something that only a person familiar with the business logic can do manually?

I googled a little on the subject, but haven't found much, with the exception of Mr. Burleson scripts (which has a history silly advice so I don't him trust) theoretical or general discussions (like the thread on AskTom). I am interested in someone who has made (or learned why this cannot be done) is sharing his experience.

Basically, I don't thnk that you will be able to get lots of benefits on the application, except if you know demand well enough to say:

This object is large enough, so randomly accessible we continue rinsing of the memory blocks, and if we kept in memory, we would avoid random disk N bed / hour.

Next, you will need to set aside cache much more as there are has blocks in the object (according to the requirements of update and the coherence of reading) and a recycle cache. And you still don't know how much extra I/o other objects would do as a side effect.

You've probably seen the articles that I wrote on the anomalies with Dungeon - here are some links if not:

http://jonathanlewis.WordPress.com/2011/03/14/buffer-States/
http://jonathanlewis.WordPress.com/2010/03/20/not-keeping/
http://jonathanlewis.WordPress.com/2011/03/16/buffer-flush/
http://jonathanlewis.WordPress.com/2011/09/26/upgrade-argh/

Concerning
Jonathan Lewis
http://jonathanlewis.WordPress.com
Author: core Oracle

What is the difference between the CACHE and the PIN code?
What is the difference between the CACHE and the PIN code?
Cached, it's when you specify CACHE as part of a CREATE or ALTER an object to say Oracle when the blocks are retrieved this object these blocks should be put to the most recently used end of the LRU (least recently used) list in the buffer cache when a full table scan. Under normal circumstances, when the blocks are retrieved in memory, in that they are placed least recently used end of the LRU list. This means that the data that was most recently viewed typically will remain available in the cache buffers for awhile. Therefore, subsequent runs of the same query should find these blocks already in the buffer cache and not to recover disk again. When you use the CACHE, you say you want to put the blocks retrieved at the end of the list, which means that they will be the first blocks "aged" off the buffer cache when more space is needed.
A few reasons to use the CACHE are:
(1) for tables of small (only a few blocks that must be read to retrieve the entire table).
(2) when you do a select statement returns a huge amount of data, and you do not necessarily have if wait for the query to run any time soon. In both cases, you give Oracle an opportunity to eliminate the blocks quickly to make room for more 'necessary' blocks instead of having to follow in the LRU list.

Pinning is when you want to keep the objects in memory and to prevent them from being aged by the normal mechanism of the LRU. You use the INVALIDHTOMEH to "pin" PL/SQL code in memory, so that users are not experiencing intermittent slowdowns when code gets years out of the shared pool and must be recharged.

The bottom line is that caching can blocks be moved quickly while pinning retains the information in memory to avoid the re-loads.

How to clear the cache and cookies?

I have a Satellite L350 lent by a friend - usually, I work on a Mac and am familiar with computers.

Is it necessary to empty the cache and cookies from time to time, and if so, how do we?

Hey Buddy,

> Is it necessary to clear cache and cookies from time to time
Not normally but I do from time to time because a lot of temporary internet files and cookies is stored so I remove to keep my own system ;)

But it s depending on the browser you are using any Web browser registers its own files internet for what browser you used?
If you use Internet Explorer, you can delete the files in the control panel > network and internet > internet options.

If you need more instructions, just after again. ;)

Will not empty the messages in my Outbox and keep messages to re-mailed every minute.

Outbox will not be empty

Will not empty the messages in my Outbox and keep messages to re-mailed every minute.

You don't have to tell us what e-mail program use. If it's Outlook Express, then this is the answer.

You have apparent dbx file corruption.

Spend most of your messages out of the Inbox and then create new folders to send and sent items box after having moved the messages you want to save to a local folder that you create.

Tools | Options | Maintenance | Store folder will reveal the location of your Outlook Express files. Note the location and navigate on it in Explorer Windows or, copy and paste in start | Run.

