3 ms·

I know many people mentioned dtrace, ZFS, mdb.. in describing Solaris' merits but it was not a very good OS in 1997 when I first got my hands on it. It was slow

I know many people mentioned dtrace, ZFS, mdb.. in describing Solaris' merits but it was not a very good OS in 1997 when I first got my hands on it. It was slow - the SPARC hardware didn't help and ZFS/MDB/Dtrace did not exist back then. But I think back then for Enterprises Solaris was the most easily accessible platform SPARC hardware came with and it had a good run for a short while. It was only when Linux started to seriously kick its ass that Sun engineers took it upon themselves to do some great work and make Solaris interesting again - I remember they had a policy - if something is slower than Linux it was a bug. I guess it all came in too late (including the Open Source part ) and people did not like being tied to SPARC hardware and CDDL was a buzz kill for whatever reasons.

bcantrill 9y ago
Ha -- isn't that ever the truth! When I came to Sun in 1996, Solaris was really good at only two things: scaling on SMP machines and not crashing while doing it. Morale wasn't necessarily high (the early history of Solaris had been brutal, and it had only been due to the budding success of 2.5 and UltraSPARC that engineers were beginning to feel forward-looking), and it took us several more years just to get the house in order. MDB came out of this period, as well as important primordial tooling like lockstat(1M). But as we were getting the OS to work as advertised, we had bigger ideas brewing. In Bonwick's case, it was ZFS, for me it was DTrace, for Tucker it was zones, for Casper it was least-priv, etc. I know that it looks like it was in response to Linux and/or as a result of the dot-com bust (and it's impossible to say to what degree these were contributing factors), but I can tell you that from the epicenter it felt much more like a natural evolution than a competitive response. That said, I know that the competition didn't hurt, and you're certainly right about Linux finally causing us to take small-systems performance and system call latency seriously!
- blinkingled 9y ago
  > scaling on SMP machines and not crashing while doing it Yep - Only ever had one time when it hosed on us (100 people team, many running JVM based applications and some doing load tests on it - IIRC it was 2.6 days.) And there wasn't a challenger in sight as far as SMP goes(h/w and s/w both).
bhangi 9y ago
In the 90s, it was frequently referred to as 'Slowlaris'. Yeah, in the early days of Solaris, I definitely used to miss SunOS 4.
X86BSD 9y ago
Agreed. When I got into IT professionally in 1990 I tolerated dealing with SunOS. It was still BSD'ish. But then when it went full re^h^h SysV I put a bullet in its head for me personally. Never used it again after that. I think the last time I touched a Solaris box was 2001 @ TWA in KC. Then 9/11 happened. It was never anything special in my eyes. I just remember hating having to try to find things and software for it. No package management, everything had to be compiled from source. Shiver. AND!!! They didn't ship with CC! It SUCKED! They unbundled it. Yeah not many fond memories of it.
- blinkingled 9y ago
  Oh yeah - you had to buy the compiler separately and GCC SPARC wasn't very good at all - I remember the software we sold sucked at performance benchmarks because we had to compile it with GCC! Solaris though was always a great platform to develop Java/JVM apps on.

bcantrill 9y ago

Ha -- isn't that ever the truth! When I came to Sun in 1996, Solaris was really good at only two things: scaling on SMP machines and not crashing while doing it. Morale wasn't necessarily high (the early history of Solaris had been brutal, and it had only been due to the budding success of 2.5 and UltraSPARC that engineers were beginning to feel forward-looking), and it took us several more years just to get the house in order. MDB came out of this period, as well as important primordial tooling like lockstat(1M). But as we were getting the OS to work as advertised, we had bigger ideas brewing. In Bonwick's case, it was ZFS, for me it was DTrace, for Tucker it was zones, for Casper it was least-priv, etc. I know that it looks like it was in response to Linux and/or as a result of the dot-com bust (and it's impossible to say to what degree these were contributing factors), but I can tell you that from the epicenter it felt much more like a natural evolution than a competitive response. That said, I know that the competition didn't hurt, and you're certainly right about Linux finally causing us to take small-systems performance and system call latency seriously!

blinkingled 9y ago
> scaling on SMP machines and not crashing while doing it Yep - Only ever had one time when it hosed on us (100 people team, many running JVM based applications and some doing load tests on it - IIRC it was 2.6 days.) And there wasn't a challenger in sight as far as SMP goes(h/w and s/w both).

bhangi 9y ago

In the 90s, it was frequently referred to as 'Slowlaris'. Yeah, in the early days of Solaris, I definitely used to miss SunOS 4.

X86BSD 9y ago

Agreed. When I got into IT professionally in 1990 I tolerated dealing with SunOS. It was still BSD'ish. But then when it went full re^h^h SysV I put a bullet in its head for me personally. Never used it again after that. I think the last time I touched a Solaris box was 2001 @ TWA in KC. Then 9/11 happened. It was never anything special in my eyes. I just remember hating having to try to find things and software for it. No package management, everything had to be compiled from source. Shiver. AND!!! They didn't ship with CC! It SUCKED! They unbundled it. Yeah not many fond memories of it.

blinkingled 9y ago
Oh yeah - you had to buy the compiler separately and GCC SPARC wasn't very good at all - I remember the software we sold sucked at performance benchmarks because we had to compile it with GCC! Solaris though was always a great platform to develop Java/JVM apps on.