13 ms·

Technically speaking, they were quite far ahead of the curve. While VMs were all the rage, they were shipping container-based products (their Sun refugees kept

Technically speaking, they were quite far ahead of the curve. While VMs were all the rage, they were shipping container-based products (their Sun refugees kept developing Solaris Zones and ZFS in this context). The technical abilities of the Joyent folks were well beyond just about anybody else for a long time. One could argue they were too early. For several years they had the technology to beat Amazon, until Amazon's massive scale washed over the whole market. The founders did well, so it's not like they "lost", but it's kinda sad to see a technically visionary set of products reach a dead end (one could maybe even view this as the final commercial death of what Sun built). Also worth noting, their stack is open source in ways that nobody else is, I think. Maybe Google is doing right with Kubernetes, but everybody else is consuming OSS software to build proprietary clouds and giving basically nothing back. Joyent were doing most of the work, and giving most of the tech away.

kbenson 7y ago
> it's kinda sad to see a technically visionary set of products reach a dead end (one could maybe even view this as the final commercial death of what Sun built). I think it's not really the end of the technology, but it is an end to a generally available public commercial offering of it. It sounds like they're still offering services, but only for "single-tenant", which after some careful reading (because I wasn't sure how to interpret that initially), I think means that they'll provide their software and managemest system for private install for large orgs in their private deployments, maybe along with a support/management contract?
- SwellJoe 7y ago
  That's basically putting technology out to pasture. When you've got a handful of large corporate customers that depend on a product, you can continue to extract money out of them for years or decades without the product having a future.
  - kbenson 7y ago
    The question is whether they are still actively pursuing enterprise clients. If they aren't, then I agree. If they are, then it's not putting it out to pasture, it's just a different business model. There are plenty of companies that do this. My brother works for one that does it for large network storage deployments using proprietary technology, and I imagine they don't want to hear from you unless you are willing to put out hundreds of thousands to millions of dollars on a deployment (and yet they still have new deployments). I definitely wouldn't consider them putting the technology out to pasture because of that.
zzzcpan 7y ago
> Technically speaking, they were quite far ahead of the curve. While VMs were all the rage, they were shipping container-based products > The technical abilities of the Joyent folks were well beyond just about anybody else for a long time. None of that was ever true. Whether they had container-based or VM-based tech was a tiny unimportant detail for a cloud company. Although over engineering in this area certainly made it impossible for them to compete with more traditional hosting companies. And to be able to compete with Amazon they needed something they didn't have - distributed systems background, not Sun background, which was anything but.
- SwellJoe 7y ago
  "And to be able to compete with Amazon they needed something they didn't have - distributed systems background, not Sun background, which was anything but." lol. "The network is the computer."
- kelp 7y ago
  I'm reminded of something a smart engineer said at work today. "Our customers don't care how we implement the system, they just care that it works"
  - insulanian 7y ago
    That's how you end up with unmaintainable solutions and code-bases which die under the cost of just trying to keep the system running, and because they lose the ability to evolve.
    - kelp 7y ago
      I guess I should have provided more context. It was in context of talking about a rewrite into another language to make something more maintainable and performant. To explain how hard that was to get right, and that if we messed it up, customers only really care that we messed it up.
- Annatar 7y ago
  Sun Microsystems literally invented the network is the computer. They were the first company ever to even begin to think in terms of distributed systems on a network scale (DEC had done some pioneering work on the hardware front prior to that with VMS, but not the network). The network file system, NFS, exists because Sun Microsystems engineers willed it so -- they wanted to multiply the computational power of smaller systems into a larger, distributed one. They were a pioneer in distributed systems computing. Not many computer companies can claim that. Even fewer companies can claim engineers which hold patents which now power the InterNet, like Radia Perlman does for the spanning tree routing protocol. If that's not distributed, I don't know what is.
pron 7y ago
I don't know if it was the immediate cause of their lack of success, but they had also brought with them from Sun the Sun attitude of doing stuff that their engineers thought was cool (and it probably was) without paying attention to how much people are willing to pay for it and whether their focus is correct from a market perspective. For example, Bryan Cantrill, Joyent's CTO, liked blaming Oracle's Larry Ellison for the demise of Solaris, on which Cantrill had worked, but by the time Ellison acquired the failing Sun, Solaris, cool technology and all, had been pretty much dead or dying for a while -- at the hands of Linux, and not much helped by the Sun attitude. Cantrill was furious and blamed Ellison of being monomaniacal about making money ("don't anthropomorphize Larry Ellison," he said) but perhaps it was Cantrill who was monomaniacal all along, not keeping his eyes on the market. Then again, luck plays such an important role in these things that nice narratives rarely give the whole picture.
