Level 2 Cache 3MB vs 6MB

Hi
Yes, I've already googled and searched this forum, plus others, and have yet to find a conclusive answer.
How much difference does 6MB of L2 Cache over 3MB make?
As you've guessed, I'm undecided between the 2.4GHz model and the 2.66GHz one.
I'm based in the UK and I've worked out the price difference for the setup I want.
MacBook Pro 2.4GHz
4GB RAM (bought from 3rd party)
250 GB HDD (7200 rpm)
= £1,294
MacBook Pro 2.66GHz
4GB RAM
250 GB HDD (7200 rpm)
= £1,483
Does an extra 3MB Cache justify spending another £189? Please take into account I'm not concerned about the extra VRAM on the GPU.
Thanks
gtds

Here's a Xbench, memory test comparison between my, returned, mb 2.4 (3MB), 4GB ram and my mbp 2.66 (6MB), 4GB ram, both unibodies running 10.5.6:
Note the STREAM (all about memory bandwidth) marks -
Macbook:
Memory Test 175.69
System 196.44
Allocate 254.23 933.60 Kalloc/sec
Fill 164.70 8007.99 MB/sec
Copy 189.87 3921.74 MB/sec
Stream 158.91
Copy 154.36 3188.22 MB/sec
Scale 148.57 3069.31 MB/sec
Add 169.70 3615.00 MB/sec
Triad 164.75 3524.50 MB/sec
Memory Test 181.43
System 207.51
Allocate 236.94 870.13 Kalloc/sec
Fill 186.60 9072.78 MB/sec
Copy 205.01 4234.43 MB/sec
Stream 161.18
Copy 149.43 3086.39 MB/sec
Scale 154.32 3188.22 MB/sec
Add 172.30 3670.29 MB/sec
Triad 171.20 3662.48 MB/sec
So, as it would would appear it doesn't mean much? However, when a application needs to maintain sustained memory bandwidth such as in some large, framed, vector math processing scenario's then computation time adds up and relative time will be saved with the larger cache, overall.
See this paper on Stream bench-marking: (Note in the article that other variables come into play such as 32-bit vs 64-bit vs shared ram, how an application is written, etc ...)
http://www.cs.virginia.edu/stream/ref.html
In my case I spoke with the company engineer that designed the application, I use, and got a recommendation for key considerations on hardware choice.
However, I exchanged my MB because, it was mis-behaving, and chose the mbp because I hear tell that Snow Leopard may off-load processes to the 'GT. As well, the extra cache would be of benefit for cutting down the time needed for running large simulations and processing routing and placement algorithms in the applications I run. Otherwise, the 2.4 MB is so close and in some case's faster in the benchmark(s), as well as equally performing when it came to memory bandwidth.

Similar Messages

Second Level Caching

I am new to flex. I created one app which has some views. Which calls a web service to get data from my database. When I run this app, as its waiting for the data, a timer icon comes up indicating its waiting. I also implemented a Gantt Chart which was written for the most part with Actionscript. I noticed it never gives me that timer and if the data is not loaded, it just does not load it. My real question is what settings did the gantt chart set so it no longer waits.
Here is the issue. I have class A which contains a list of Class B objects. If in my app, I instantiate a new Class B object and persist it, I noticed that when I reload the app the new Class B object is no longer listed in my Class A object. I did some reading and learned this had something to do with Second Level Cache. Because I updated the Class B, the Class A was not updated and so when I reloaded the App, the App just got what was in Cache which was the old Class A object containing the old list of Class B objects. Well, I turned off Second Level Cache and walla, my problem went away. Cool. Oh, wait, when I loaded the Gantt Chart, it no longer would load the data. It didnt wait for it. So, when I turned the Second Level Cache back on, now my Gantt Chart loaded the data on the second refresh because now the data was in L2 Cache. So, now I am wondering how to solve my problem. One idea was to turn off or no whatever is turned off or on in my Gantt chart so it does wait for the data. But I didnt write the Gantt Chart code and I am too new to know how to do this. Would Someone help me with this?

