Adobe, really ?

[Jean Moniatte]

Funny quote from the Adobe document twetted by Adam earlier:

The Wall Street Magnate development team experimented with open source CFML engines, so the

team understands the value of the Adobe ColdFusion platform. The first implementation of

wallstreetmagnate.com was prototyped with an open source implementation of ColdFusion. This was

suitable for demonstrating the key elements of the application but ran into scalability challenges as the

complexity of orchestrating information required for the application led to persistent engine locking

errors on the server that the developers were not able to easily resolve. After spending several months

trying to fix locking and concurrency issues, Wall Street Magnate decided to use the Adobe

ColdFusion server to overcome these challenges. 

http://wwwimages.adobe.com/content/dam/Adobe/en/products/coldfusion/pdfs/cf11/TurningUpTheHeat.pdf

I wonder which open source engine it was :-)

[Igal @ Lucee.org]

I wonder which open source engine it was :-)

not ours if they experienced such issues...  unless they are really clueless on how things are supposed to work.

Igal Sapir
Lucee Core Developer
Lucee.org

--
You received this message because you are subscribed to the Google Groups "Lucee" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lucee+un...@googlegroups.com.
To post to this group, send email to lu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/CAG-7QUu69Hu8mpWrBVyVcgu9Vcf%3D2Uk5jHCYCTLZBoP8W4a7LQ%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[Gert Franz]

And instead of asking for help, they spent month trying to solve the issue themselves...

Gert

Sent from somewhere on the road

[Andrew Penhorwood]

My bet is clueless on how things are suppose to work.  Having worked as a contractor for most of my CFML career the amount of poorly written apps is huge.  During that time most of the people I worked with had never programmed anything before written CFML code.

Andrew Penhorwood

[Andrew Dixon]

Where these not the guys who were part of Adobe's Keynote at cf.objective last year saying how wonderful cfclient was?

Kind regards,

Andrew

about.me
mso - Lucee - Member

To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/016ce3cb-bbdb-404a-8fbe-ff14da40d90a%40googlegroups.com.

[Kai Koenig]

Correct. That's telling enough, isn't it? :)

To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/CAG1WijXUVyw3EY1sV7KAHxxW_6qFLvWgzoLOP7MhFWHBUK0f4A%40mail.gmail.com.

[Andrew Dixon]

Indeed it is!!!

Kind regards,

Andrew

about.me
mso - Lucee - Member

To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/1C82957D-5B7E-403D-9739-55C58D24E081%40gmail.com.

[Denard Springle]

I wonder which open source engine it was :-)

not ours if they experienced such issues...  unless they are really clueless on how things are supposed to work.

I know and have worked with the developers on the team that developed that project. I can say, with all confidence, that they are not clueless developers. It's been a couple/few years since this all went down, so from what I remember it was an early release of Railo, and they did have scalability issues at the time. They also did reach out to the Railo team at the time from what I recall and were still unable to resolve all the issues. The move to ACF was not without its own share of issues, either - scalability still being one of them. I wasn't (officially) part of the project so I'm speaking purely from memory and conjecture, but from what I recall the move to ACF was done reluctantly. But, at the end of the day, it ended up being the right move for that project, long-term, for lots of reasons that have nothing to do with which engine it ran on.

That being said, Adobe is misrepresenting the nuance involved, but given this is a marketing piece and not an actual evaluation of the differences between the engines, and is a bit of a hatchet job against 'open source' in general (which Adobe has been doing for years), it's not very surprising. Still, knocking quality developers simply based on this marketing piece seems... I dunno... childish.

-- Denny

[Gert Franz]

Denny,

I honestly can't remember these guys to ever reach out to us. And I tried really hard. Well anyway. If the project is going well, they made the right move...

Gert

Sent from somewhere on the road

--

You received this message because you are subscribed to the Google Groups "Lucee" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lucee+un...@googlegroups.com.
To post to this group, send email to lu...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/c385642b-c01e-47a7-bf23-98ec5bfc7566%40googlegroups.com.

[Igal @ Lucee.org]

knocking quality developers simply based on this marketing piece seems... I dunno... childish.

first of all, I didn't knock anyone.  I presented the logical possibilities based on what I knew.

so I knew from personal experience that Lucee, and before it Railo, is more scalable than ACF because I used to run ACF before I switched to Railo.  when I switched from ACF8 to Railo3 I managed to reduce my web farm by two thirds for one application, simply because Railo required far less resources.

