Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
In regard to our caching discussion, Frank, I think you will like the following article. The prior article about two essential filters is interesting too. http://www.onjava.com/pub/a/onjava/2004/03/03/filters.html?page=1 One thing seems certain: there is complete serverside cache control. -- You can lead a horse to water but you cannot make it float on its back. Heaven has changed. The Sky now goes all the way to our feet. ~Dakota Jack~ This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
snip I suspect that it is relatively small and, when you introduce sophisticated state and caching options, it may be faster. Relative to what? To the web server dealing with it? I would suspect it's actually relatively BIG compared to that. I'm certainly willing to be proved wrong though. /snip As we agree, we just need to get the data. I think there should be a lot of standard testing out there that is more reliable than I am probably going to do off the top. I will look around a bit and maybe someone will chime in with some help here. snip That might be true... one other point to remember though is that the more unusual things you do, the harder it will be to get people able to maintain it. I fight standards at work as much as the next guy just because creative people don't like standards forced on them, but clever solutions usual equal difficult to maintain solutions. I don't think I'm telling you something you don't know :) But, that's not my argument, so don't spend any time on it. Just another side track we could go down :) /snip I am 100% in favor of keeping maintenance low. That was my original primary objective going here. I could not believe how much time was being spent on URLs, base tags, etc., etc., with all the concomittant confusion throughout the development process. snip I think the bigger hit is reading the danged thing. This obviously is especially so when there is an ongoing use of changing the JSP page. This has no penalty with ProtocolPages. /snip I just mean the more complicated parsing rules that go with JSP, as well as everything else. snip That's what I like, woman that are easy to please :) (You were left SO wide open there I just couldn't resist!) /snip Heh, heh! Remind me never to fence with you, Frank. Actually, and this is really off topic, in my milieu I am dearly loved because I have really learned how to be a friend, husband, etc. This has been the largest accomplishment by far in my life and the one I most cherish. snip * (Does the app server pass the response back to the web server to serve, or does it serve it directly to the client at that point? I'm actually not sure, but let's be optimistic and say the web server is out of the equation at this point, although I suspect that's NOT the case) In any case, images are returned to client- /snip I think that the ResourceAction class actually acts as the web server and that is why the return is null. The class writes to the responses output stream and that is all the server does, right? FileInputStream fis = new FileInputStream(fileName); BufferedInputStream bis = new BufferedInputStream(fis); byte[] bytes = new byte[bis.available()]; OutputStreamos= response.getOutputStream(); bis.read(bytes); os.write(bytes); os.flush(); os.close(); snip There is clearly more work done with the second chart because the app server is now involved. How costly is all that work? /snip Again, I am not sure about this, and it is the sort of thing that one typically makes mistakes about if one does not go by the data. So, I am going to hold judgment on it, but my intuitions are not running in the same direction as yours. I keep remembering how the following of good OOP principles can make Java applications faster that C applications doing the exact same thing, although most think that would not be the case. Just the other day I was reading about a guy who tried and failed to do some imaging work in early Java and had to go to C. Now Java is faster in that area. snip You could make the argument that because any servlet-based application is incurring those costs with most requests, what's the big deal about adding a few more? To a degree that would be a reasonable argument, but scalability is most certainly at risk because viewed from the point of view of the app server, a single page really corresponds to 10 app server requests. /snip I think there are trade-offs that will need to be measured. For example, I can envision with some of the stuff I do creating whole pages dynamically with images. The new Java imaging stuff is really, really, cool on this. Flash has its side too. snip Even if it's all done in the most efficient way, those ten requests look, for all intents and purposes, like 10 simultaneous USERS (assuming 1 request per user). So, maybe your app server can handle 100 concurrent requests... If the web server was allowed to serve the images, your app server still has 100 slots available to service requests, which corresponds generally to 100 concurrent users... If it's serving 10 images for each physical user though, now you can only service 10 concurrent users, so you've reduced your overall server capacity (as viewed by outside clients) by 90%. Ouch. /snip Now just one moment, Bub! LOL ;-) You really are not seeing the ways you can save time here. For
