Slashdot Mirror
A Look At Microsoft's 'Mini Internet' For Testing IE
MrSeb writes "With the grandiose bluster that only an aging juggernaut can pull off, Microsoft has detailed the Internet Explorer Performance Lab and its extraordinary efforts to ensure IE9 is competitive and IE10 is the fastest browser in the world. Here are a few bullet points: 128 test computers, 20,000 tests per day, over 850 metrics analyzed, 480GB of runtime data per day, and a granularity of just 100 nanoseconds. The data is reported to 11 server-class (16-core, 16GB of RAM) computers, and the data is stored on a 24-core, 64GB SQL server. The 'mini internet' has content servers, DNS servers, and network emulators (to model various different latencies, throughputs, packet loss)."
Why not just use the real internet?
Beaten by Chrome and Firefox.
Granularity of 100 nanoseconds: What does that mean?
"As you can see, my young apprentice, your friends have failed. Now witness the firepower of this fully ARMED and OPERATIONAL battle station! "
What is this now? MSDN blogs? Seriously, gimme old slashdot back please!
until they add some zombied computers and malware control servers.
oh, wait.. these are windows test systems. never mind. some kind soul probably already added them
And still unable to correctly implement CSS3 and HTML5
I couldn't resist. But with all the work and effort and resources going into this, how is it that operations a tiny fraction of this can generate fast, reliable and standards complaint browsers better than MSIE?
Microsoft, the problem isn't that you're not spending enough money. It's that you're not doing it right.
These massive pages are a real benchmark for any browser. Or Google .. they seem to be logging every page I go to now. Or eBay with all that horrible bloat. Or Facebook, which is seriously clunky with to many competing scripts. ...
If it's their own little internet they should be using some of the most bloated, unresponsive web sites on the internet to test on. I don't think when IE10 comes out I'll be surfing their own tiny little internet.
A feeling of having made the same mistake before: Deja Foobar
16-core, 16GB of RAM computers
They might want to try running IE on smaller/less robust computers. If you need all that to run it, some might argue its s-l-o-w.
It means that each millisecond of ping is divided into over 9000 parts.
Microsoft, Howard Hughes is calling. Yes, he's read the MSDN article on IEPL and he'd really like his Spruce Goose back.
Whatever the result I hope microsoft do well that in turn will push competitors and we the users should hopefully benefit. Though I feel IE has a long way to go..
Wasn't this a plot-point in the Hitch-Hiker's Guide to the Galaxy series? Having an artificial universe on-site so they could go exploring but still be able to come back for long lunches...
Al Gore is going to be *pissed* when he hears about this.
To mention the in-house installs of SWEN, TDSS, Melissa, and ILOVEYOU.
Silence is a state of mime.
"And the Chevy Vega was thoroughly tested for millions of miles before being released to the public." GM has spoken!!
Sorry, but gray text on gray background is making my eyes bleed.
If Mozilla was doing half this to test their bloated piece of crap, you'd all be shitting your pants with glee.
...and they still can't see why users hate their software.
Nobody question's commitment to software quality and testing methodology. (You have to admit that they're pretty good about that nowadays. This isn't IE3 on NT4)
Microsoft's attitude towards web standards, web technologies, and their even continued ambition to make the internet microsoft-centric are what make IE a significantly less useful tool to your average end user.
Firefox has pretty much unmatched flexibility and extensibility.
Chrome is fast, simple, and very secure, and takes care of itself.
Hell, even opera is more useful (Opera is the most underrated browser ever)
It's also really really hard to shake the idea that microsoft really doesn't care about HTML5 or any other emerging web standards. The problem is, they don't seem to care much about their own techs. Silverlight and company are pretty much only seriously used when microsoft makes some back room deals with other large companies.
Meanwhile firefox and chrome (and apple with safari) are busy trying to build tech and standards that will be used in the next generation of web applications. All microsoft provides is another god damn hurdle in testing your site to make sure it works with the blue E.
On the business software/Enterprise side, IE shines. Why? Because you can publish policies and settings that configure just about every behavior and aspect of everything that IE does and chose weather to enforce or simply make them default. This is a godsend to anyone who has to manage more than a handful of machines. (Oh hey, we need to turn on or of x obscure feature for just this address because X reason.. For 5,000 workstations) Firefox and Chrome claim to have this (Publishing, enforcing settings via AD) but in practice their implementations fall very short of IE.
