Slashdot Mirror
Talk to the IBM Linux Hackers
We've all heard plenty about IBM's investment in Linux, but we don't hear much from -- or about -- the actual Linux developers at IBM. This interview is not with one person, but with a number of IBM Linux people spearheaded by Dave Hansen, who volunteered to help us with this interview. Of the group responding to your questions, Dave says, "There are more people, but the majority of the group's skills are represented. No surprise that we'll have our responses reviewed before we send them back to you, but we'll try to expedite that.
"A little background:
The group's experience is pretty broad. Most members were Sequent employees who worked on Dynix/PTX before IBM acquired Sequent (we are still mostly based in Beaverton, OR). Not everyone was with Sequent; Matt Dobson and Dave Hansen came into the group last summer, right out of college. A few of our Austin colleagues are long time IBM employees who worked on the AIX kernel before moving to Linux. Ask about anything from the rmap VM, to PTX's crashdump facilities, to life in Portland :)
- Martin Bligh - Large IA32 system VM scalability, Specweb, NUMA
- Dave Hansen - Big Kernel Lock, SMP locking, Specweb99
- Matt Dobson - NUMA API/ Documentation
- Rick Lindsley - SMP/NUMA Locking, Performance Tuning
- Patricia Gaughen - Discontiguous Memory for NUMA
- Bill Irwin - VM/rmap hacker
- Hanna Linder - dcache_lock scalability
- Janet Morgan - I/O Scalability
- Ruth Forrester - database/performance/scalability"
Now that Linux has been ported to run on high-end machines under virtualization, when will we see a kernel tuned for (e.g.) scalability to 64-128 processors natively?
I'm out of my tree just now but please feel free to leave a banana.
Will IBM try to blend aspects of AIX and Linux together, or will the 2 development paths remain discreet?
"I'm just here to regulate funkiness."
ok, this isn't directly related to the linux kernel, but are there any plans to replace CDE with GNOME (like Sun) or other free desktop environment?
Do you even lift?
These aren't the 'roids you're looking for.
What *new* feature(s) are you most excited about developing?
What compiler do you use for the kernel - gcc or xlc?
Do you even lift?
These aren't the 'roids you're looking for.
Care to tell us if there is anything in Linux kernel or other GNU programs that can profit from your previous AIX experience? How big a paradigm shift was it for you to move from AIX to Linux? Were you pleasantly surprised, unpleasantly surprised or not surprised at all with what you saw in Linux kernel? Did you learn anything in Linux that you would like to see implemented in AIX?
You can't handle the truth.
For the women in the group:
What are your opinions regarding the shrinking number of women in the industry? (actually I believe the numbers are rising again in schools)
Linux seems to be drifiting, very slightly and relatively slowly now, towards a place where a couple of companies exert a kind of defacto control over it.
The licenses are still open, but as a practical matter, most of the core development is being done by corporate people that are concentrated in a few companies like RedHat, and to a lesser extent, IBM.
Do we low end users have cause to worry about this? Does IBM worry about the control that a company like RedHat has over IBM's Linux initiative? And is it really possible, as a pratical matter, for technology as complex as GCC to be forked by volunteers?
Is there any work on the file system limitations or software RAID device restrictions (I was told no more than 12 devices per RAID setup)? What changes are being made.
My work is particularly interested in filesystems > 64 TB and RAID with > 20 devices.
IBM will be using linux to help sell their hardware. Other companies have tried this (VA Linux, which owns Slashdot, once had linux hackers on their payroll). Obviously, IBMs hardware is in a different league as an x86 clone, but do you have any thoughts on Open Source business models and their validity? Once the kernel is running smoothly, will you be disposable since the "Open Source community" can continue development for free?
Do you even lift?
These aren't the 'roids you're looking for.
Why does IBM feel that Linux is important to IBM, and how important does IBM feel Linux is to the future of computing in general?
...interesting if true.
Is Linus accepting your changes well? How directly do you submit patches, and what are your experiences on the overall Linux kernel development style?
I think, therefore thoughts exist. Ego is just an impression.