No, it was not the backend. And I still do not have a good answer. I ended up turning off 2nd Level Caching so my backend updates were available instantly. However, as stated before, my gantt Chart did not wait for the data to be available and so when it rendered the chart, it rendered it with no data. Since the chart extended UIComponent and updateDisplayList which was what was being called to render the chart to begin with, I added a public function which called updateDisplayList and then added a event on the callresult so when the data was done loading, I called the new function which called the updateDisplayList which rendered the chart a second time but this time with the data. Its not graceful because it actually renders the chart twice at startup. But it did get the job done. Maybe one day when I am better at Action Script, I will understand how to get this component to wait til the data is loaded before it renders it. But I am not there now. Of course, maybe someone will understand my question and help me. that would be nice too.

V8 Application Level Caching (Singleton Object_?

I have a requirement to have our data base lookups cached in an object which is
shared for all sessions accessing the application. Normally this would be accomplished
via a singleton object. Howerver in the spirit of bea v8, I would like to use
V8 controls to populate the caches.
I have created a control called LookupImpl and a database control called LookupDB.
LookupDB is a member variable of LookupImpl. LookupImpl has an init() method
which will retrieves data from LookupDB and caches it for later use.
If I place LookupImpl on a page flow it works fine. But if I manually instantiate
LookupImpl (x = new LookupImpl())) in a java class (my singleton class intended
to share one copy of the information among all users) , when LookupImpl tries
to run a method on its LookupDB member variable, the member variable is null and
the process blows up.
I have also read in an article
http://dev2dev.bea.com/products/wlworkshop81/articles/wlw_internals.jsp
stating that controls cannot be manually instantiated.
So my questions are:
What would be the best practice for doing sharing application level cached data?
It is true that a java class cannot instantiate controls?
Can a control be defined to be a “singleton”?
Thanks for all input
gr

I have a requirement to have our data base lookups cached in an object
which is shared for all sessions accessing the application. Normally this
would be accomplished via a singleton object. Howerver in the spirit of
bea v8, I would like to use V8 controls to populate the caches.That's an odd approach. Normally you'd have a singleton, hang a cache object
in the JNDI tree, or even have an EJB that handled this task.
Peace,
Cameron Purdy
Tangosol, Inc.
http://www.tangosol.com/coherence.jsp
Tangosol Coherence: Clustered Replicated Cache for Weblogic
"gerry rodman" <[email protected]> wrote in message
news:3fcfb309$[email protected]..
>

Coherence as a hibernate 2nd level cache

Hello,
I'm trying to use Coherence as a hibernate 2nd level cache but I'm getting the following exception when there are hundreds of concurrent DB transactions.
java.lang.IllegalStateException: Exceeded maximum number of waiting threads (Status=1)
at com.tangosol.net.cache.OverflowMap$Status.waitForAvailable(OverflowMap.java:4029)
at com.tangosol.net.cache.OverflowMap.prepareStatus(OverflowMap.java:2152)
at com.tangosol.net.cache.OverflowMap.beginKeyProcess(OverflowMap.java:1873)
at com.tangosol.net.cache.OverflowMap.getInternal(OverflowMap.java:580)
at com.tangosol.net.cache.OverflowMap.get(OverflowMap.java:330)
at com.tangosol.coherence.component.util.CacheHandler.getLease(CacheHandler.CDB:3)
at com.tangosol.coherence.component.util.CacheHandler.ensureLease(CacheHandler.CDB:4)
at com.tangosol.coherence.component.util.daemon.queueProcessor.service.grid.ReplicatedCache.updateResource(ReplicatedCache.CDB:18)
at com.tangosol.coherence.component.util.CacheHandler.put(CacheHandler.CDB:11)
at com.tangosol.coherence.component.util.CacheHandler.putAll(CacheHandler.CDB:8)
at com.tangosol.coherence.component.util.SafeNamedCache.putAll(SafeNamedCache.CDB:1)
at com.tangosol.coherence.hibernate.CoherenceCache.put(CoherenceCache.java:90)
at org.hibernate.cache.impl.bridge.BaseGeneralDataRegionAdapter.put(BaseGeneralDataRegionAdapter.java:48)
at org.hibernate.cache.UpdateTimestampsCache.preinvalidate(UpdateTimestampsCache.java:70)
Anyone have any ideas?
The timestamps cache has the default replicated cache scheme.
Thanks,
Ivelin