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
I think the worst case is 22 versus 32, Frank. with 10 images. See your note and then my reasoning below that. snip Even if it's all done in the most efficient way, those ten requests look, for all intents and purposes, like 10 simultaneous USERS (assuming 1 request per user). So, maybe your app server can handle 100 concurrent requests... If the web server was allowed to serve the images, your app server still has 100 slots available to service requests, which corresponds generally to 100 concurrent users... If it's serving 10 images for each physical user though, now you can only service 10 concurrent users, so you've reduced your overall server capacity (as viewed by outside clients) by 90%. Ouch. I fully acknowledge those are rough, worst-case numbers... I certainly don't mean to imply that your approach is 90% worse. Not at all! Just trying to illustrate the problem, as I see it, in certain environments. /snip app server = (AS) struts server = (SS) req = request -- = pass res = response With ResourceAction ___ First case HTML = req (AS) res (AS) = 2 Second image JPEG (say) = req (AS) -- res (SS) = 3 . Tenth image JPEG (say) = req (AS) -- res (SS) = 3 WIthout ResourceAction ___ First case HTML = req (AS) res (AS) = 2 First image JPEG (say) = req (AS) res (AS) = 2 Second image JPEG (say) = req (AS) res (AS) = 2 . Tenth image JPEG (say) = req (AS) res (AS) = 2 This is 22 versus 32. Apparently you forgot (I think?) that the app server has to handle ten images too. They don't just go out with the page, although we are looking at this in a very oversimplified sense. There is no question that the AS is quicker with HTML than the SS, but I am not so sure about the images. The SS may be faster. There is lots of room here for tuning. -- You can lead a horse to water but you cannot make it float on its back. ~Dakota Jack~ You can't wake a person who is pretending to be asleep. ~Native Proverb~ Each man is good in His sight. It is not necessary for eagles to be crows. We are poor . . . but we are free. ~Hunkesni (Sitting Bull), Hunkpapa Sioux~ This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
Dakota Jack wrote: I just mean the more complicated parsing rules that go with JSP, as well as everything else. Ok, gotcha. But, this only applies for the first access to the JSP, right? From then on it's a servlet invocation (which is more expensive than returning just a plain'old HTML document, right? But now I am splitting hairs) Heh, heh! Remind me never to fence with you, Frank. Actually, and this is really off topic, in my milieu I am dearly loved because I have really learned how to be a friend, husband, etc. This has been the largest accomplishment by far in my life and the one I most cherish. That's all that counts in the end. I have a wife of ten years and two great kids (well, most of the time they are great anyway), and although I have to remind myself of it sometimes because I get so wrapped up in other things that aren't as important, they are without a doubt my greatest accomplishment. I think that the ResourceAction class actually acts as the web server and that is why the return is null. The class writes to the responses output stream and that is all the server does, right? I thought so too at first, but upon further reflection I'm not so sure... If a request comes in to the web server and then it forwards it on to the app server, that would mean at some very low level that the web server was passing along the connection to the app server... I'm not so sure it's anything that complex... It may be that the app server renders the response stream, but then passes it back to the web server to return to the client. The bottom line though is that we're talking a level low enough that I don't know the answer for sure. Again, I am not sure about this, and it is the sort of thing that one typically makes mistakes about if one does not go by the data. So, I am going to hold judgment on it, but my intuitions are not running in the same direction as yours. I keep remembering how the following of good OOP principles can make Java applications faster that C applications doing the exact same thing, although most think that would not be