IMHO, IBM makes some of the best mobile hardware out there -- one of the professors I support raves about his ThinkPad 600, that went with him into the Israeli desert for several months and is still running strong, no service required -- but the linux support for that hardware has been, um, erratic at best. Yes, we've been occasionally been able to purchase the odd model with linux preinstalled (usually it's *more* expensive than the comparable model with MicroSoft preinstalled, grr) but an awful lot of the hardware (mini-pci modems, &etc) is rather difficult to drive with a penguin behind the wheel. Why does IBM's linux enthusiasm fade so quickly at the small (physical) end of the hardware scale? Is there momentum underway to change this?
Need a UNIX/Linux/network guru in the Boulde
Do you see any place for Linux on the desktop? If so, will you be able to fend off the PC hardware group better than the OS/2 group did in the past?
Fascism should more properly be called corporatism, since it is the merger of state and corporate power.
As many people here, I am a huge Linux fan, but I am so much so that I am trying to figure out how to get into the professional Linux world when I graduate.
I attend Clemson University and am in the Computer Information System (CS + business) program (and doubled in Political Science). My goal is to become a Linux sys admin, or perhaps some other Linux guru type job. The work that IBM is doing with Linux is also very appealing to me.
So, how did you get your job, and what would you recommend as the path to follow for us geeks just getting started in the professional world as to how to get into Linux? How can I become as entrenched with Linux as the professionals at IBM? I have had two internships (not with IBM, nor with Linux, but with other CS stuff), but how can I get an entry-level job in a Linux intensive environment like IBM? How can said job lead me into a career where I can be deeply involved in the Linux world?
Lawrence Lessig is my personal hero.
I would love to know what distros are popular on your desktops, at work and at home.
Also explaining why you chose that distrobution would be great.
Thanks
When you were starting out as a group, did you encounter a lot of friction and resistance from middle and/or upper management about your wanting to work on Open Source projects for IBM? If so, what did you do to overcome the objections and become the team you are now? I think the answer to this would help a lot of other people in other companies get mainstream acceptance of the idea of OSS in corporate environments.
-----
"You spilled my egg... I needed that egg."
What features do you find linux most lacking in?
(If we don't examine our weaknesses, we will be crippled)Solaris 9 is getting great reviews. Between the strengths of the traditional open source community and IBM's resources, do you see a point in the next several years where you expect Linux to surpass Solaris in all of its core strengths? Or does Solaris have some unique values which will allow Sun to continue to position itself to advantage, at least for some applications? Please answer this as a technical rather than marketing question.
___
"with their freedom lost all virtue lose" - Milton
Most everyone in the world would say that Linux/Unix is not as user friendly as say OSX and Windows/add suffix here.
With that in mind do you guys see this complexity as a bonus or a henderance to all *nixs moving forward, and please let us know why you feel this way.
Neck_of_the_Woods
#/usr/local/surf/glassy/overhead
Now that IBM has endorsed the Free Source movement, how about working on venerable BSD line, upon which AIX is based? It's much closer to the IBM way of doing things and should be more familiar, as well as being a techically strong base on which to do future work.
I'm one of the few people who really enjoyed the OS/2 desktop and it's features. Have any of the former OS/2 developers been contributing to Linux?
:)
Specifically, the user interface and accessability people - OS/2 was very polished - does IBM see a benefit by offering this expertise to the GNOME/KDE projects?
If so, how does this tie into IBM's vision of Linux of the desktop, if you have one?
Here at my job we have quite a few older IBM machines (PPC RISC IBM 7248 and 7006) and I was searching for a Linux distribution that I could install them - I found a few for the 7248s but nothing for the 7006. I'm planning on building a cluster out the old iron, they may not be fast but they are semi-retired and there are a lot of them.
My question is this, along with bringing linux to the newer machines, are there plans to support the older machines?
Ash nazg durbatuluk, ash nazg gimbatul Ash nazg thrakatuluk agh burzum-ishi krimpatul
About a year ago it was announced on the Linux for AS/400 web page that "OS/400 V5R1 will support the first version of Linux for AS/400's." I haven't heard much since then, and there doesn't seem to be much information about Linux on AS/400s on the IBM web page. Can you comment about if, and how, porting of Linux to AS/400 machines is developing? Thanks.
