|
Home > Archive > Commerce Server Solution Sites > December 2004 > caching
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
|
|
| Fadi Meqbel 2004-12-01, 2:47 am |
| I am developing a dynamic application built on top of the retail2002, I
noticed performance issues and I adjusted the AppPool and it is still slow.
I figured caching would be Ideal but I do not have enough tutorial and I
browsed alot to know what needs to be cached (i.e. the CMS contents, the
catalog tables) doing LRU caching or I should be able just adjust the LRU
cache in the web.config
| |
| Robert Bogue [MVP] 2004-12-01, 7:47 am |
| You'll have to be more specific about the performance problems you're
encountering, where they seem to be coming from, etc. for us to be able to
help recommend caching strategies.
Rob
"Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
news:2F65F5D8-3837-4AB6-A9F8-ACEB8E39FA2B@microsoft.com...
>I am developing a dynamic application built on top of the retail2002, I
> noticed performance issues and I adjusted the AppPool and it is still
> slow.
> I figured caching would be Ideal but I do not have enough tutorial and I
> browsed alot to know what needs to be cached (i.e. the CMS contents, the
> catalog tables) doing LRU caching or I should be able just adjust the LRU
> cache in the web.config
| |
| Fadi Meqbel 2004-12-01, 5:51 pm |
| They seem to be coming from whenever the application have to pull several
data from the database and from when the page has many contents. for example,
Basket.aspx takes about 10 seconds-14 seconds to load, the default page takes
it 25 seconds to start. I tried using the output cache but because I have a
menu selection on the left hand side, if you select anything the same page
will come out for the rest of the items that u will select.
Also if i look at the performance monitor whenever a process of a page
happens, it spikes it to 100 page/sec and the %processor time
Thanks,
Fadi
"Robert Bogue [MVP]" wrote:
> You'll have to be more specific about the performance problems you're
> encountering, where they seem to be coming from, etc. for us to be able to
> help recommend caching strategies.
>
> Rob
>
> "Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
> news:2F65F5D8-3837-4AB6-A9F8-ACEB8E39FA2B@microsoft.com...
>
>
>
| |
| Robert Bogue [MVP] 2004-12-02, 8:48 pm |
| If the processor time is spiking then it's probably a code issue with your
code that caching may not be able to fix. We run 250 pages per
second/processor/server on a .NET based eCommerce site that we built. That
is with a full catalog index so caching is completely ineffective.
Do you think that you're posting back to the same page repeatedly?
Rob
"Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
news:2BAEEEB6-CF36-47BB-A337-0228A97BE9BA@microsoft.com...[vbcol=seagreen]
> They seem to be coming from whenever the application have to pull several
> data from the database and from when the page has many contents. for
> example,
> Basket.aspx takes about 10 seconds-14 seconds to load, the default page
> takes
> it 25 seconds to start. I tried using the output cache but because I have
> a
> menu selection on the left hand side, if you select anything the same page
> will come out for the rest of the items that u will select.
>
> Also if i look at the performance monitor whenever a process of a page
> happens, it spikes it to 100 page/sec and the %processor time
>
> Thanks,
> Fadi
>
>
> "Robert Bogue [MVP]" wrote:
>
| |
| Fadi Meqbel 2004-12-03, 5:52 pm |
| In some of the classes I have implemented, I do have few methods that each
one of them posts back. What do you think the solution is in this case? let
me know if you want me to email you an example.
Thanks,
Fadi
"Robert Bogue [MVP]" wrote:
> If the processor time is spiking then it's probably a code issue with your
> code that caching may not be able to fix. We run 250 pages per
> second/processor/server on a .NET based eCommerce site that we built. That
> is with a full catalog index so caching is completely ineffective.
>
> Do you think that you're posting back to the same page repeatedly?
>
> Rob
>
> "Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
> news:2BAEEEB6-CF36-47BB-A337-0228A97BE9BA@microsoft.com...
>
>
>
| |
| Robert Bogue [MVP] 2004-12-03, 8:46 pm |
| I think you need to determine the root cause of your performance issue. My
suggestion is commenting out components until you determine which one is
causing the performance problem.
"Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
news:C705BD3D-4B75-4A06-A37B-1D7517415A0C@microsoft.com...[vbcol=seagreen]
> In some of the classes I have implemented, I do have few methods that each
> one of them posts back. What do you think the solution is in this case?
> let
> me know if you want me to email you an example.
>
> Thanks,
> Fadi
>
> "Robert Bogue [MVP]" wrote:
>
| |
| Fadi Meqbel 2004-12-05, 2:46 am |
| Rob,
I elimnated the repeated postbacks and this definitely improved the
performance,
However when I check the performance monitor i still see the Processer time
is high and the page/sec went down. Can you please tell me if there is
something I have to do to the IIS 6.0 to fix that problem or what are the
factors that I need to check so I can have a better processor Time.
Thanks,
Fadi
"Robert Bogue [MVP]" wrote:
> I think you need to determine the root cause of your performance issue. My
> suggestion is commenting out components until you determine which one is
> causing the performance problem.
>
> "Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
> news:C705BD3D-4B75-4A06-A37B-1D7517415A0C@microsoft.com...
>
>
>
| |
| Robert Bogue [MVP] 2004-12-07, 7:46 am |
| Fadi:
I think fundamentally you need to determine why the processor is spiking...
However, I think that you need to realize that a processing spike isn't a
problem. It indicates the system is working. It's only a problem when it's
sustained and it creates a scalability problem.
What is your target number of simultaneous users? How long can a page take
when at that number of users?
Rob
"Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
news:403F07A9-7231-4958-83D7-D66643DC04C2@microsoft.com...[vbcol=seagreen]
> Rob,
> I elimnated the repeated postbacks and this definitely improved the
> performance,
> However when I check the performance monitor i still see the Processer
> time
> is high and the page/sec went down. Can you please tell me if there is
> something I have to do to the IIS 6.0 to fix that problem or what are the
> factors that I need to check so I can have a better processor Time.
>
> Thanks,
> Fadi
>
> "Robert Bogue [MVP]" wrote:
>
| |
| Fadi Meqbel 2004-12-07, 5:49 pm |
| My target users is 25,000. I have not done any stress testing. Any stress
testing tool you recommend me to use.
"Robert Bogue [MVP]" wrote:
> Fadi:
>
> I think fundamentally you need to determine why the processor is spiking...
> However, I think that you need to realize that a processing spike isn't a
> problem. It indicates the system is working. It's only a problem when it's
> sustained and it creates a scalability problem.
>
> What is your target number of simultaneous users? How long can a page take
> when at that number of users?
>
> Rob
>
> "Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
> news:403F07A9-7231-4958-83D7-D66643DC04C2@microsoft.com...
>
>
>
| |
| Robert Bogue [MVP] 2004-12-08, 5:51 pm |
| 25,000 simultaneous users is way too many. That may be the total number of
users but certainly not concurrent ... a few hundred concurrent users is a
pretty high number for most sites.
As for a stress testing tool, we index our sites for a full text index to
stress test it. It's amazing the kind of load you can put on a box with an
indexer.
Rob
"Fadi Meqbel" <FadiMeqbel@discussions.microsoft.com> wrote in message
news:0B66B115-861B-4738-B953-25F48295602B@microsoft.com...[vbcol=seagreen]
> My target users is 25,000. I have not done any stress testing. Any stress
> testing tool you recommend me to use.
>
> "Robert Bogue [MVP]" wrote:
>
|
|
|
|
|