- bcantrill 7y ago
  To be clear, I have never blamed Ellison for the demise of Sun or even Solaris -- but I do blame Oracle for re-proprietarizing OpenSolaris, and thereby ripping up the social contract of copyright assignment in an open source project. If it must be said, this wasn't an isolated incident -- and in hindsight, was merely the front of a conga line bad Oracle behavior. As for my own putative ignorance of market demands, one of Sun's growth areas at the time of acquisition in terms of revenue was the group that I had co-founded within Sun[1][2] -- in part because we had listened carefully to the market, and built a product that it very much wanted. Finally, if we want to engage in the perennial musing as to the demise of Sun, I think for my part I still broadly stand by my analysis from 2011.[3] [1] http://dtrace.org/blogs/bmc/2008/11/10/fishworks-now-it-can-be-told/ http://dtrace.org/blogs/bmc/2008/11/10/fishworks-now-it-can-... [2] http://dtrace.org/blogs/bmc/2010/07/25/good-bye-sun/ http://dtrace.org/blogs/bmc/2010/07/25/good-bye-sun/ [3] https://news.ycombinator.com/item?id=2287033 https://news.ycombinator.com/item?id=2287033
- kragen 7y ago
  Solaris's death at the hands of Linux can be attributed in part not just to "the Sun attitude" but specifically to the attitude of Bryan Cantrill personally. In response to Dave Miller explaining how Linux's networking was technically superior (at that moment) due to more care for things like TLB footprints and system call overhead, and also had more accessible technical support, Cantrill famously decided to respond by making fun of Miller for being a nerd: "Have you ever kissed a girl?" https://www.osnews.com/story/28261/have-you-ever-kissed-a-girl/ https://www.osnews.com/story/28261/have-you-ever-kissed-a-gi... https://web.archive.org/web/20110721130404/https://cryptnet.net/mirrors/texts/kissedagirl.html https://web.archive.org/web/20110721130404/https://cryptnet.... Admittedly, this was in 1996, and it's possible Cantrill regretted this remark, or the nerd-hating attitude behind it, at some point in the intervening 23 years. But I've never seen him admit it. I see he's participating in this thread, so if you see him not replying to this message, that's him doubling down.
kenned3 7y ago
This. Well put. when eveyone was hooked on VM's and other nonsense they had 'branded zones' running. The learning curve was high, but worth it once you had it running. Sad to see them leave. wonder what i will do with my SmartOS based media server now?
- jclulow 7y ago
  For what it's worth, we're still maintaining SmartOS and it's still open source!
kelp 7y ago
Just as a side note here, Google is doing Kubernetes as a commoditize your compliment strategy (https://www.gwern.net/Complement https://www.gwern.net/Complement) so they can sell other cloud compute services. This makes it easier for people to migrate their on-prem workloads to Kubernetes, and then to the cloud. This commoditize much of what Amazon was attempting with ECS. Forcing Amazon to fast follow with EKS, arguably a worse and much more complex to deploy product than Google GKE. It makes it so it's hard to make money off container orchestration, but you can still make money off what Google is good at, or at least thinks they are good at. Selling compute, proprietary databases, and stuff like TPUs. My point is that yes, Google is advancing the whole industry by releasing Kubernetes as opensource, but it's also a strategy to sell more of GCP and deny some revenue to competitors. It remains to be seen if it works. GCP is still a distant 3rd in the cloud race and has a long way to go to improve their enterprise support. (Their usability is much nicer though...)
Annatar 7y ago
"but it's kinda sad to see a technically visionary set of products reach a dead end" It's not a dead end since any company, public or private can use the code and the product (I for instance build my own modified SmartOS from source) to power their infrastructure. Technically there is nothing to prevent anyone from taking the product and the code and starting the next Amazon. I know of at least one internet service provider in the UK who is leveraging SmartOS to provide commercial, for-profit service. There are probably more such businesses, but this is something which squarely falls under "I'm keeping my mouth shut and making money" category.
- insulanian 7y ago
  What would you state as the top reason(s) to choose SmartOS vs Linux for those use-cases?
  - Annatar 7y ago
    1. full-blown virtual UNIX servers running at bare metal speeds (joyent-branded zones). 2. ability to run virtual Linux servers are bare metal speeds (and their applications). 3. simple and easy container management of 1. and 2. with imgadm and vmadm. 4. OpenZFS - easy capacity growth, checksummed data, silent data corruption detection and self healing. 5. lightning fast, faster than GNU/Linux performance on the same hardware. 6. same application portfolio as GNU/Linux (14,000 applications available with a simple pkgin command invocation). 7. ability to create virtual routers and switches out of thin air with dladm, join and partition physical links in terms of percentages, simplifying network topology in virtualized environments. 8. OS paranoid about protecting and self-healing itself and the data it hosts. 9. ability to pick between KVM (slower) or bhyve for paravirtualization of non-Linux OS's like Microsoft Windows, OpenBSD, FreeBSD et cetera (although this process is poorly documented). 10. NFS V3, V4, fiberchannel and iSCSI which work correctly. There is one major downside (I don't consider it as such, but I know others do): one must pick the hardware compatible with SmartOS from the Joyent's hardware compatibility list, not the other way around: https://eng.joyent.com/manufacturing/bom.html https://eng.joyent.com/manufacturing/bom.html. But when it does fire up over the network and once one groks it, holy shit, it's like being taken into the future with a time machine.