Hello,
Actually it's happening during read from a replicated cache too.
java.lang.IllegalStateException: Exceeded maximum number of waiting threads (Status=1)
at com.tangosol.net.cache.OverflowMap$Status.waitForAvailable(OverflowMap.java:4029)
at com.tangosol.net.cache.OverflowMap.prepareStatus(OverflowMap.java:2152)
at com.tangosol.net.cache.OverflowMap.beginKeyProcess(OverflowMap.java:1873)
at com.tangosol.net.cache.OverflowMap.getInternal(OverflowMap.java:580)
at com.tangosol.net.cache.OverflowMap.get(OverflowMap.java:330)
at com.tangosol.coherence.component.util.CacheHandler.getLease(CacheHandler.CDB:3)
at com.tangosol.coherence.component.util.CacheHandler.getCachedResource(CacheHandler.CDB:10)
at com.tangosol.coherence.component.util.CacheHandler.get(CacheHandler.CDB:1)
at com.tangosol.coherence.component.util.SafeNamedCache.get(SafeNamedCache.CDB:1)
at com.tangosol.coherence.hibernate.CoherenceCache.get(CoherenceCache.java:65)
at org.hibernate.cache.impl.bridge.BaseGeneralDataRegionAdapter.get(BaseGeneralDataRegionAdapter.java:44)
at org.hibernate.cache.StandardQueryCache.get(StandardQueryCache.java:130)
at org.hibernate.loader.Loader.getResultFromQueryCache(Loader.java:2188)
at org.hibernate.loader.Loader.listUsingQueryCache(Loader.java:2151)
at org.hibernate.loader.Loader.list(Loader.java:2121)
at org.hibernate.loader.criteria.CriteriaLoader.list(CriteriaLoader.java:118)
at org.hibernate.impl.SessionImpl.list(SessionImpl.java:1597)
at org.hibernate.impl.CriteriaImpl.list(CriteriaImpl.java:306)
at org.hibernate.impl.CriteriaImpl.uniqueResult(CriteriaImpl.java:328)
There must be a configuration option to control the number of waiting threads, right?
I guess it's not a hardcoded constant.
The statistics of the cache we are reading from are:
0.99 hit probability
550 303 gets
1469 puts
Size: 1270This is for 5 mins period.
This cache holds the configuration of the application, so there are almost no writes.
Thanks,
Ivelin

ADF - Jdev 11 - application level caching

Hi All,
Our system user SOA calls to populate UI's.
So in order to make performance faster, i am planning to store LOV'S values in application scope.
So i am looking at a solution where i can populate the lists in managed bean which is stored in application scope, when application loads.
How i can achieve this solution, can i have servlet listener. Can i get access to bean when application loads.
Regards
Praveen

Hi Frank,
I am planning to have context listener, where i can store my attributes at application level.
Instead of making frequent SOA calls for populating LOV'S, i can just get the attribute and set it.
Regards
Praveen

CS5 Photoshop crashes... Have tried everything! Please help!!!