In Windows XP, Win2K & Win2K3 the OE user files (DBX and WAB) are by default marked as hidden. To view these files in Windows Explorer, you must enable Show hidden files and folders under start | Control Panel | Folder Options icon | Opinion, or in Windows Explorer. Tools | Folder options | View.

With OE closed, find the DBX files for the items in the Outbox and sent and delete them. New ones will be created automatically when you open OE.

After you're done, followed by compacting your folders manually while working * off * and do it often.

Click Outlook Express at the top of the the folder tree so no folders are open. Then: File | Work offline (or double-click on work online in the status bar). File | Folder | Compact all folders. Don't touch anything until the compacting is completed.

General precautions for Outlook Express:

Do not archive mail in the receipt or sent items box. Create your own user-defined folders and move messages you want to put in them. Empty the deleted items folder daily. Although the dbx files have a theoretical capacity of 2 GB, I recommend all a 300 MB max for less risk of corruption.

Information on the maximum size of the .dbx files that are used by Outlook Express:
http://support.Microsoft.com/?kbid=903095

Disable analysis in your e-mail anti-virus program. It is a redundant layer of protection that devours the CPUs, slows down sending and receiving and causes a multitude of problems such as time-outs, account setting changes and has even been responsible for the loss of messages. Your up-to-date A / V program will continue to protect you sufficiently. For more information, see:
http://www.oehelp.com/OETips.aspx#3

Why you don't need your anti-virus to scan your email
http://thundercloud.NET/infoave/tutorials/email-scanning/index.htm

Note that for some AV programs, it may be necessary to uninstall the program and reinstall in custom Mode and uncheck analysis when the option is the result of e-mail messages.

Compact often as specified above.

And backup often.

Outlook Express Quick Backup (OEQB Freeware)
http://www.oehelp.com/OEBackup/default.aspx

Host memory consumption is higher than the comments its use (incl. Cached) and overhead?

I'm under ESXi 5.1. I don't have any custom reserve configured for memory. I'm running in CVS 'Sandy Bridge' mode.
I noticed earlier than expected host memory consumed on many of my guests. This example comments running Server R2 2012 with the latest version of VMware tools running.
VMware reports
The guest operating system reports
My understanding is that the "Standby" and "Modified" on the client contains the ' cache ', and the 'Cached' memory prompt is the typical response to this question that I see when people seek only to the memory of 'In Use'. In my case, I'm factoring in the memory cache, my guest shows 3805 MB of real memory, yet my host shows 7.68 GB of memory consumed host, with only 68 MB of general consumption. According to my calculations, the virtual machine should be only consume about 4.28 GB of memory.
Post edited by: bradduane Replace the values of text with screenshots.

I think I found my answer in the guide 'Understanding memory resources management' here: http://www.vmware.com/files/pdf/mem_mgmt_perf_vsphere5.pdf

Section "when to retrieve memory host" page 15:

«In the high State, the use of global memory machine virtual comments is smaller than the size of the host memory.» Whether or not the host memory is overcommitted, the hypervisor will not recover memory via hot-air balloon or permutation. (It is true that when the limit memory VM is not set).

If memory free host down to the low threshold, the hypervisor begins to find the memory using the balloon. Balloon happens before free memory has actually reached the soft threshold because it takes time for the balloon pilot to allocate and PIN memory physics reviews. Usually, the balloon pilot is able to retrieve the memory in a timely order to keep free memory in the host above the soft line. »

I checked esxtop, pressing "m" for the display of the memory, and I see that hosts VMKMEM shows 'State', so the only memory of way either currently retrieved is by sharing the page. Only when I start to run out of memory of the host, that this State becomes 'soft' or 'hard', balloon or Exchange will be held in order to free the "free" memory of comments. ESXTOP show my 'minfree"than 940 MB, while it shows that I physically 6960MB of free memory.

I have to empty my legacy disk cache and cache before made a big media? Or my computer will be made faster if I do not clear the cache?