Finding God in a Dog
What can IBM tell us about the company's future plans for voice recognition under Linux, and in particular, if IBM found supporting the ViaVoice SDK economically unviable, is there any chance that they could open source the code so that volunteers could continue to maintain/develop it?
As evidence of the open source community's interest in the SDK, check out the projects in Sourceforge that rely on it. I just hope that IBM doesn't let the Linux ViaVoice SDK go the way of Blender !!!
IBM's C/Set C++ compiler (becoming Visual Age C++) was one of the best C++ implementations of its time for both performance and standards compliance. Now the product's been discontinued for Windows & OS/2 but still exists for AIX & S/390. Is there any hope for Linux support for this compiler?
I've read a great deal about IBM shifting from the old-bad-guy to new-good-guy with their commitment to Linux. So far, the community seems fairly convinced that IBM's intentions are honest and that they are in fact bringing a lot of credibility and respect to open source and Linux. However, historically, IBM attacked the unified UNIX which GNU/Linux is in some ways becoming. As the developers, you are on the front line of IBM's test as a Linux-friendly force. I'd like to ask what efforts you are making to keep this the case.
What's everyone's stance on the eternal Windows versus Linux battle? Are you firmly entrenched in the everything Linux standpoint, or more towards the Windows for idiots and Linux for power users standpoint? In addition to this do you feel that linux will soon be capable of capturing a significant amount of marketshare in the personal desktop market segment, or do you feel that the place for Linux is in server applications?
How are the different flavors (Redhat, SuSE, etc.) of Linux affecting IBM's development efforts? What are the teams' favorite(s)? Why?
Ruger
With all of this investment in Linux, does IBM put it's money where it's PR is and utilise Linux on the desktop and servers worldwide?
When will CATIA on Linux see daylight?
The only shops that i know of that are going to V5 are changing to NT from AIX.
Is Linux support part of IBM's professional services and if so can this be explained in some more detail. I'm sure there are many qualified people who would and could make a lifelong career here, but IBM being the huge monolith it is, makes it difficult to know whose door to knock on.
What steps are you taking (if any) to evangelize Linux on "low-end" PowerPC systems such as the Apple Power Macintosh line? Is there any chance of seeing low-end PowerPC-based motherboards with out-of-the-box Linux support from IBM in the near future?
In general, the Itanium has seen quite poor adoption rates and even Intel/HP admit that their initial public silicon is really only fit for software developers and platform work (due to low performance, and altready established players in the 64-bit field).
Do you think that IPF64 line will see any kind of broad industry adoption? Will it become just like rest of the (non-embedded) processor architectures designed since the x86 -- constantly fighting for 5% of the market? Do you think the AMD Hammer architecture will be a meaningful player in the field?
Since your group is a key player in the Linux on IA64 and GCC for IA64 projects, can you give us any status information about recent developments in the project since the IA-64 Summit?
Thanks,
-Chris
While meeting with a fellow software consultant recently, I queried him on his plans to expand his business. Specifically, I was curious to know what information infrastruture he plans to grow around. He returned a puzzled look. I had asked if he had considered Linux as a possible server platform, to handle internal source code control, email, and file shares. I suddenly diminished in his eyes. He responded as if I had just asked him if he wanted to buy used Yugo. To him, Linux isn't business worthy... isn't a real stable, capable OS... and worse yet... a threat to the software industry!
Linux needs positive marketing. So my question is, what can you do to champion Linux with IBM's giant huge marketing machine? Not every IT person in the world reads Slashdot (I'd venture to say less than 1%), but a large number of IT people read MS marketing. So the perception most IT people have about Linux is through a Microsoft lens.
Check out my podcast: DreamStation.cc Video Game Show
What were the major porting problems the 390? Do many people use it? How has management accepted Linux in the 390 world?