Hi Guys! I am having big problems. I cannot use Adobe Photoshop (CS5) ....(nor Corel Paintshop Photo Pro x3 "Sorry for mentioning that on here!!! )
I am running a 1 month old Toshiba Satellite C660-1F1 with Windows 7 Home Premium 64-bit with 2GB RAM Intel Core i3-380m processors 2.53 Ghz / 3rd Level Cache 3MB, 320 GB Hard Disk with Intel Graphics Media Accelerator HD (shared memory up to 761 MB).
I have tried the full paid-for version and the trial version downloaded from the Adobe site.
Both programmes (CS5 and Corel PSPx3) installed and can be opened.
With photoshop when i open it the following message comes up:
"Photoshop has encountered a problem with the display driver, and has temporarily disabled GPU enhancements. Check the video card manufacturer's website for the latest software. GPU enhancements can be enabled in the performance panel of preferences."
I can usually perform one or two actions before the programme crashes and a message appears to say that it has stopped working. It will crash straight away if I click on the 'Help' menu and then on 'System Info'. Before it crashes when the buttonbs are still responsive ...if I go to the performance panel in preferences and scroll over 'detected video card' the discription box says:
"Your system display driver has caused Photoshop to crash. Update or reinstall the driverand check enable OpenGL drawing to retry."
When I follow these steps it just crashes as usual. I have deleted and reinstalled the drivers.I have deleted and reinstalled the programmes countless times. My drivers are all up-to-date. Both software packages claim to and seem to support 64-bit operating systems with my specs. Both say they support Intel HD Graphics. I do not know what else to do or try. I have even performed a fresh reinstallation of my windows 7 64-bit op system. I was looking to upgrade to 4 GB RAM and then 8GB RAM later but not before i can solve this problem.
Is there something simple/obvious I am missing or is it most likely faulty hardware / bad graphics card??? Please help!
Just in case it is relevant .... with Corel Paintshop Photo Pro x3 it installed and opens as usual but will crash the second you try to save something.The cursor remains normal but no buttons, tabs or menus are responsive at all. There is just a ping sound every time I click the mouse / touchpad. i must then use 'end task' in task manager to close it down, although it says the programme is running! (It does not say 'not responding').
All info / advice / suggestions welcome!!! Thanking you in advance!
John

Intel doesn't have the best OpenGL implementation. Okay, that may be being overly kind.
Your best bet, especially if you have already loaded their latest driver, is probably to deconfigure OpenGL usage by Photoshop.
First, try this:
1. Edit - Preferences - Performance
2, Click the [Advanced Settings...] button
3. Set Mode to Basic.
4. Uncheck Vertical Sync.
5. Close and reopen Photoshop.
If this still doesn't help, try this:
1. Edit - Preferences - Performance.
2. Uncheck [ ] Enable OpenGL Drawing.
3. Close and reopen Photoshop.
Good luck.
-Noel

Caching on DB level

Hi,
Our scenario is like one frequently accessed table which seldom changes is to be cached so that instead of retrieving the data from the filers each time a user requests for data it retrieves from cache for performance impprovement.
so instead of using folowing method
create table md_cube_64
storage (next . . . )
cache;
Alter table md_cube_64 cache;
what other methods are there for table caching?
is there any database level caching ?
Regards
Asif
Message was edited by:
Mohammad Asif

1) Assigning table to keep pool
2) Using different table type (not heap table but for example index organized table, single table hash cluster) to reduce necessary i/o amount to get table data
3) if table data never changes and you are reusing table data much in the same session then reading them into a pl/sql collection structure in the first run and after that using that collection instead of querying actual table
I'm sure there are other possibilities...
Gints Plivna
http://www.gplivna.eu

Less Cache

Why is there less cache in the newer models. Here the spec for the old version.
The Apple MacBook "Core 2 Duo" 2.2 13-Inch (Black - Late 2007/Santa Rosa) features a 2.2 GHz Intel "Core 2 Duo" processor (T7500), with two independent processor "cores" on a single silicon chip, a 4 MB shared "on chip" level 2 cache, an 800 MHz frontside bus,
Here the spec for the latest model. Is there a big difference , there say you getting more and more speed. But to me it look like there going backwards.
2.1GHz or 2.4GHz Intel Core 2 Duo processor with 3MB on-chip shared L2 cache running 1:1 with processor speed
800MHz frontside bus

Intel hasn't reduced the cache on the Penryn processor per se. They offer Penryn CPUs with both 3MB and 6MB caches, dependent on CPU speed, for computers using an 800MHz bus. The 2 lower speed CPUs (2.1GHz and 2.4GHz) come with 3MB while the next two steps (2.5 and 2.6GHz) come with the full 6MB cache. Heat may have been one consideration.
http://www.intel.com/products/processor_number/chart/core2duo.htm
For Apple, I suppose this may make it slightly easier to add a wee bit more distance between the MB and the MB Pro (middle and higher end ones anyway).