Currently I went quite a few large compositions, and I wondered just from a perspective of rendering speed, is it better to flush caches? Or my comps will be rendered more quickly if I didn't clear the caches?
Logic suggests it will, my comp info in the cache and so it can be more quick to leave them, but everything also I understand a rendering that passes more quickly if she was a pleasure to work with own cache?
I'm also ready for someone to say that it makes absolutely no difference at all!
Very much appreciated.
Chris

The whole point of caches is to make the treatment more quickly so that the items do not need to be rerendered.

The only reason to clear the cache, it's a bad result was mistakenly cached.

SSD for the Scratch, Cache and swap file?

Certainly, startup disk and RAID3 9-drive (7K 3000) the map of Areca-16. HDDs but... I was wondering, as I refine my construction, on the use of SSDS Strath, Cache and the pagefile.
It seems to me that constant random access to these files (small for most), could be somewhat competing for overall throughput (they have been configured on the RAID3 alongside media and project). In an ideal configuration, it would be a practical or advantageous to have 2 x SATA3 SSDs RAID0 for Cache, the Scratch and the pagefile? (and whatever else reconstructible temporary structures). In addition, in order to move the traffic far form the Areaca PCIe lanes, would it be more effective to RAID0 on SATA3s to the mobo? What would be the ideal size for these disks (min and max)? 2 x 60GB would do? Or 2x120GB? More?
I have also heard that, because of the nature of the scratch and cache, it would be better to have them on two separate drives? Or is it just an urban legend?

Frederic,

Bays of drive and the system configuration questions are good, but if I or someone had to answer us would be influenced by our own test files, projects, we have worked on, controller CPU, RAM, the card (or the lack of) and much more.

Given that what you're building has several similarities to mine, I'll share some of my own conclusions. To the topic specific answers like "how many disks in a RAID before it's inconsequential", it depends on so many things you should probably stop asking questions like this, because it may depend on certain things - CODEC, number of layers, transitions, filters, etc..

Things I'll comment based on similarities between the system you are creating and mine (X 58, 6-core CPU, 24 GB of RAM, controller Areca w / large cache (2 GB or more)):

1 move the pagefile seems really make a big difference on my system. So therefore, I keep in my table of OS/programs.

2. faster I ran used two large paintings with the projects, the media and all the files "scratch" on a table and "hide media" and released on the second. While the difference was not huge, the fastest result for PPBM5 comparative analysis was achieved but put 'media cache DB' on my OS/programs, which use SSDS.

So, how do proceed you?

1. load your operating system and programs on your disk boot or start table.

2. set up your Areca and drives in RAID 0, load some concrete projects similar to what you'll be using this system for, and test. Run PPBM5 too, it's a very useful common denominator that can be used to modify your own system and do it before / after comparison of the changes you are trying. IMO, a drive of 9 or 10 RAID 3 or 5 (or 6 besides) works well enough for everything. [Readers here who have a big RAID 0 out of their motherboard, I don't know if this is true for you or not, my test was done with all the great paintings hanging on a controller from Areca]

3. for answering your question about "Hard drives how much you" on your own system and with your own projects, simply repeat step #2 using a small RAID 0 array. Latest Areca SAS RAID controllers so fast, that I would say that you can count on very similar performance with a RAID 0 array with one less than road when you build the dos system to desired PARITY RAID; with the exception of RAID 6, of course, who would actually perform '2 disks' more small.

4 configure your final bar selected with parity, load your projects and you're done!

Another way, you can go forward? (Less time, less optimized, version case potentially filled with several discs, heat and energy than the 'optimal' generation) Note that I tested with a few large RAID 0 arrays that will be in my final build, and I use these disks for the storage of off-site backup.

1 build your big array with PARITY from the outset.

2. start to use the system and if it works well for you, you are finished.