Necessity is the plea for every infringement of human freedom. It is the argument of tyrants; it is the creed of slaves.
I want to konw when IBM will begin to support Linux at an OS for it-self instead like an tools for selling hardware.
/SOHO printer (the only Lexmark printer driver was marked experimental at linuxprinting). By doing this, IBM it's the only company in the world do not offer support for there printer to the Linux Home users.
IBM never release any desktop tools for Linux. The Lotus SmartSuite never hit the Linux Desktop, nor Notes. By doing this, IBM it's the only company with Microsoft do not port they office suite to Linux.
The IBM's Lexmark company never provide any driver for the home
IBM also dicontinue the support for ViaVoice under Linux!
IBM also doesn't support they desktop and notebook hardware under Linux or other free OS! IBM even take the FreeBSD partition number to use it for it's "suppend to disk" on the bios of ThinkPad making them unusable for FreeBSD !
I really think that's IBM only take the Linux part they need now, and don't look at the future.
The grid computing it's right there at our door, and already desktop begin to replace mainframe in batch processing. We are now using lost cpu cycle do process job under the Windows and Linux desktop. If IBM doesn't investigate this oppurtunity (and the Linux Desktop) in the neer future, IBM will just don't get the next wave and will begin to clash !
Ceci n'est pas une Signature !
Does IBM have any plans to create its own Linux distribution? Doing so would certainly generate a great deal of publicity for IBM's Linux strategy, as well as serve to make Linux a household name. What are your thoughts on this?
With high-end pc servers closing the reliability and performance gap, what advantage is their to moving to Linux for S/390?
A example, widely sited, states that great cost savings can be achieved by consolidating 100's if not 1000's of pc servers. This only works, if the servers being replaced are under extremely low work loads. Many argue that a much greater savings could be achieved in consolidating serives on existing x86 systems.
The only example that've heard that makes any sense, is using linux to extend the cabilities of Mainframe OS's (i.e. using linux, to run apache to create on-line reports, replacing printouts).
This Sig. is False.
There are a few debates raging in linux kernel development these days. I was just interested in hearing IBM's take on a few of these subjects:
:)?
[1] Rik van Riel VM (RMAP) or Andrea Arcangeli VM?
[2] Is a new kernel build system needed for the linux kernel?
[3] Modules/InitRAMFS or regular ol' monolithic kernel?
[4] Which journaling file system does IBM see becoming the de facto journaling file system for Linux? In other words, what's so cool about JFS
I apologize for sneaking in a few too many questions.
w o r l d w i d e w e b e r
From the brief bios, and Sequent pedigree, it looks like there is a lot of focus on high-end features like NUMA, async I/O and the like.
Other commercial organizations, notably SGI, are also putting forth effort in those areas. There is actually quite a bit of overlap.
Since these are "open source" projects, do you collaborate with your traditional "enemies" such as SGI and Sun on Linux? What is your management's attitude towards that type of collaboration? If not, do you "look" at the work the others are doing in comparison to what you are doing?
Learning HOW to think is more important than learning WHAT to think.
How will IBM's recently announced layoff affect the development effort around Linux? Will resources shift away from development/support for a free "product", and be added to revenue-generating projects?
What you have done for the IBM-specific (and strategic) technologies is great, but how will your work benefit the average PC user?
In other words, what are the areas of the Kernel where IBM invested resources that x86-based machines will benefit from, in terms of performance, scalability, robustness, etc.?
I'd really love to see Linux JFS and LVM as thoroghly integrated as in AIX so that you can dynamically grow LVs and JFS filesystems while they are mounted and the system is running with users on it. Also to be able to dynamically shrink mounted filesystems would be nice too, but even AIX doesn't have that yet.
Also while we are on the topic of filesystems, how about adding ACL support to Linux JFS like the sgi's XFS has... so that we can have huge globs of JFS space exported via Samba to Windoze client workstations and have all the NT security (right-click menu stuff) work seemlessly so we can rid ourselves completely of NT fileservers.
Why isn't IBM making more of an effort to recruit developers directly from the Linux community, as opposed to hiring people who have very little if any working familliarity with the platform?