the case. Just the other day I was reading about a guy who tried and failed to do some imaging work in early Java and had to go to C. Now Java is faster in that area. snip I too await the data :) But, I think you'd have to agree that for your approach to wind up being faster, much like when Java programs are faster than C programs, it must be due to some hidden optimization going on. I mean, on an operation-per-operation basis, C will ALWAYS beat Java... Simply put, there will always be less machine code ops going on with a C program at the lowest levels (assuming they algorithmically equivalent) than a Java program. But, because a Java program can be optimized at runtime, that's where the speed gains occur that you can't get with C. Something similar must be happening if your approach winds up being faster. All things being equal (ceteris paribus), that's the only logical conclusion to reach. That doesn't make it the right one of course :) The data will be the decider... I think there are trade-offs that will need to be measured. For example, I can envision with some of the stuff I do creating whole pages dynamically with images. The new Java imaging stuff is really, really, cool on this. Flash has its side too. No argument there... it's the same with OOP vs. strictly procedural code... if we assume we're always talking a straight compile to machine code and similar optimizations in both, procedural code should generally be faster for the same reason I talked about above. Clearly though, the benefits of an OOP approach outweigh any difference. Same could be true here. Now just one moment, Bub! LOL ;-) You really are not seeing the ways you can save time here. For example, there is such a thing as caching, pragmas and expiry headers which can be set with a response in a way that the meta tags just cannot handle. But now your pushing those caching decisions back on the browser, right? I thought one of your basic premises was to not trust the browser to construct URLs and such? Wouldn't you have the same distrust for caching? (and probably worse since that is at least at the users' discrection) There will be savings of creating no calls where pure HTML would be lost. There will be other things like this too. Remember too that the ResourceAction class is acting as a multithreaded alternative mini-server. Indeed, the approach allows us to get the images, for example, from some other server that is maximized to do just this. Conceivably that could be quicker for cached images. Remember I said conceivably. The ability to be flexible can make for great rewards in efficiency and fluidity that are not immediately obvious. Granted, some additional flexibility might outweigh any problems. If you rolled my BLOBServerAction into your ResourceAction, then you could transparently serve images from
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
Dakota Jack wrote: app server = (AS) struts server = (SS) req = request -- = pass res = response You lost me here already... What's the difference between the app server and the struts server? Isn't Struts running IN your app server? With ResourceAction ___ First case HTML = req (AS) res (AS) = 2 Second image JPEG (say) = req (AS) -- res (SS) = 3 . Tenth image JPEG (say) = req (AS) -- res (SS) = 3 WIthout ResourceAction ___ First case HTML = req (AS) res (AS) = 2 First image JPEG (say) = req (AS) res (AS) = 2 Second image JPEG (say) = req (AS) res (AS) = 2 . Tenth image JPEG (say) = req (AS) res (AS) = 2 This is 22 versus 32. Apparently you forgot (I think?) that the app server has to handle ten images too. They don't just go out with the page, although we are looking at this in a very oversimplified sense. I don't see how you got 22 OR 32! :) The first request from the client is for the HTML document, right? So that's one request. The browser then sees ten img tags, regardless of what they point to. So for each one it makes a request. That's 10 requests, right? So it's 11 in all from the client to the server (ignoring for the moment whether server means app server alone or web server in front of app server, or whatever other configuration you might dream up). The only different between the two approaches we've been discussing is what on the server is going to handle each of those 11 requests... Is it a web server sending back a static HTML page for the first request, and then an image for each of the subsequent 10 image requests, or is it a web server returning the HTML page and then an app server returning the images, or an app server returning the page AND the images? There is no question that the AS is quicker with HTML than the SS, but I am not so sure about the images. The SS may be faster. There is lots of room here for tuning. Let me ask you this question... If you are accessing a web site, and you connect directly to the Internet, is that, ignoring things like caching and such, generally going to be faster than going through a proxy? I'd hope you would say yes. Now, clearly, if the proxy is doing caching and/or other optimizations, it might turn out to be faster, but that further proves my point: the web server is like the proxy in this example. -- Frank W. Zammetti Founder and Chief Software Architect Omnytex Technologies http://www.omnytex.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
snip I think that the ResourceAction class actually acts as the web server and that is why the return is null. The class writes to the responses output stream and that is all the server does, right? I thought so too at first, but upon further reflection I'm not so sure... If a request comes in to the web server and then it forwards it on to the app server, that would mean at some very low level that the web server was passing along the connection to the app server... I'm not so sure it's anything that complex... It may be that the app server renders the response stream, but then passes it back to the web server to return to the client. The bottom line though is that we're talking a level low enough that I don't know the answer for sure. /snip I am certain on this one, because you can do this sort of thing *without* the web or app servers at all. I do this fairly frequently with code not unlike and heavily borrowing in principle from Jason Hunters HttpMessage and HttpsMessage in COS. The ResourceAction sends the response and ends the whole process by returning null. snip I too await the data :) But, I think you'd have to agree that for your approach to wind up being faster, much like when Java programs are faster than C programs, it must be due to some hidden optimization going on. I mean, on an operation-per-operation basis, C will ALWAYS beat Java... /snip Well, maybe on on an operation basis. An operation by any other name is still an operation. However, I don't disagree and would merely quibble about the language and the description. snip Simply put, there will always be less machine code ops going on with a C program at the lowest levels (assuming they algorithmically equivalent) than a Java program. /snip Well put! Yes! snip But, because a Java program can be optimized at runtime, that's where the speed gains occur that you can't get with C. /snip At the very least this is a main place to gain speed: the Tortoise and the Hare come to mind. snip ceteris paribus /snip Heh, I meant to tell you last time, this is Latin, not Greek. LOL ///;-) www.m-w.com Main Entry: ce·te·ris pa·ri·bus Pronunciation: 'kA-tr-s-'par--bs, 'ke-, 'se- Function: adverb Etymology: New Latin, other things being equal : if all other relevant things, factors, or elements remain unaltered snip But now your pushing those caching decisions back on the browser, right? I thought one of your basic premises was to not trust the browser to construct URLs and such? Wouldn't you have the same distrust for caching? (and probably worse since that is at least at the users' discrection) /snip The answers are no, yes, no. Setting caching in the response object is not equivalent to setting caching in the meta tags. This is why the ResourceAction has an edge. Note also that the setting of cache, pragma and expires are runtime alterable, and can override the meta tags, in ResourceAction. I left those decisions out of the code I sent you. Did you notice where I added in it response to someone's query on that? snip Granted, some additional flexibility might outweigh any problems. If you rolled my BLOBServerAction into your ResourceAction, then you could transparently serve images from WEB-INF *or* a database, transparently to the user and front-end. That's a nice bit of flexibility to be sure. /snip And, if you imagined more radical uses of images for whole pages, etc., then you might start thinking about BufferedImages cached in sessions, etc. snip I leave the leg-work to you :) /snip You got it! Jack -- You can lead a horse to water but you cannot make it float on its back. ~Dakota Jack~ You can't wake a person who is pretending to be asleep. ~Native Proverb~ Each man is good in His sight. It is not necessary for eagles to be crows. We are poor . . . but we are free. ~Hunkesni (Sitting Bull), Hunkpapa Sioux~ This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
Too late when I sent this. Let me make the necessary alterations to the nomenclature. Sorry! web server = df. (WS) app server = df. (AS) request= df. req response = df. res = df. passing the control With ResourceAction 1.0 WS req WS res HTML [2] 1.1 WS req AS res [3] 1.2 WS req AS res [3] 1.3 WS req AS res [3] 1.4 WS req AS res [3] 1.5 WS req AS res [3] 1.6 WS req AS res [3] 1.7 WS req AS res [3] 1.8 WS req AS res [3] 1.9 WS req AS res [3] 1.10 WS req AS res [3] Total 32 Without ResourceAction 1.0 WS req WS res HTML [2] 1.1 WS req AS res [2] 1.2 WS req AS res [2] 1.3 WS req AS res [2] 1.4 WS req AS res [2] 1.5 WS req AS res [2] 1.6 WS req AS res [2] 1.7 WS req AS res [2] 1.8 WS req AS res [2] 1.9 WS req AS res [2] 1.10 WS req AS res [2] Total 22 However, let me note, once again, that we can make it 22 to 22 by simply sending the attributes that are relevant back to a different server. For example, we could have img src='http://blahblahblah.com/ResourceAction.do?file=whatever.gif' Doing this, if we are talking about serving images to a large-scale site, we could get rid of both the WS and the AS and use a SCS (small custom server) optimized for this situation. I do this sort of thing constantly, *sub rosa*, on my sites. This is probably quicker than using WS to serve the images, and certainly so if the images are in any way dynamic in nature and if we make use of the multithreading opportunities that crop up in this situation. But, this is going afield. And, this is only looking at the upside too. Jack -- You can lead a horse to water but you cannot make it float on its back. ~Dakota Jack~ You can't wake a person who is pretending to be asleep. ~Native Proverb~ Each man is good in His sight. It is not necessary for eagles to be crows. We are poor . . . but we are free. ~Hunkesni (Sitting Bull), Hunkpapa Sioux~ This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
Dakota Jack wrote: I am certain on this one, because you can do this sort of thing *without* the web or app servers at all. I do this fairly frequently with code not unlike and heavily borrowing in principle from Jason Hunters HttpMessage and HttpsMessage in COS. The ResourceAction sends the response and ends the whole process by returning null. I agree, obviously you can take Tomcat for instance and use it to serve everything... I have a production app that is a single server running Tomcat and Oracle and that's it, no web server anywhere, everything is served from Tomcat whether it's a JSP, an image, an HTML document or whatever else. The question that's in my mind though is what happens when you have a web server in front of Tomcat? Just rendering to the response in a servlet might not be enough in that case... Think of a proxy analogy... Does the web server almost appear like a proxy? In other words, a request comes in to the web server, does it (a) pass the connection to the app server to fulfill, at which point it's done and can service more requests, or (b) does it ask the app server for the resource, whatever it is, wait for the response from the app server and send it along to the client when the app server is done responding? Same idea as a network proxy. The point being, just because the app server CAN serve everything, doesn't necasserily mean it WILL with a web server in front. But again, I don't know the answer here, it's just a question in my mind. ceteris paribus /snip Heh, I meant to tell you last time, this is Latin, not Greek. LOL ///;-) Really?? Well, I have something to yell at my Macroeconomics professor for then! I know for sure she said it was Greek! :) Funny aside... My Macroeconomics professor... her last name, and I couldn't have made this up, is Economopolous. That just rules! But now your pushing those caching decisions back on the browser, right? I thought one of your basic premises was to not trust the browser to construct URLs and such? Wouldn't you have the same distrust for caching? (and probably worse since that is at least at the users' discrection) /snip The answers are no, yes, no. Setting caching in the response object is not equivalent to setting caching in the meta tags. This is why the ResourceAction has an edge. Note also that the setting of cache, pragma and expires are runtime alterable, and can override the meta tags, in ResourceAction. I left those decisions out of the code I sent you. Did you notice where I added in it response to someone's query on that? I did notice, but my point is that the browser settings would override any tags or headers you set. I might be wrong about that, but that would be my expectation. After all, what good is a setting in my browser that says don't cache anything if a web site designer can come along and overrule that? Surely the FOSS community would be up in arms over their loss of freedom, right?!? ;) -- Frank W. Zammetti Founder and Chief Software Architect Omnytex Technologies http://www.omnytex.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