As for RAID 3, 5, 30 and more, I use my table of data for editing more video, so go with RAID 5 was a simple choice for me. That being said, it seems to me that in my system (using a large cache ARECA controller and drives of 1 TB 7200 32 MB cache) a player "a less" RAID 0 seems to work only about the same as a RAID 5, so I don't think that the ineffectiveness of the RAID 5 or RAID 3 for large files (aka video editing) may be noticeable.

Kind regards

Jim

Play (and keeping to synce) multiple video streams?
Hi all

I wonder if the following is possible with Flash Media Server:

We store technical workshops at conferences, and we are looking for a way for people to watch them later with a very good user experience. Currently, we give presenters Camtasia to record their actions on the screen, which then encode us into a single video file. We also use a video camera to record the presenter of the size upwards, which encode us in a second video file. So now we have two video files we would like to keep synchronized to display to the user at the same time.

Is it possible to use Flash Media Server as well as a custom player to do this?

Basically, I don't think it would be too difficult to create a single set of playback controls that sent the time codes of reading request again to two separate video streams. The problem I see would be what would happen if one of the staggered creeks and had to pause the buffer while the other current continued to advance? In this case the two streams would become out of sync. Our current workaround is to make us two streams into a single video file, which is very demanding work and does give us no flexibility on the road, but that's all we could do to ensure that they remain synchronized when the user skipped around or that he stopped for buffering.

Any thoughts?

Thank you
Brian
Unfortunately, there is no really accurate way to keep streams synchronized with FMS.

Since there is no timecode to speak of, there is nothing to synchronize at other times netstream. The only way I know for ns synchronization is break time and restart the stream as they get out of sync, but as a FMS stream does not remain in the cache and buffer empties as soon as you interrupt the flow, you have to refresh whenever it is resumed. Even with a buffer of zero, ago yet the time it takes for the first few KB of data to the client, so there is always some default alignment.

Is FMS? I wonder if using an application server http to simulate the flow can be a viable solution (example http://flashcomguru.com/index.cfm/2005/11/2/Streaming-flv-video-via-PHP-take-two ). With progressive video, the data is cached and you have access to bytesloaded/bytestotal and time netstream, so I think that you would be able to achieve synchronization you want.

Transactions involving multiple caches and a database
Hi all

I'm curious to know if the following is possible thanks to the support of transaction consistency.

I have the need to write data in two caches and a database of my Tomcat container, and the entire operation must be atomic.

Example:

Write in the database (it is NOT via a dumps)
Write caching 1 (uses continuous writing to database)
Write to the cache 2 (uses continuous writing to database)

If all fails, the entire transaction needs to restore. This feels like an XA transaction, but looks like it doesn't work as I expect because the cache must be the last resource, but I have two caches. The order of operations is not important.

Thank you

Rob
When you use write-through is here (as I know) no way to get a fully transactional behavior (assuming you have more than one cache node) since each node is responsible for the persistence of its own data items (they each will use a separate connection to the database).

If you are using as a reason ' hiding beside of "this can be done using XA transaction. As long as the two caches belong to the same cache service they count as a resource unique "last"...

/ Magnus

How to delete pictures on the phone and keep about iCloud

Pls will have this problem and a little confused. How can I remove photos from my iPhone 6 and keep these photos on my phone X someone help pls simple format pls

Hello

When you use iCloud photo library, your iCloud storage can work to free up storage on your iPhone by keeping versions space saving on your iPhone and the full-size originals in iCloud. However, unable to save pictures to iCloud instead of have them on your iPhone at all.

Another option is to import all existing photos from your iPhone to computer, and then delete the originals of your iPhone, freeing up a space for new.

The following support article explains the difference between the storage (on your iPhone) device and iCloud:
- On the storage on your device and in iCloud - Apple Support
You can also find these resources useful:

How can I erase history navigation, cache and cookies

How can I erase history navigation, cache and cookies?

You'd which using your Safari browser. Perhaps under the menu Safari or in the privacy of the Safari preferences option.

Items cached and keep pool

Similar Questions

Maybe you are looking for