Need input on data cache

Hi All,
This is the situation..
Data cache = 3mb (default)
Data block size = 20kb
so number of blocks a memory segment (data cache) can hold is 157
What will happen If user is retrieving 300 blocks in one retrieval ?
I know swapping will takes place for multiple retrievals.
I know that agg cache in ASO will increase till OS say no it.
Will Essbase tries to increase data cache....?? if so how it works??
My understanding is Essbase tries to increase data cache and it may touch the wrong segment in memory to create a segment violation to crash!!
As per the modern OS, virtual memory management using pages and segments ...I think this will takes place!!
Please give your inputs!!
Thanks,
Jeeth
Edited by: Jithendra on Mar 28, 2011 7:12 PM
Edited by: Jithendra on Mar 28, 2011 7:14 PM
Edited by: Jithendra on Mar 28, 2011 7:15 PM

Hi,
I just did an initial analysis....
It is always 3mb. (never increased it)!!
XCP log says it stopped at executing the maxl!! ..like (Maxl pending)
and its an segmentation violation!!
from my understanding segmentation violation is something to do with kernal memory allocation, virtual memory and physical memory.
I cannot try the same on prod server (pls don't ask why the settings are default etc ..I'm new to the system)
I just provided an intial analysis telling that at this point of time...the highly visible point to me is data cache in buffered I/o.
The segmentation violation can also happen at total level of availability of RAM.
but in our case we are using only 1/3rd of the Total available RAM.
memory in all applications = (1/3rd of available RAM)
and its a Maxl operation...
by seeing al this i suspected the application cache.(data cache)
I posted only to gather little mor info and to have a debate to get clear understanding.
Thanks for your support and input...
Please advice
Thanks,
Jeeth
Edited by: Jithendra on Mar 29, 2011 1:37 AM

KM Scheduler Task Cache issue

We have created couple of custom schedulers in Enterprise Portal. These schedulers are worked fine in our QA and DEV environment. But when we move the same transport to Production the scheduleru2019s tasks are not showing up in KM scheduler list. But after restarting J2EE couple of times it's showing up in KM scheduler task. This is the behavior we are seeing whenever we move the scheduler to production.
Here are the some technical details about the scheduler,
We have created one Portal Development component and 10 Scheduler Tasks under this component. All the tasks are not created in a single release. Initially when we develop we used to have 2 scheduler tasks under this component, when we move this transport to production we didn't faced any issue. Even after adding 4 to 5 scheduler task we didn't had any issues. After adding more than 6 to 7 whenever we add one we are getting the same issue.
Question: 1 - Is there any limit to for adding Scheduler task in one Portal Component?
This time we have restarted our server more than 5 times but scheduler not showed up. We ask our manger to logon to the portal server who never navigated to KM screen, we ask her to navigate and she did. For her schedulers are showed up. After she navigated to the KM screen everybody is able to see the scheduler. This is not the first time it happens like this, the same situation happen for the previous release also. In previous release we ask our Technical Architect to login to see KM schedulers. After his login everybody is able to see the schedulers.
After doing all these is what I obverse is, it seems KM scheduler has some server level cache for the user and it's giving old screen whenever we are trying to login and see. After the transport, the cache is not getting updated with the new version.
We guess it might be something we need to do with cache.
Is there any idea why the system is behaving like this.

Hi,
I am facing similar issue. How did you resolve it?
Regards,
Apurva

JNDI cache in a Cluster

Lets say that we have 2 machines in a cluster, If I cache in the JNDI, which
          is accessible from both Servers,
          1. when they update the same cache then how can we put a lock so that they
          don't mess up other user's data.
          2. Can I hide this cache for the other machine in the cluster?.
          3. What is the best way to do session level caching?. For Example, If I use
          the sessionID of the Weblogic's HttpSession, It is too big to store it.
          4. What is the maximum objects can be stored in the JNDI? If I store 400-500
          objects in the JNDI , whether performance will be affected?!
          5. Do I have to rebind cache if I update the data?
          Thanks
          /selvan
          Captura Software Inc