kbenson 7y ago

> it's kinda sad to see a technically visionary set of products reach a dead end (one could maybe even view this as the final commercial death of what Sun built). I think it's not really the end of the technology, but it is an end to a generally available public commercial offering of it. It sounds like they're still offering services, but only for "single-tenant", which after some careful reading (because I wasn't sure how to interpret that initially), I think means that they'll provide their software and managemest system for private install for large orgs in their private deployments, maybe along with a support/management contract?

SwellJoe 7y ago
That's basically putting technology out to pasture. When you've got a handful of large corporate customers that depend on a product, you can continue to extract money out of them for years or decades without the product having a future.
- kbenson 7y ago
  The question is whether they are still actively pursuing enterprise clients. If they aren't, then I agree. If they are, then it's not putting it out to pasture, it's just a different business model. There are plenty of companies that do this. My brother works for one that does it for large network storage deployments using proprietary technology, and I imagine they don't want to hear from you unless you are willing to put out hundreds of thousands to millions of dollars on a deployment (and yet they still have new deployments). I definitely wouldn't consider them putting the technology out to pasture because of that.

zzzcpan 7y ago

> Technically speaking, they were quite far ahead of the curve. While VMs were all the rage, they were shipping container-based products > The technical abilities of the Joyent folks were well beyond just about anybody else for a long time. None of that was ever true. Whether they had container-based or VM-based tech was a tiny unimportant detail for a cloud company. Although over engineering in this area certainly made it impossible for them to compete with more traditional hosting companies. And to be able to compete with Amazon they needed something they didn't have - distributed systems background, not Sun background, which was anything but.

SwellJoe 7y ago
"And to be able to compete with Amazon they needed something they didn't have - distributed systems background, not Sun background, which was anything but." lol. "The network is the computer."
kelp 7y ago
I'm reminded of something a smart engineer said at work today. "Our customers don't care how we implement the system, they just care that it works"
- insulanian 7y ago
  That's how you end up with unmaintainable solutions and code-bases which die under the cost of just trying to keep the system running, and because they lose the ability to evolve.
  - kelp 7y ago
    I guess I should have provided more context. It was in context of talking about a rewrite into another language to make something more maintainable and performant. To explain how hard that was to get right, and that if we messed it up, customers only really care that we messed it up.
Annatar 7y ago
Sun Microsystems literally invented the network is the computer. They were the first company ever to even begin to think in terms of distributed systems on a network scale (DEC had done some pioneering work on the hardware front prior to that with VMS, but not the network). The network file system, NFS, exists because Sun Microsystems engineers willed it so -- they wanted to multiply the computational power of smaller systems into a larger, distributed one. They were a pioneer in distributed systems computing. Not many computer companies can claim that. Even fewer companies can claim engineers which hold patents which now power the InterNet, like Radia Perlman does for the spanning tree routing protocol. If that's not distributed, I don't know what is.

pron 7y ago

I don't know if it was the immediate cause of their lack of success, but they had also brought with them from Sun the Sun attitude of doing stuff that their engineers thought was cool (and it probably was) without paying attention to how much people are willing to pay for it and whether their focus is correct from a market perspective. For example, Bryan Cantrill, Joyent's CTO, liked blaming Oracle's Larry Ellison for the demise of Solaris, on which Cantrill had worked, but by the time Ellison acquired the failing Sun, Solaris, cool technology and all, had been pretty much dead or dying for a while -- at the hands of Linux, and not much helped by the Sun attitude. Cantrill was furious and blamed Ellison of being monomaniacal about making money ("don't anthropomorphize Larry Ellison," he said) but perhaps it was Cantrill who was monomaniacal all along, not keeping his eyes on the market. Then again, luck plays such an important role in these things that nice narratives rarely give the whole picture.

