Hi Shariq, Yes that is why we are using the cache. the cache will be distributed. When node 1 changes the RXT it will flip the parity bit in the cache entry that correspondence to the changed RXT, then the listener will be called in node 2 since the entry was changed. And within the listener the changes will be loaded to the artifact configuration map.
Regards, Pulasthi On Tue, Sep 16, 2014 at 11:38 AM, Shariq Muhammed <sha...@wso2.com> wrote: > On Tue, Sep 16, 2014 at 11:28 AM, Pulasthi Supun <pulas...@wso2.com> > wrote: > >> Hi Azeez, >> >> Yes that can be done, that would reduce the number of DB calls further. >> We can simply keep an entry for each RXT type that will keep a parity bit >> as a value. Once a update is done the listener can query the exact RXT >> type that was changed and only load that value. >> > > One more thing, we need to consider a cluster scenario as well right. If > we update an RXT from node-1, then node-2 should also invalidate the cache > and look into the DB. > > >> >> Regards, >> Pulasthi >> >> On Tue, Sep 16, 2014 at 11:04 AM, Afkham Azeez <az...@wso2.com> wrote: >> >>> For this type of problem, caching coupled with loading only the changed >>> parts is the way to go. >>> >>> On Tue, Sep 16, 2014 at 10:53 AM, Pulasthi Supun <pulas...@wso2.com> >>> wrote: >>> >>>> Hi All, >>>> >>>> Currently when working with RXT's each time the loadGovernanceArtifacts >>>> method is called it in turn will call the DB to load the RXT's, this >>>> is to load any changes done to the RXT's . This a major performance >>>> hit especially in APIM since when loadGovernanceArtifacts is called >>>> multiple times a large number of DB calls are executed. After looking into >>>> this came up with the following solution to reduce the number of database >>>> calls. >>>> >>>> We currently load all the artifact configurations to a map that keeps >>>> all the artifact configurations against the tenant id. If we can find out >>>> when the artifact configurations are updated we can limit the number >>>> of times the database is queried by loading the configurations only when >>>> there has been a change. To track this we will keep a small distributed >>>> cache which will keep a entry for each tenant and will act as parity >>>> bit and add a listener for this cache for cache entry update. The >>>> lister will then load the changes into the map that keeps the artifact >>>> configurations. This way will remove the need to query database every time. >>>> WDYT? >>>> >>>> >>>> Regards, >>>> Pulasthi >>>> >>>> -- >>>> -- >>>> Pulasthi Supun >>>> Software Engineer; WSO2 Inc.; http://wso2.com, >>>> Email: pulas...@wso2.com >>>> Mobile: +94 (71) 9258281 >>>> Blog : http://pulasthisupun.blogspot.com/ >>>> Git hub profile: https://github.com/pulasthi >>>> >>> >>> >>> >>> -- >>> *Afkham Azeez* >>> Director of Architecture; WSO2, Inc.; http://wso2.com >>> Member; Apache Software Foundation; http://www.apache.org/ >>> * <http://www.apache.org/>* >>> *email: **az...@wso2.com* <az...@wso2.com> >>> * cell: +94 77 3320919 <%2B94%2077%203320919>blog: * >>> *http://blog.afkham.org* <http://blog.afkham.org> >>> *twitter: **http://twitter.com/afkham_azeez* >>> <http://twitter.com/afkham_azeez> >>> *linked-in: **http://lk.linkedin.com/in/afkhamazeez >>> <http://lk.linkedin.com/in/afkhamazeez>* >>> >>> *Lean . Enterprise . Middleware* >>> >> >> >> >> -- >> -- >> Pulasthi Supun >> Software Engineer; WSO2 Inc.; http://wso2.com, >> Email: pulas...@wso2.com >> Mobile: +94 (71) 9258281 >> Blog : http://pulasthisupun.blogspot.com/ >> Git hub profile: https://github.com/pulasthi >> > > > > -- > Thanks, > Shariq > Senior Software Engineer > -- -- Pulasthi Supun Software Engineer; WSO2 Inc.; http://wso2.com, Email: pulas...@wso2.com Mobile: +94 (71) 9258281 Blog : http://pulasthisupun.blogspot.com/ Git hub profile: https://github.com/pulasthi
_______________________________________________ Architecture mailing list Architecture@wso2.org https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture
