Announcement

Collapse
No announcement yet.

Benchmarking ZFS On FreeBSD vs. EXT4 & Btrfs On Linux

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Originally posted by kraftman View Post
    Those SAP official benchmarks didn't proof a thing. It was two unrelated benchmarks done on different hardware with different database used. Neither proof nor paper describing Linux scale worse then Solaris on such workload.
    Linux used faster hardware, faster CPU and faster RAM, and still lost that SAP benchmark. That is testament to Linux had low cpu utilization: 87%. Whereas Solaris had 99% cpu utilization. That is the reason won, on slower hardware.

    You are contradicting yourself. You accept benchmarks like FreeBSD vs Linux vs OpenSolaris without problem. Even though they do not use same software or gcc version: one of them compiles to 64bit, the other compiles to 32 bit. One uses gcc v3.xx and the other uses gcc v4.xx. And you think those benhcmarks are fair because Linux wins. You even posted a benchmark where one Linux fanboy compared one old 800 MHz sparc Solaris vs a intel dual core 2.4GHz Linux and you think that is a good benchmark because Linux won.

    You accept benchmarks on different hardware and different software when Linux win, and reject all benchmarks when Linux loose.

    You sir, you are a humbug and a liar.


    Originally posted by kraftman View Post
    Solaris developers lie and FUD, so I don't believe they will be fair with others and I don't believe they'll saying true things about Solaris. I'm also not interested in proving you anything. Like I said, I can post a link sometimes etc.
    I have posted several links to non-Solaris developers. Even Linux developers. But you reject even those links, even to Linus T.

    If a link is negative for Linux you reject it, it does not matter if it is Linus T that says so. If a link is good for Linux you immediately accept it. It does not matter if it is a Linux fanboy doing the worst and biased comparison in history - "it is a good and fair link"

    Have I told you are a Humbug and FUDer? You earlier confessed you do FUD.



    Originally posted by kraftman View Post
    It seems papers backing up your claims doesn't exist. At least, I couldn't find them. Now, could you show them to backup your claims?
    Dont play silly. Ive posted lots of research papers. You saw them.


    If I'm correct, I don't have to specify you a thing. In this link I gave someone claims Solaris is dead. I'm just pasting, why are you asking me?
    You are not correct. If you want me to show you something, then you must specify what you want me to show. How can I otherwise defend Solaris? Would like if government did say:
    -Kraftman, you are accused of a horrible crime. We will not tell you which crime. Do you confess? We will not specify.


    I'm aware of this. However, how is this different of showing some Solaris guys or fanboys articles? Or public relations talk? Those are usually even worse then some blog post. Afaik you even gave a link to Bonwick's blog and he FUDs.
    Regarding Bonwick, there are many experts that thinks he is correct. Bonwick (the ZFS main architect) has written negative things about Linux, but that is not FUD. That is relevant criticizm. There are lots of experts agreeing with Bonwick on his Linux criticism. Bonwick is not making things up nor lie out of nothing - he has most probably talked to other experts and Enterprise sysadmins.

    Regarding my links, do you consider my links to Linus T, Andrew Morton, etc - are FUD and lies? Do you think I have created my own web site and made up those interviews? Do you think I lie and FUD about those interviews?

    Comment


    • Your exactly words have been:

      In my opinion, if Oracle really wants to bet on BTRFS, Oracle should have assigned a whole team of developers to BTRFS. But guess what? There is only one single developer on BTRFS. Oracle is not interested in BTRFS, now that they got the best: ZFS. Come on, do you really expect BTRFS is a serious attempt by Oracle? There is only ONE full time developer! Jesus. Who are you trying to fool?
      Which is totally fud, follow the links I posted, you could see that btrfs is activelly developed by a more than 1 dev, at least 2 or more from oracle, red hat, ibm...

      If you now want to rephrase yourself telling me that you pointed 1 full time payed dev by oracle go ahead, do whatever you want. But the reality is that you have pointed bad information, and someone really you and trusting you (it's hard to believe that someone does) can learn wrong data.

      Comment


      • sorry , 1min.

        And if someone is reading and trusting you (it's hard to believe that someone does) then he/she could learn wrong data.

        Comment


        • This is not a guess. A Senior Storage expert says this. As other experts also says.
          A 2008 opinion, even on 2008 otherdevs had said that this guy was worng, i posted you data, but obviously you have pointed that

          http://lxer.com/module/newswire/lf/view/103802/

          is lying.

          Comment


          • Kebabbert wrote:
            "In my opinion, if Oracle really wants to bet on BTRFS, Oracle should have assigned a whole team of developers to BTRFS. But guess what? There is only one single developer on BTRFS. "

            Originally posted by Jimbo View Post
            Which is totally fud, follow the links I posted, you could see that btrfs is activelly developed by a more than 1 dev, at least 2 or more from oracle, red hat, ibm...
            How is that FUD? I read in the interview by the BTRFS main architect, he said he was the only full time paid developer at Oracle. That is not "a team". So, where do I FUD? Do you mean Chris Mason is not to be trusted in the interviews?


            Originally posted by Jimbo View Post
            If you now want to rephrase yourself telling me that you pointed 1 full time payed dev by oracle go ahead, do whatever you want. But the reality is that you have pointed bad information, and someone really you and trusting you (it's hard to believe that someone does) can learn wrong data.
            As I told you: I post information that I have read, from credible sources. This piece of information "one full time paid developer at Oracle on BTRFS" is from Chris Mason himself. Do you think he is not credible? Should I not trust him? Sorry I do not agree with your viewpoint in that case, I deem him credible. He said so, it is not FUD nor "bad information"

            If you can prove there is a whole team of full time paid developers at Oracle, then you have disproved me and showed me wrong. If you can not prove there is a full time team, then maybe it is correct that only Chris Mason is full time paid developer at Oracle? Especially as he says so, himself. I think he knows more on who is working on BTRFS, than you?

            Comment


            • more easy for intellects like you

              "There is only one single developer on BTRFS"

              Comment


              • Originally posted by Jimbo View Post
                A 2008 opinion, even on 2008 otherdevs had said that this guy was worng, i posted you data, but obviously you have pointed that

                http://lxer.com/module/newswire/lf/view/103802/

                is lying.
                You posted this link earlier, and I answered to your post. Did you not read my answer on this? Here is my answer on your link:
                http://phoronix.com/forums/showpost....1&postcount=44

                Comment


                • Originally posted by Jimbo View Post
                  more easy for intellects like you

                  "There is only one single developer on BTRFS"
                  I am refering to my earlier post on this; there is only one full time paid developer at Oracle.

                  After writing that sentence once, I refer to that sentence again, but only write down parts of the whole sentence. Are you playing dumb?

                  If you talk about Linux horizontal scalability, you sometimes only write "linux scalability" and you mean "Linux horizontal scalability" - but you do not write every single word. Everyone does that. It gets a pain to write down every single word in a long sentence, such as "one full time paid developer at Oracle working on BTRFS" - dont you think?

                  Comment


                  • You pointed that there was 1 dev, wanting to look it as it was an under development project. Rephrase whatever you like.

                    You arguments have nosense on a true open source project like btrfs. How many full time kernel dev has linux foundation and are payed by linux fundation?

                    Comment


                    • Originally posted by Jimbo View Post
                      You pointed that there was 1 dev, wanting to look it as it was an under development project.
                      Do you mean BTRFS is not under development? I must have missed that. When did Oracle release v1.0? Do you have links, or are you making this up?


                      Originally posted by Jimbo View Post
                      Rephrase whatever you like.
                      I was quoting BTRFS main architect who said in an interview, he was the only one full time developer at Oracle. I am not making that up. If you think I lie about this, I am not. Again, I only write here what other, more credible people say. For instance, Linus T, Andrew Morton, SAP official benchmarks, BTRFS main architect, researchers, etc.


                      Originally posted by Jimbo View Post
                      You arguments have nosense on a true open source project like btrfs. How many full time kernel dev has linux foundation and are payed by linux fundation?
                      I am not questioning if there are Linux developers. Why do you imply I question that?

                      I was only quoting BTRFS main developer who said he was the only full time developer at Oracle - if you think this is a lie and FUD - fine for you. But according to wikipedia, it is not FUD nor lies. And I also said that in my view point, if Oracle is really serious with selling a product, Oracle should dedicate a whole team, not a single developer (according to Chris Mason). How can you draw the conclusion I do FUD?!?!?! I do not FUD. If you continue to accuse me of FUDing, then it is you that FUDs and lies about me. Stop say I FUD, every assertion I did was true.

                      (Guesses and opinions are not FUD. False assertions are FUD. I never did false assertions. I only wrote opinions and guesses. Again: opinions are never wrong or FUD, but facts and assertions can be wrong or FUD).