bcantrill 7y ago
To be clear, I have never blamed Ellison for the demise of Sun or even Solaris -- but I do blame Oracle for re-proprietarizing OpenSolaris, and thereby ripping up the social contract of copyright assignment in an open source project. If it must be said, this wasn't an isolated incident -- and in hindsight, was merely the front of a conga line bad Oracle behavior. As for my own putative ignorance of market demands, one of Sun's growth areas at the time of acquisition in terms of revenue was the group that I had co-founded within Sun[1][2] -- in part because we had listened carefully to the market, and built a product that it very much wanted. Finally, if we want to engage in the perennial musing as to the demise of Sun, I think for my part I still broadly stand by my analysis from 2011.[3] [1] http://dtrace.org/blogs/bmc/2008/11/10/fishworks-now-it-can-be-told/ http://dtrace.org/blogs/bmc/2008/11/10/fishworks-now-it-can-... [2] http://dtrace.org/blogs/bmc/2010/07/25/good-bye-sun/ http://dtrace.org/blogs/bmc/2010/07/25/good-bye-sun/ [3] https://news.ycombinator.com/item?id=2287033 https://news.ycombinator.com/item?id=2287033
kragen 7y ago
Solaris's death at the hands of Linux can be attributed in part not just to "the Sun attitude" but specifically to the attitude of Bryan Cantrill personally. In response to Dave Miller explaining how Linux's networking was technically superior (at that moment) due to more care for things like TLB footprints and system call overhead, and also had more accessible technical support, Cantrill famously decided to respond by making fun of Miller for being a nerd: "Have you ever kissed a girl?" https://www.osnews.com/story/28261/have-you-ever-kissed-a-girl/ https://www.osnews.com/story/28261/have-you-ever-kissed-a-gi... https://web.archive.org/web/20110721130404/https://cryptnet.net/mirrors/texts/kissedagirl.html https://web.archive.org/web/20110721130404/https://cryptnet.... Admittedly, this was in 1996, and it's possible Cantrill regretted this remark, or the nerd-hating attitude behind it, at some point in the intervening 23 years. But I've never seen him admit it. I see he's participating in this thread, so if you see him not replying to this message, that's him doubling down.

kenned3 7y ago

This. Well put. when eveyone was hooked on VM's and other nonsense they had 'branded zones' running. The learning curve was high, but worth it once you had it running. Sad to see them leave. wonder what i will do with my SmartOS based media server now?

jclulow 7y ago
For what it's worth, we're still maintaining SmartOS and it's still open source!

kelp 7y ago

Just as a side note here, Google is doing Kubernetes as a commoditize your compliment strategy (https://www.gwern.net/Complement https://www.gwern.net/Complement) so they can sell other cloud compute services. This makes it easier for people to migrate their on-prem workloads to Kubernetes, and then to the cloud. This commoditize much of what Amazon was attempting with ECS. Forcing Amazon to fast follow with EKS, arguably a worse and much more complex to deploy product than Google GKE. It makes it so it's hard to make money off container orchestration, but you can still make money off what Google is good at, or at least thinks they are good at. Selling compute, proprietary databases, and stuff like TPUs. My point is that yes, Google is advancing the whole industry by releasing Kubernetes as opensource, but it's also a strategy to sell more of GCP and deny some revenue to competitors. It remains to be seen if it works. GCP is still a distant 3rd in the cloud race and has a long way to go to improve their enterprise support. (Their usability is much nicer though...)

Annatar 7y ago

"but it's kinda sad to see a technically visionary set of products reach a dead end" It's not a dead end since any company, public or private can use the code and the product (I for instance build my own modified SmartOS from source) to power their infrastructure. Technically there is nothing to prevent anyone from taking the product and the code and starting the next Amazon. I know of at least one internet service provider in the UK who is leveraging SmartOS to provide commercial, for-profit service. There are probably more such businesses, but this is something which squarely falls under "I'm keeping my mouth shut and making money" category.

insulanian 7y ago
What would you state as the top reason(s) to choose SmartOS vs Linux for those use-cases?
- Annatar 7y ago
  1. full-blown virtual UNIX servers running at bare metal speeds (joyent-branded zones). 2. ability to run virtual Linux servers are bare metal speeds (and their applications). 3. simple and easy container management of 1. and 2. with imgadm and vmadm. 4. OpenZFS - easy capacity growth, checksummed data, silent data corruption detection and self healing. 5. lightning fast, faster than GNU/Linux performance on the same hardware. 6. same application portfolio as GNU/Linux (14,000 applications available with a simple pkgin command invocation). 7. ability to create virtual routers and switches out of thin air with dladm, join and partition physical links in terms of percentages, simplifying network topology in virtualized environments. 8. OS paranoid about protecting and self-healing itself and the data it hosts. 9. ability to pick between KVM (slower) or bhyve for paravirtualization of non-Linux OS's like Microsoft Windows, OpenBSD, FreeBSD et cetera (although this process is poorly documented). 10. NFS V3, V4, fiberchannel and iSCSI which work correctly. There is one major downside (I don't consider it as such, but I know others do): one must pick the hardware compatible with SmartOS from the Joyent's hardware compatibility list, not the other way around: https://eng.joyent.com/manufacturing/bom.html https://eng.joyent.com/manufacturing/bom.html. But when it does fire up over the network and once one groks it, holy shit, it's like being taken into the future with a time machine.