I still don't understanding the 32 and 22... What do the [2] and [3]'s represent? Dakota Jack wrote: Too late when I sent this. Let me make the necessary alterations to the nomenclature. Sorry! web server = df. (WS) app server = df. (AS) request= df. req response = df. res = df. passing the control With ResourceAction 1.0 WS req WS res HTML [2] 1.1 WS req AS res [3] 1.2 WS req AS res [3] 1.3 WS req AS res [3] 1.4 WS req AS res [3] 1.5 WS req AS res [3] 1.6 WS req AS res [3] 1.7 WS req AS res [3] 1.8 WS req AS res [3] 1.9 WS req AS res [3] 1.10 WS req AS res [3] Total 32 Without ResourceAction 1.0 WS req WS res HTML [2] 1.1 WS req AS res [2] 1.2 WS req AS res [2] 1.3 WS req AS res [2] 1.4 WS req AS res [2] 1.5 WS req AS res [2] 1.6 WS req AS res [2] 1.7 WS req AS res [2] 1.8 WS req AS res [2] 1.9 WS req AS res [2] 1.10 WS req AS res [2] Total 22 However, let me note, once again, that we can make it 22 to 22 by simply sending the attributes that are relevant back to a different server. For example, we could have img src='http://blahblahblah.com/ResourceAction.do?file=whatever.gif' Doing this, if we are talking about serving images to a large-scale site, we could get rid of both the WS and the AS and use a SCS (small custom server) optimized for this situation. I do this sort of thing constantly, *sub rosa*, on my sites. This is probably quicker than using WS to serve the images, and certainly so if the images are in any way dynamic in nature and if we make use of the multithreading opportunities that crop up in this situation. But, this is going afield. And, this is only looking at the upside too. Jack If we are talking about dynamically-created resources, then I would tend to agree with your view. But we have, at least as far as I was concerned, been talking about strictly static resources. In that case, your basic premise boils down to, as I see it: An app server running ResourceAction can serve resources more efficiently than a web server. Again, strictly talking about static resources, I would be absolutely SCHOCKED to learn this is the case under most circumstances. That would be like saying a Cadillac could beat a NASCAR vehicle in 1 ten-lap race... It might be able to under some circumstances, like the NASCAR driver being drunk!, and certainly there are some very nice trade-offs to driving the Caddy like more room and a better stereo, but in general you wouldn't expect the Caddy to lose. A bit of hyperbole there, but the underlying point is what's important. -- Frank W. Zammetti Founder and Chief Software Architect Omnytex Technologies http://www.omnytex.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
snip The question that's in my mind though is what happens when you have a web server in front of Tomcat? Just rendering to the response in a servlet might not be enough in that case... /snip *Before* ResourceAction returns null, the response output stream has been written, flushed, and closed. The only thing that the app server or the web server have left to deal with is that null. There is no wrapper in this case and no proxy in the sense you are talking. The OutputStream from an HttpResponse object writes to the client. snip The point being, just because the app server CAN serve everything, doesn't necasserily mean it WILL with a web server in front. /snip But, in this case, the OutputStream does and there is no pass it on functionality in there that would incorporate any reference or use of the web or app server. The fact that this OutputStream ends the process might be one of the factors favoring ResourceAction. snip ceteris paribus /snip Heh, I meant to tell you last time, this is Latin, not Greek. LOL ///;-) Really?? Well, I have something to yell at my Macroeconomics professor for then! I know for sure she said it was Greek! :) Funny aside... My Macroeconomics professor... her last name, and I couldn't have made this up, is Economopolous. That just rules! /snip LOL Economopolous! Hilarious Remember My Big Fat Greek Wedding where the Greek guy has a way of turning everything to Greek history, etc.? Well, Ms. Economopolous is clearly Greek and her name is Economic-city. (Plato's Republic was really Politia which means The City. Republic comes from Republica which was a Latin translation.) Anyway, this is ALL ceteris paribus. (You can tell Latin from the endings, ibus is the dative plural.) snip I did notice, but my point is that the browser settings would override any tags or headers you set. I might be wrong about that, but that would be my expectation. After all, what good is a setting in my browser that says don't cache anything if a web site designer can come along and overrule that? Surely the FOSS community would be up in arms over their loss of freedom, right?!? ;) /snip The good is that the web site designer knows when a change has been made and the assumption is that you are going to see what the web site designer has to offer. No? Jack -- You can lead a horse to water but you cannot make it float on its back. ~Dakota Jack~ You can't wake a person who is pretending to be asleep. ~Native Proverb~ Each man is good in His sight. It is not necessary for eagles to be crows. We are poor . . . but we are free. ~Hunkesni (Sitting Bull), Hunkpapa Sioux~ This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
snip On Sun, 30 Jan 2005 14:11:24 -0500, Frank W. Zammetti [EMAIL PROTECTED] wrote: I still don't understanding the 32 and 22... What do the [2] and [3]'s represent? /snip A total of three possible processes (1) getting the request; (2) passing the request to another server; (3) handling the response. If you have them all, you have a three. If only 1 and 3, then you have a two. snip If we are talking about dynamically-created resources, then I would tend to agree with your view. But we have, at least as far as I was concerned, been talking about strictly static resources. /snip If there are static resources, then we can get it down to 22 versus 22 by sending the images to a separate server. Not only can we do this, but we can send the images to a super efficient separate server if we are talking about static images only. snip An app server running ResourceAction can serve resources more efficiently than a web server. /snip Not that an app server is faster under any circumstances tha a web server. That really is not close to true. I've seen the stats on that one and I would doubt that they will ever be the same or close to the same. I would be as SCHOCKED as you (is this an Italian-Jewish SHOCKED? ///;-) ) in that case. What I am talking about is a custom server for images which gets rid of a LOT of baggage, including WEB-INF but having the same protections as being under WEB-INF. snip Again, strictly talking about static resources, I would be absolutely SCHOCKED to learn this is the case under most circumstances. That would be like saying a Cadillac could beat a NASCAR vehicle in 1 ten-lap race... It might be able to under some circumstances, like the NASCAR driver being drunk!, and certainly there are some very nice trade-offs to driving the Caddy like more room and a better stereo, but in general you wouldn't expect the Caddy to lose. /snip In this case the analogy, IF apt, is the reverse. The custom server is the NASCAR. All the doodads needed on an app or a web server can be pealed off and serious savings with multithreading, parsing presumptions, etc. can be realized. snip A bit of hyperbole there, but the underlying point is what's important. /snip I enjoyed the ride in the caddy. Had the stereo on a good jazz station in my mind with Lead Belly growling at me. The metaphor is apt but really, when you are talking a mini-quick-custom server, reversed. I am actually surprised that there are not more of these little speedy and specialized servers around. Jack This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
Dakota Jack wrote: The good is that the web site designer knows when a change has been made and the assumption is that you are going to see what the web site designer has to offer. No? Jack I concur with the assumption, but I don't see it making any difference... Remember that what we affectionately refer to as the web these days is really a lot more than what it was originally. One of the things that was originally intended is that the client is in complete control. If the user wants to request a resource from a server, and they tell their browser via a setting that no, I WANT you to go ask that server EVERY SINGLE TIME for the resources, NEVER use what you might have in the local cache, then they should be allowed to do that, and whatever the web site creator wants you to do is irrelevant. Same idea when the user can override fonts and colors and the like with their own local settings. Nowadays though, us web app/site designers think WE know how best a client should view our site, and we actually go out of our way to make it so... how many times have you visited a site where the font is too small and the usual font size adjustments don't make any difference? So you have to go in to setting and uncheck that User Font Sizes Specified By Site option. Annoying, and not what was originally intended. This is of course all only relevant to the extent that it supports my point, that anything you do on the server side to try and control caching is either (a) useless because the end user can override it anyway or (b) not in keeping with the spirit of the web, at least, not as originally intended. Now I'm off on a bit of a tangent though :) -- Frank W. Zammetti Founder and Chief Software Architect Omnytex Technologies http://www.omnytex.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[OT] Re: JSP under /WEB-INF folder PROTOCOL PAGES (ProtocolPages)
