Also, if we are talking about the use of remote cache in JCS, 1) There is a separate JVM running the remote cache server, but is missing in your diagram 2) In your diagram, I think application at the lower right hand corner of each page should be C, not B. Otherwise, you would have both a static data store called B and an application also called B. 3) All A, B and C are considered local caches. Only the one in (1) is considered the remote cache.
Hanson On 5/11/05, Hanson Char <[EMAIL PROTECTED]> wrote: > Hari, > > I am starting to think we are talking about different things here. We > should clarify the terminology. > > 1) Are you referring to the use of a) remote cache server or b) > lateral cache in JCS ? > 2) What does it mean by "Remote Cache Synchronization" ? Is it an > extra step involved in the use of lateral cache ? For AFAIK, there is > no such thing in using the remote cache server. > > H > > On 5/11/05, Hariprasath Manivannan <[EMAIL PROTECTED]> wrote: > > Hi Hanson, > > > > I have attached a word doc depicting the problem. > > > > > > > > Regards, > > Hari > > > > > > On 5/11/05, Hanson Char <[EMAIL PROTECTED]> wrote: > > > What do you mean by "synchronization" ? Do you mean when B tries to > > > update it's local cache, or C tries to update it's local cache, or > > > both ? In all cases, A at most will only be affected by a bunch of > > > cache item delete requests. > > > > > > Hanson > > > > > > On 5/11/05, Hariprasath Manivannan <[EMAIL PROTECTED] > wrote: > > > > Hi Hanson, > > > > > > > > Thanks for reply. > > > > > > > > To clarify things about my query : The client A is again another J2EE > > > > application. I referred to client because it uses the data from my > > datastore > > > > application B. > > > > > > > > The scenario that I am worried is: If a new client application C wants > > to > > > > use data from my store(using cache), then a cache will be setup at > > client C > > > > and at the contents will also be stored in the cache at "B"(My data > > store > > > > application). When synchronization happens, the cache at A will updated > > to > > > > have the stuff that "C" requested. > > > > > > > > Regards, > > > > Hari > > > > > > > > > > > > On 5/11/05, Hanson Char < [EMAIL PROTECTED]> wrote: > > > > > When you say "client" side A and "server" side B, do you mean A is a > > > > > client of server B, both using JCS (in remote cache mode) ? > > > > > > > > > > If so, when B synchronizes with the remote cache, the corresponding > > > > > cache items are flushed from A, so the client will never hold data not > > > > > relevant to it. > > > > > > > > > > Hanson Char > > > > > > > > > > On 5/11/05, Hariprasath Manivannan < [EMAIL PROTECTED]> wrote: > > > > > > Hi, > > > > > > > > > > > > I am planning to use the remote cache feature of JCS in my J2EE > > based > > > > > > application. Basically my core application is a static data-store. > > Since > > > > the > > > > > > data is static, I decided to use the remote cache feature as this > > can > > > > reduce > > > > > > the RMI calls and network traffic. > > > > > > > > > > > > Now if I have a cache and both server side and client side, then I > > may > > > > face > > > > > > issue during the remote caches synchronization. The server side > > cache > > > > will > > > > > > be a global cache containing all the cached objects. This is fine. > > But > > > > once > > > > > > synchronization is complete, the client cache's will become similar > > in > > > > > > content to the server side cache. So for any given client > > application > > > > cache > > > > > > will be holding data which is not relevant to that application. > > > > > > > > > > > > Is there any way where by I still maintain the remote cache, but > > > > synchronize > > > > > > with primary cache server "selectively" - based on a application ID > > or > > > > > > something like that. > > > > > > > > > > > > Regards, > > > > > > Hari > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > Cheers, > > > > Hariprasath Manivannan, > > > > M : (006) 016 970 8470. > > > > http://kl-diary.blogspot.com > > > > > > > > > > > > > > > -- > > Cheers, > > Hariprasath Manivannan, > > M : (006) 016 970 8470. > > http://kl-diary.blogspot.com > > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
