Oh i guess i should have stated that when I request the data from my web service (coldfusion cfc), 760,000 is about the number of records (in total) that i'm expecting to get (200 requests with 2100 records in each request). I have to parse through those records to create the Charts. I'm pulling the records back with a remoteobject request. I should also further say that the 760,000 records are all for the initial request. subsequent requests will be a lot less... but still along the lines of 12000 records every minute.
records are coming back in the format of: machineID, propertyid, minuteid, propertyvalue 1,1,213,98 1,2,213,78 2,1,213,94 Everything is numbers Thanks for the explanations jester. --- Jonathan Hirschi <[EMAIL PROTECTED]> wrote: > Well, the way I'm planning this out is that at the > moment, i need to have 60 chart objects with > essentially up around 40 or so series per chart > (estimated - could go as high as 60). Each series > is > an array of numbers. Each array has 60 values. > > I'm still noodling out how to best put this > together... > I need still need to manage time in all this for > labels on the charts, but the time element is going > to > be synced across alll of the charts, so I've been > playing around with updating the labels and > horizontal > values manually.. (not sure if that's the way to go > yet) > > so for example (hope this makes sense): > > Chart = cpu:object > > series = machine1:Array .... machine60:Array > > machine1:Array > 98 > 100 > 45 > 67 > etc.. > > machine2:Array > 56 > 34 > 98 > 45 > etc... > > --- JesterXL <[EMAIL PROTECTED]> wrote: > > > Sure; Flex 2 gets all of Flash 8's bitmap features > > with the added benefit of > > a fast as nuts AVM. As far as the data, though, > not > > sure. I've spent 3 > > years pushing Flash to it's limits and know where > > she breaks, but garbage > > collection & tons of data, I haven't really found > > the ceiling yet in 8.5. > > What I have found is "fixed in the next build", so > > it all sounds promising. > > > > Flash in general has problems creating millions of > > objects because of the > > overhead associated with the actual creation of > > "Object" in Flash. There is > > a lot of overhead, and while it's significant for > > the hundreds, for the > > thousands & hundreds of thousands, those bytes > start > > to add up. This is > > significantly improved in 8.5. No, I don't have > > details; basically, the > > amount of overhead is significantly reduced; it > was > > a big problem in Flash > > hitting the "hundreds of MovieClips" limit, and > now > > works great... well, as > > best an alpha can be. Besides, the way it was > > explained no sense... > > something about byte headers, and trait objects, > and > > all kinds of other > > low-level insanity you have to be a genious to > > understand... or a Comp Sci > > grad I reckon. > > > > Max memory? Nope; I've had Flash take up to 2 > gigs > > once, it was awesome! > > > > Again, haven't really pushed 8.5 yet since she > isn't > > done. > > > > As far as porting, yeah, I agree. We've used a > > Flash 8 loaded SWF in 2 > > projects to enable file-upload, and while it works > > for the most part, it's > > not eloquent, and not being eloquent ticks > > developers off. > > > > Actually, 720,000 doesnt' sound to bad... what > > properties do these data > > points have? Like an ID and Name, or a long list > or > > what? > > > > As far as GC, hell no; if Adobe published how > their > > GC worked, and changed > > it, we'd all be f00ked. I think Tinic's article > is > > the only one that has > > lasted: > > > http://www.kaourantin.net/2005/09/garbage-collection-in-flash-player-8.html > > > > There was a comment on another article where he > > described how using 1 member > > variable vs. many locals was better... even though > > locals were optimized for > > local registers in Flash Player 7 making functions > > faster... it's all > > confusing. > > > > There was an in depth article that disappeared > from > > the face of Google; I > > haven't seen it in 2 years, but don't know if it > was > > worth reading or not. > > > > Anyway, let us know about the data points; what > > their properties are, I can > > run some tests when I'm done with this project. > > > > > > > > ----- Original Message ----- > > From: "Jonathan Hirschi" <[EMAIL PROTECTED]> > > To: <flexcoders@yahoogroups.com> > > Sent: Friday, January 20, 2006 5:46 PM > > Subject: Re: [flexcoders] Performance question - > > please help! > > > > > > So in terms of Flex 2, would it be able to > possibly > > handle this amount of data effectively since it > uses > > flash 8? > > > > It sounds like you are saying that 1.5 has trouble > > displaying that amount of data because of ram > usage > > problems? does that same problem exist in flex 2? > > Is there a max on how much memory flash player 7 > is > > able to use effectivly? is that max much larger > in > > flash 8.5 player? > > > > It sounds to me like it might be easier to port > this > > thing to flex 2 than it would be to try and plug > in > > to > > flash 8 bitmap features and draw my own graphs. > > > > as far as data points, well, it's 200 machines * > 60 > > properties * 60 minutes = 720,000 data points > that > > i > > need to hold in memory at any given time. I'm > also > > being asked to update the data points at a rate of > > once per minute ie 200 machines * 60 properties = > > 12,000 updated data points per minute with an > > eventual > > goal to get those updates down to once every 10 > > seconds.. > > > > is that something that is doable by flex 2 even? > > They've got an app here written in mfc (c++) that > > can > > do it, but they'd like to have it accessible from > > anywhere on the web... > > > > Have there been any performance papers published > on > > flash describing memory usage, etc? > > > > Thanks! > > > > --- JesterXL <[EMAIL PROTECTED]> wrote: > > > > > Answering from the 30,000ft view, my suggestion > is > > > to utilize Flex 1.5 & > > > Flash 8's bitmap features. Flash Player 7, > which > > is > > > what Flex 1.5 outputs, > > > cannot handle the amount of data points you want > > > with acceptable > > > performance, either in code or displayed. While > > > paging can be effectively > > > employed, you'd probably lose the visual > === message truncated === __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- Flexcoders Mailing List FAQ: http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt Search Archives: http://www.mail-archive.com/flexcoders%40yahoogroups.com Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/flexcoders/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