I marked my response as OT, I think we're going down that road (not exactly unusual for us)... Dakota Jack wrote: What I was getting at is the fact that if I return a page to the browser that have ten images, all referencing ResourceAction, what's happening is that the browser is making ten separate requests TO THE APP SERVER, whereas in a typical setup, these requests would be handled by the fronting web servers. It's clearly more resource-intensive in your approach. /snip I am not clear what part of the process you are referring to, Frank. We both agree that the first delivery of the page is via the front server (I tend to only use the back server anyway). I wasn't clear on that part I guess, but it doesn't change my argument. I guess your saying the pages aren't even JSPs, which is fine. Doesn't really change anything though except that there's no servlet involvement to serve the initial page, which is slightly better. That's cool... If there are ten calls to ten images, as you assume for discussion purposes, then there will be ten calls either way. That's correct. Just basic HTTP here :) I think you are saying that in addition there will be a penalty of a pass to a server that can handle Servlets or an equivalent technology that will respond to the ProtocolPage by routing the call to some Action, Command, or whatever in some language, in the way I suggest. Is that right? If so, let me know and we will go from there after you confirm. That's it exactly. The web server has to forward the .do request to the app server, which then serves the images, 10 total servlet invocations in all. Yes, there would be ten requests either way, but one assumes that the web server can serve static content more efficiently than any Action you can devise. Perfectly logical argument, ceteris paribus (I love using Greek in real conversations!), because that's exactly what web servers are optimized to do. I wouild need, as you would too I assume, more information on the actual penalty. Absolutely. The only assumption I'm making is that there IS a penalty. There are any number of factors that would go into how big or small it is, whether it's enough to outweigh the benefits of the approach. I make no attempt here to reach any conclusion on the magnitude of the problem, if it is a problem at all. I suspect that it is relatively small and, when you introduce sophisticated state and caching options, it may be faster. Relative to what? To the web server dealing with it? I would suspect it's actually relatively BIG compared to that. I'm certainly willing to be proved wrong though. I don't dismiss what you are saying. Don't get me wrong. I just have learned to get the data and then to see what the real difference is. And I would agree that's the right frame of mind in all things :) It may in fact be such a small penalty that it's worth ignoring. I have a suspicion it's not, but without empirical data, it's just a hypothesis. When considering costs and so on, I am not sure whether the balance goes to which side. I would suspect, from my experience, that software maintenance and so on would clearly outweigh the hardware and associated requirements. That might be true... one other point to remember though is that the more unusual things you do, the harder it will be to get people able to maintain it. I fight standards at work as much as the next guy just because creative people don't like standards forced on them, but clever solutions usual equal difficult to maintain solutions. I don't think I'm telling you something you don't know :) But, that's not my argument, so don't spend any time on it. Just another side track we could go down :) I am surprised at this. You may be right, but my sense is that this difference is not really that important when everything else is taken into account. Even if you had to cluster multiple machines instead of one, say, as a ratio, that would seem to be *probably* cheaper as a GUESS. I don't know. We could look at some data and if you have any handy I would love to see it. I don't have any handy, but I agree, some real data here would be better than us bantering back and forth for the next week :) I suppose the simplest thing to do is set up a test where you serve 10 static images from a web server vs. 10 images from your ResourceAction with a web server in front of the app server and see what the overall response time is. It would be a little rough, but at least we could tell if there is a perceivable difference or not. Then we get into whether thousands of imperceivable hits accumulate enough over time to hurt us. That's obviously far trickier to guage, but that's the kind of things us enterprise architects have to bother about. :) I think the bigger hit is reading the danged thing. This obviously is especially so when there is an ongoing use of changing the JSP page. This has no penalty with ProtocolPages. Not