Lets say that we have 2 machines in a cluster, If I cache in the JNDI, which
          is accessible from both Servers,
          1. when they update the same cache then how can we put a lock so that they
          don't mess up other user's data.
          2. Can I hide this cache for the other machine in the cluster?.
          3. What is the best way to do session level caching?. For Example, If I use
          the sessionID of the Weblogic's HttpSession, It is too big to store it.
          4. What is the maximum objects can be stored in the JNDI? If I store 400-500
          objects in the JNDI , whether performance will be affected?!
          5. Do I have to rebind cache if I update the data?
          Thanks
          /selvan
          Captura Software Inc

Cleaning the cache

How do I clean my cache files? is that something we do as a rutine?
Is the program MacKeeper safe to clean the cache files?
Am I suppost to keep my firewall enable? this is my first Mac and as happy as I am I would like to keep it in great shape!! Thank you so much

Use your machine normally.
From the X Lab's Maintaining Mac OS X
System and User cache cleaning are troubleshooting steps, not part of regular maintenance.
Caches are used to improve the performance of both the operating system and applications. Removing System and User cache files for other than troubleshooting purposes defeats the purpose of cache files, which will be rebuilt automatically at your next restart or login, adding time to those processes.
Potential side effects of System-level cache cleaning are discussed in our "Side effects of System cache cleaning" FAQ.
Accordingly, cache cleaning can be an important troubleshooting step, such as when used in our"Resolving Disk, Permission, and Cache Corruption" FAQ, but it is not regular maintenance.

Caching in obiee 11g

My requirment has an executive dashboard which has 30 reports on it.This dashboard can be accessed by 1000 users.
My requirment here is when the users open the dashbard page it has to hit the cache and the reports in the dashboard should appear immediatley without taking time.The cache also should be updated daily.
My first queation here is:
1) we dont have any prompts or data security .The report data is same whoever opens the report.So is it fine if we use ibots to clear the cache daily and run the reports for a user and if anyother user opens the dashboard since it is the same reports will hit hit the cache or not?
2) when i tried its not hitting the cache when a different user opens it.Is there a solution where there is a group level cache and users in that group hits the same cache entry.

1. Yep, fine to use IBots to clear / re-seed the cache.
2. I would check - can you validate if caching is working at all? You mentioned there are no filters or any type of user security turned on, so I don't understand why the queries wouldn't be cachine. Probably best to turn the logging level up to 5 for a couple of users and ask them to hit the dashboard, then take a look at the logs and see if things are getting cached (and if not, hopefully why). It seems like in a simple environment with no filters and/or security that it should be hitting the cache (if caching is indeed properly working).
One other quick check for caching - just go in to the BI admin tool and see if cache entries are showing up there.
Hope this helps,
Scott

Setting the size of a shared entity cache

Hi,
I am using WLS 7.1 and have a few entity beans that use a shared
application-level cache. I would like to set the size of this cache using
the admin console.
As this property is specified in the weblogic-application.xml file, I
attempted to access the page in the Admin console Deployment Descriptor
Editor that lets you modify this deployment descriptor.
However, clicking on the Weblogic-application node in the deployment
descriptor editor gives me an Error 404--Not Found message.
Any ideas why this is happening?
Thanks in advance.
Santosh
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.754 / Virus Database: 504 - Release Date: 6/09/2004

Hi,
I am using WLS 7.1 and have a few entity beans that use a shared
application-level cache. I would like to set the size of this cache using
the admin console.
As this property is specified in the weblogic-application.xml file, I
attempted to access the page in the Admin console Deployment Descriptor
Editor that lets you modify this deployment descriptor.
However, clicking on the Weblogic-application node in the deployment
descriptor editor gives me an Error 404--Not Found message.
Any ideas why this is happening?
Thanks in advance.
Santosh
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.754 / Virus Database: 504 - Release Date: 6/09/2004

Caching in Web Logic

Hi,
I wanted to find out if the BEA WebLogic Server does any HTTP-level caching at all with respect to both entire documents (eg: gif images, static HTML files) as well as partial documents (eg: JSP fragments).
Any inputs would be much appreciated,
Thanks,
Rishi.

Hi, I also wanted to know if BEA Web Logic does any kind of Java Server Faces (JSF) caching.
Thanks,
Rishi.

Level 2 Cache 3MB vs 6MB

Similar Messages

Maybe you are looking for