Speaking from experience (I worked at Big Blue for about 2 years), the typical new-hire at IBM knows very little about Unix and other core technologies that form the foundation of your company, and its products...I can name several that didn't even know what a network card was, who gained full-time positions as regulars within the company while top-level engineers and other knowledgable employees were getting pink slipped for cost reductions.
There are employees at IBM who are just now learning what a network card is, while thousands of seasoned Linux developers go unemployed.
What does IBM plan to do about the experience gap within the company?
What is the history of linux at IBM? Was there a grass roots thing amongst developers and/or did senior management suddenly see the light, was it a mixture of these? It's surprised me and many others how suddenly and how thoroughly IBM has become penguinized. So a background story from within would be very interesting to read. Thanks.
Of those to whom much is given, much is required.
I don't think I am alone when I say that IMHO near-perfect, generalized MS Office compatibility would be the biggest boost Linux could receive in the desktop front. Everybody knows what IBM has done to help Linux succeed in the server market, from multi-million ad campaigns to huge contributions on the development effort. Could IBM consider developing a standardized library and API to MS Office file handling, which would enable any Linux app to transparently read/write those formats? This is the sort of thing which unpaid hackers have problems developing (witness the uncoordinated efforts of GNOME, OpenOffice and KDE coders), but would be near-trivial to a professional dev team such as IBM's.
As an IBMer with a thinkpad, I'd like to ask a very straightforward, down to business question.
When will I have Linux running on my Thinkpad, supporting my internal IBM needs?
Seems to me that there will be some initial development cost, perhaps a bit of training, but in the end, a Linux machine would be more easily maintianed. Between exporting xterms, allowing support to telnet or ssh in and needing to be root to really screw stuff up, I think there's a great deal of leverage there. I've looked into the C4EB (Client for E-Business, for those not in IBM speak) stuff, but I can't find out how to run my Lotus 1-2-3, Wordpro, Freelance, etc. (including secure tunneling aka working from home) without kludging it. Certainly, it's fit for some needs, but my job places me in a less flexible position.
My background: I'm a hardware engineer, been playing with Linux since Slackware 96 (in 1996) as a user and admin. I won't allow an unsupported distribution, like Debian or Slackware, to touch my machine, as so many IBM specific things make assumptions.
How about a CD (or 4) that I could boot to that could 1) shrink my Win2k partition 2) set aside my hibernate partition so hibernating will stop blowing away my Linux partition 3) install Linux, Gnome or KDE, Wine and integrate Lotus 1-2-3, Wordpro and Freelance. Sure, I could do this on my own, investing the weeks or months that I had when I was a student, but I'm married now and my management won't support a hardware engineer playing with software on that level.
I think it'd be sweet to be able to hit a shortcut that would export to a bot, signal a telnet or whatever, so an automated script could diagnose simple problems and queue for human review if necessary.
Now that IBM is rebulding their federal services
groups, does IBM have any plans to work with Red Hat
for A NIAP Common Criteria evaluation (making it hunky-dory, from a security perspective to use Linux.) How about FIPS 140-2 for the IBM-Linux crypto cards?
Elegance is for tailors. -A. Einstein
I'd heard that IBM was basically running Linux inside of an AIX VM, as opposed to directly on the hardware. Is this true, and if so, what do you have to watch out for in terms of performance tuning in order to deal with the vastly different set of latencies presented by a VM, rather than bare x86 metal?
"In a 32-bit world, you're a 2-bit user. You've got your own newsgroup, alt.total.loser." -Weird Al
It's not just the services aspect that's important, though that's where the actual money is. The strategic importance of Linux to IBM is hard to underestimate, simply because none of IBM's competition can take advantage of Linux the way IBM can. Having a commodity OS with open specifications, and one that works as well as Linux does, can do no harm to IBM but can hurt its various competitors. Linux is everything that OS/2 wanted to be, and more (strategically). In other words, aside from the advantages it affords, Linux is great for IBM because it doesn't destabilize IBM - it provides the missing link.
posting anonymously to keep my job