                      Comment


                      • No, no, no you lie, you are under heavy fudding masturbation! you don't even know what you have written.

                        (opinion)In my opinion, if Oracle really wants to bet on BTRFS, Oracle should have assigned a whole team of developers to BTRFS (end opinion).

                        (assertion, fud begins) But guess what? There is only one single developer on BTRFS. Oracle is not interested in BTRFS, now that they got the best: ZFS. Come on, do you really expect BTRFS is a serious attempt by Oracle? There is only ONE full time developer! Jesus. Who are you trying to fool? (end assertion, fud ends)



                        There is no mention to Chris Mason interview, there is a clear intention to undervalue btrfs development. There is no earlier message speaking about one full time developer and lot of other devs, there is no earlier message about Chris Mason interview. You obviously have no idea how btrfs development is going. So You fud.

                        Fud is not about lying intentionally, is more like spreading false information because of ignorance or disrespect about something.

                        Comment


                        • "There is only one full time paid developer" So?

                          Comment


                          • Originally posted by doubledr View Post
                            "There is only one full time paid developer" So?
                            Some proof? Afaik there are many more.

                            Comment


                            • Originally posted by kebabbert View Post
                              Because you have not read this thread from the beginning, I post this article again. Here is a link to a research paper about how XFS, JFS, ReiserFS, etc does not protect your data - which is what I have claimed earler.
                              http://www.zdnet.com/blog/storage/ho...ta-at-risk/169
                              So, when I post to blog it's wrong, but when Kebabbert posts to blogs it's good?

                              [QUOTE]I have also posted several research papers which backs up my claim about ZFS giving good protection, whereas common filesystems (including Raid-5 and raid-6) do not offer good data protection.

                              Do you want me to repost those research papers for you?[QUOTE]

                              I'd love to you repost them. Btw. I want to see proofs there backing all you have claimed before.


                              Nevermind Kraftman, claiming I do not post to research papers., He just lies about me
                              FUD. I'm claiming your not posting to papers which backups your claims.

                              Comment


                              • Originally posted by kebabbert View Post
                                Linux used faster hardware, faster CPU and faster RAM, and still lost that SAP benchmark. That is testament to Linux had low cpu utilization: 87%. Whereas Solaris had 99% cpu utilization. That is the reason won, on slower hardware.
                                Solaris run different database, Oracle database. Afaik Solaris machine had twice amount of RAM more then Linux machine. Solaris server was also much more expensive then Linux server. It's possible, like it was said before Solaris server was much more expensive, because somebody cared to optimize this server for maximum CPU utilization. It's not a proof it scales better then Linux

                                You are contradicting yourself. You accept benchmarks like FreeBSD vs Linux vs OpenSolaris without problem. Even though they do not use same software or gcc version: one of them compiles to 64bit, the other compiles to 32 bit. One uses gcc v3.xx and the other uses gcc v4.xx. And you think those benhcmarks are fair because Linux wins.
                                Why are you lying? I don't think such comparisons are fair and I don't treat them as proofs. Stop lying, please.

                                You even posted a benchmark where one Linux fanboy compared one old 800 MHz sparc Solaris vs a intel dual core 2.4GHz Linux and you think that is a good benchmark because Linux won.
                                Yes, and you even posted two SAP unrelated papers, how is it different?

                                You accept benchmarks on different hardware and different software when Linux win, and reject all benchmarks when Linux loose.
                                You want me to accept benchmarks done on a different hardware. I consider it's fair benchmarking defaults like Phoronix does and if I claim Linux is faster it's not, because of Phoronix benchmarks, but because I believe so and because I saw things which suggest this.

                                You sir, you are a humbug and a liar.
                                It's you who's lying.

                                I have posted several links to non-Solaris developers. Even Linux developers. But you reject even those links, even to Linus T.
                                I don't consider those were proofs or papers, those were personal opinions which neither proof nor show Linux code is worse then Solaris.

                                Regarding Bonwick, there are many experts that thinks he is correct. Bonwick (the ZFS main architect) has written negative things about Linux, but that is not FUD. That is relevant criticizm. There are lots of experts agreeing with Bonwick on his Linux criticism. Bonwick is not making things up nor lie out of nothing - he has most probably talked to other experts and Enterprise sysadmins.
                                Papers, please. Linux dev said Bonwick is a FUDer and a lier. You're lying now saying Bonwick is not making things up nor lie.

                                Comment

                                Working...
                                X