Does that mean they have only porn sites with midget porn? And a mini 4chan, populated with toddlers?
Then technically 40% of its traffic is pure Pr0n.
... why we need IPv6. ;-)
...and Internet Explorer still can't render things right.
Oh the hilarity!
By the way, does the current IE still assume a fixed width per letter when rendering buttons, so that buttons-with-very-long-texts-on-them render with extra space to the left and right?
Hey don't blame me, IANAB
can it run Crysis?
That really isn't that many machines for testing. When I was testing PART of the components for the SQL team we had hundreds of physical machines (500+) and thousands of virtual machines (2000+) running daily, weekly and on-demand tests. Of course the main thing that escaped them was that there was no way for the dev team to check all those results, so most were useless.
...to prove how quickly it fails the ACID3 test?
One thing I'm wondering - given how in Windows 7 and beyond, MS seems to be making IPv6 the default home networking protocol, any idea whether this mini internet they are experimenting w/ is an IPv6 internet, or an IPv4 intranet (likely w/ private addresses?)
Seems to me that that's what they should be doing. That way, they don't have to duplicate their efforts later to verify IPv6 compatibility.
They aren't seeing all sorts of malware, trijans, Adverts and ISP Throttling on their 'mini internet'.
therefore it is not indicative of what users will experience.
What about all the sites with 20+ hit counters etc? How does it work then? We need to know this.
...is apparently barely enough to run IE10, if I understand it correctly. Now that being Microsoft, could anyone be surprised?
Is the editorializing in the lead sentence of the article really necessary? If you have an opinion to share; state the fact and save the editorial for a concluding sentence.
As it is written, my reflex is to discount the entire post as biased crap.
Forget about this, what are they doing to make email HTML rendering standards-compliant? Seriously, Outlook 2007 is a bigger pain in the a$$ than IE6 was!
With each version of IE, we're being told that this is it. And after using it for a few seconds, I always end up going back to Firefox
... and IE still sucks ass after all those test, gadgets and hardware. I'll stick with eithe chrome or firefox thank you.
Have gnu, will travel.
Well, that's all fine and dandy... but shouldn't they make it standards compliant first? What a waste of time!
I've noticed when connecting my ADSL Router in LAN the Windows is a lot lot lot slower. I mean often seconds slower than my several years older Ubuntu box. The Ubuntu box loads up the pages in Router almost instantly.
It's something in Windows, regardless of the browser it just happens. I've noticed the problem with Windows Vista and Windows 7, can't remember if the problem existed in Windows XP.
While slashdot mocks the computer industry marketing for describing computers using a single metric, you seem to be quite happy with that when it comes to browser performance.
An example: Chrome (v8 engine) has this reputation for amazing speed, but IE9 absolutely grinds Chrome into the dust when it comes to simply repositioning elements on screen; something which today's web apps spend a lot of their time doing. You can feel it too if you know what you're looking for. I don't follow IEs development as closely as Chrome or Firefox, but IE must be hardware accelerating these translations.
I fully expect Google to focus on performance cases which help their specific apps. Again, a conflict of interest, akin to Microsoft pre-caching masses of junk, so that Office can appear to start up much faster than the competition.
that must suck to license that server, especially with the new model coming out......
I am all for this testing, but in practice I think it will lack one very important metric that Microsoft can not measure. How well IE10 works with a site riddled with poorly written code.
I am Bennett Haselton! I am Bennett Haselton!
MSDN blogs are often very technically detailed, written by people who know this stuff from the inside, and if it's about a topic that's of general computing interest then it seems that's a good thing. And the blog in question is chock full of some really good detailed stuff about how they're doing performance testing, reasons why the lab is architected the way it is, detailed graphics on how they measure performance, how they analyze it...on and on.
Frankly, this seems more akin to old Slashdot than a lot of the nonsense we see here today. (That story the other day about a girl sent home from school because her lunch wasn't healthy, and then quickly called into question over what happened? Really? What was that topic even doing on Slashdot in the first place?) Whatever you think about Microsoft, having this extremely detailed look into how one of the world's biggest software vendors (or are they the biggest now?) goes about performance testing, and how they ensure consistent results, should be really, really interesting to anyone involved in IT.