I also knew that these guys didn't come to us for help since I joined Railo, so if that's the case then these events took place a very long time ago.  the Adobe piece made it seem like this just happened now, so the most logical explanation for me was that it wasn't our engine.

if anything -- then Adobe misrepresented the facts to the point where no one could identify the case unless they knew this project by name -- don't blame me for it.

Igal Sapir
Lucee Core Developer
Lucee.org

To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/319D04BD-0774-4F4F-B85D-60A7870AF86D%40rasia.ch.

[Denard Springle]

Gert - It wasn't developed by the company, but by a third party who shall remain nameless (unless they choose otherwise). In any case, this is probably why you don't remember the particular company, because it was a third party you do know.

Igal - same goes for you. However, calling developers clueless is knocking them. Regardless of what you know or don't know... and perhaps that's the point - you don't know anything more than what Adobe told you. You should know to take anything Adobe spews out with a giant grain of salt. Especially a marketing hit piece like this.

That being said, I'm not trying to have a pissing contest with anyone - neither of you or anybody else. I was merely providing some context to the discussion, which seemed to be lacking. My apologies if I stepped on some toes - that wasn't my intent.

-- Denny

[Gert Franz]

No worries Denny,

The only reason I wanted to know it is because I am interested in the problem and if there is a solution. 

But if they are happy now, that's fine.

Gert

Sent from somewhere on the road

To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/f285b7df-716c-4ce8-9b9f-25c1ca8eb875%40googlegroups.com.

[Denard Springle]

Hey Gert,

   Understood. Sadly, like I said, this all happened 2-3 years ago. Why Adobe is bringing it up now I have no clue... probably because it is the only example they could come up with where a company went *from* open source *to* ACF. Either way it's a case of them grasping at straws as far as I'm concerned. The reasons for the actual migration had less to do with technology than with the bias of manager types. I believe the problems could have been resolved given the time and effort... but someone above the pay grade of the developers insisted on going another direction. Like I said originally - the migration was made reluctantly (on the part of the developers) and scalability was still an issue even after the initial migration, but the customer was happier drinking the Adobe kool-aid... and they stroked the checks. In short, I doubt there was anything you guys could have done differently.

-- Denny

[Adam Cameron]

On Saturday, 7 March 2015 16:48:01 UTC, Igal wrote:

knocking quality developers simply based on this marketing piece seems... I dunno... childish.

first of all, I didn't knock anyone.  I presented the logical possibilities based on what I knew.

so I knew from personal experience that Lucee, and before it Railo, is more scalable than ACF because I used to run ACF before I switched to Railo.  when I switched from ACF8 to Railo3 I managed to reduce my web farm by two thirds for one application, simply because Railo required far less resources.

I wonder whether it's worth a bunch of us getting together and knocking out some tests to fire off under load, and compare similarly provisioned & configured Lucee and ColdFusion boxes?

The last few tests I did under load between CF11 and Railo 4.x had CF outperforming Railo, I'm afraid. But... not by a great deal, and the code being tested wasn't exactly run-of-the-mill application code.

Railo used to be a lot (a lot!) faster than CF, but I'm not sure that it automatically is any more. I say Railo instead of Lucee because these tests were 6-8 months ago, and it was Railo I was running. Also I was running on Java 7 not Java 8, so this might change things as well.

I'd like to get together some more real-world tests, in a more controlled environment, and put both through their paces. Then publish the results.

My problem is that whilst I can write (or help write) the code, I have nothing but my shitty laptops to run it on. And that's not a realistic environment to run stuff on for benchmarking. Also I'm not a load tester by trade, so my approach of getting JMeter running lots of threads/users running lots of requests might not be a sensible approach to running the actual tests...?

-- 

Adam

[Peter Boughton]

Maybe there's a JVM-tuned OS image somewhere that can be used?

Or, didn't Adobe release an EC2 image for CF? Could that be used as-is
for CF, but also cloned and re-purposed for Lucee? (I have no idea if
there are licensing issues there, or if EC2 images can be used anywhere
other than on Amazon's servers.)

Obviously whatever is done needs to be completely open and reproducible
by anyone - ideally something like a self-contained ISO that people can
install to VirtualBox/etc to run themselves.

It would be great if this could also be done as [part of] a "here's how
to do performance testing" guide that developers could use as a general
reference for configuring their own setups.