Per my subject-line: They're both more highly optimized than std. builds of FireFox are (specifically for Windows OS) by using more "radical" optimization options than the std. builds of FF do!
Links for downloads:
---
WATERFOX: (64-bit highly optimized FireFox)
http://waterfoxproj.sourceforge.net/
---
PALEMOON: (32 & 64-bit highly optimized FireFox)
http://www.palemoon.org/
---
* Enjoy - &, I hope that helps your issues with FireFox being "slow" etc/et al...
APK
P.S.=> I use WaterFox here myself on Windows 7 64-bit, & I even like it (although Opera's my fav. & in a 64-bit alpha build too no less) - it's my "2nd fav. browser" here in fact!
... apk
Got SOASTA, Microsoft? Got a cloud for your test Internet?
It's just awes... Waiting on cache...
Chrome team did no need any of these to build hell of a faster browser then IE9..
probably MS should have asked them for tip..
... well, let's hope that doesn't make it into the requirements to run the thing.
IE10 will be great! Great software, great test lab.
Guys, will it work on my Ubuntu???!
the real internet has virusez ;)
alternatively, they could focus on writing good algorithms....
Why does Microsoft spend tens of millions of dollars on Trident when they could not just adopt Webkit like everyone else? Then their improvements would help EVERYONE.
I totally understand why Microsoft wants to have it's own browser in Windows. I *DO NOT*, nor will I ever, understand why it needs to have it's own ground-up rendering engine.
NIH syndrome is certainly the main culprit here, I have NO doubt.
Too late, the very article summary itself is obviously written by an aging MS-hating neckbeard.
Grandiose Bluster? Microsoft's developers are simply proud of their work. Why make a petty comment about a viable and competent testing environment?
Oh, who am I kidding. I know I'm a lamer because I don't log in.
... apk
128 test computers, 20,000 tests per day, over 850 metrics analyzed, 480GB of runtime data per day, and a granularity of just 100 nanoseconds.
What is supposed to be impressive about this? 128 is not many computers, 20,000 tests is dog slow, 480 GB of data is just feeble, and I would be embarrassed to admit ever timing to a granularity as crude as 100 ns. 1/3 ns is more like it.
This is a puff piece and just serves to reinforce my impression that Microsoft's engineering culture died long ago. It fell down and can't get up.
As for the chance of Microsoft coming up with the world's fastest browser, it isn't going to happen, sorry, You actually need a lot of skilled, dedicated software artists working tirelessly to make that happen. In the lower galleys at Microsoft, mercilessly whipped by HR and beancounters, sneered at by legions of fat and happy partners, always waiting for that dread U10, Microserfs just don't have the right stuff. Not now, or ever again.
Have you got your LWN subscription yet?
They want it badly now.
Just give Mozilla and Opera more money to be the default search engine in there. No need to reinvent what opera already did.
I came here to say this (except with an 's' instead of 'z').
My main concern about IE7,8 and 9 is not about the render speed. Is the speed of the program itself. It takes much more time to startup, to open new tabs, to do ANYTHING. I cannot notice if a page render more milliseconds faster than on Firefox, but bothers me when I want a new tab and this take seconds.
Yep, things are getting out of hand in browser land:
user@host:/tmp/lynx1998$ time make -j3
blah blah blah...
Welcome to Lynx!
make[1]: Leaving directory `/tmp/lynx2-8-1/src'
real 0m14.920s
user 0m41.457s
sys 0m1.983s
user@host:/tmp/lynx2-8-7$ time make -j3
[...]
Welcome to Lynx!
make[1]: Leaving directory `/tmp/lynx2-8-7/src'
real 0m49.232s
user 0m46.933s
sys 0m2.085s
And that's on an i5! Anyone know a good gopher client?
Why the hell does M$ even *care* about IE? I can't see how it improves their bottom line, or anything..why not just sub with, say, Chrome, or so.. (Although i wouldn't be surprised if they signed up Opera instead of a (IMHO) better one ;-(
that's a tiny testbed for PCs IMO.