It's a UNIX system!

The all-seeing eye of DTrace

noreply@blogger.com (Henkis) — Tue, 03 Jan 2012 15:11:00 +0000

I was recently involved with a problem related to backup software running on Solaris, as part of a general health check of the system I stumbled on something interesting that was not visible using conventional tools.

This tuned out to be a good opportunity to put my DTrace skill to work together with a few finished scripts. Once again it struck what how amazing this tool is, you can really see everything that is going on in your system and as it turns out, you can even see problems that does not even exist. Since this was so much fun and a good example I will walk through the steps again:

The thing that caught my eye was the output from errinfo of the DTrace toolkit. There are a very high rate of system calls returning in error, namely close() with -9 "Bad file number", as seen with errinfo:

whoami        ioctl   22     13  Invalid argument                 
init          ioctl   25    212  Inappropriate ioctl for device   
awk           stat64    2    520  No such file or directory
java          lwp_cond_wait   62   3492  timer expired                                 
processx      close    9 102073391  Bad file number

Syscall errors are in itself normal can be seen on any systems, but usually not several thousand per second. As the error message indicates this happens when a close() is issued on a file handle (Integer) that does not represent an open file for that process, which at first look seems like a quite useless operation.

We can also see that close() is by far the most used system call here:

# dtrace -q -n 'BEGIN { close=0;total=0 } syscall::close:entry \
{ close = close + 1 } syscall:::entry  { total = total + 1 } END \
{ printf("%d close calls of %d total calls\n",close,total) }'

309530 close calls of 426212 total calls

Looking at which file descriptor the process is trying to close shows that there is an even distribution of close between 0 and 65536 and the only successful calls where to numbers lower than 1024 where numbers normally used unless a process has a very high amount of open files.

# dtrace -n 'syscall::close:entry { this->fd = arg0 } syscall::close:return \
/ errno!= 0 / { @failed = lquantize(this->fd,0,65536,16384) } syscall::close:return \
/errno == 0/ { @good = lquantize(this->fd,0,65535,1024) }
dtrace: description 'syscall::close:entry ' matched 3 probes
           value  ------------- Distribution ------------- count    
             < 0 |                                         7        
               0 |@@@@@@@@@@@@@                            414811   
           16384 |@@@@@@@@@                                294912   
           32768 |@@@@@@@@@                                294912   
           49152 |@@@@@@@@@                                294912   
        >= 65536 |                                         0        

           value  ------------- Distribution ------------- count    
             < 0 |                                         0        
               0 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 12459
            1024 |                                         0

The processes responsible only lives only a short while, but by using dtruss i could trace system calls based on the process name:

 15889/1:  fork1()   = 0 0
 15889/1:  lwp_sigmask(0x3, 0x0, 0x0)   = 0xFFFF 0
 11385/1:  getpid(0x0, 0x1, 0x1CD0)   = 15889 0
 11385/1:  lwp_self(0x0, 0x1, 0x40)   = 1 0
 11385/1:  lwp_sigmask(0x3, 0x0, 0x0)   = 0xFFFF 0
 11385/1:  fcntl(0xA, 0x9, 0x0)   = 0 0
 11385/1:  schedctl(0xFFFFFFFF7F7361B8, 0xFFFFFFFF7F738D60, 0x11A340)
 = 2139062272 0
 11385/1:  lwp_sigmask(0x3, 0x0, 0x0)   = 0xFFFF 0
 11385/1:  sigaction(0x12, 0xFFFFFFFF7FFDEE20, 0x0)   = 0 0
 11385/1:  sigaction(0x2, 0xFFFFFFFF7FFDEE20, 0x0)   = 0 0
 11385/1:  sigaction(0xF, 0xFFFFFFFF7FFDEE20, 0x0)   = 0 0
 11385/1:  getrlimit(0x5, 0xFFFFFFFF7FFDED90, 0x0)   = 0 0
 11385/1:  close(0x3)   = 0 0
 11385/1:  close(0x4)   = 0 0
 11385/1:  close(0x5)   = 0 0
 11385/1:  close(0x6)   = 0 0
....
 11397/1:  close(0xFFFF)   = -1 Err#9

The process is issuing close on all numbers between 0x3 to 0xFFFF in a loop, as expected the first few are actually open and closed correctly but the other was majority is returning error -9.

If we look in the beginning of the trace we can see a fork, followed a little later by getrlimit(0x5,...), if we look at what that arguments to getrlimit means:

# egrep "RLIMIT.*5" /usr/include/sys/resource.h
#define RLIMIT_NOFILE   5               /* file descriptors */

The process is checking the limit of file descriptors and then closes the whole possible range which seems a little unnecessary since almost none of them are open. But this was just after a fork, and a forked process inherits all the open files of it's parent, this might not be what you want so a close is in order. There are however no easy way of getting a list of all used file descriptors so what we see here is a brute-force approach of making sure none are open before continuing. This would probably not have been noticed if it weren't for the unusual high limit of file descriptors.

# plimit $(pgrep processx|head -1) | grep nofiles
nofiles(descriptors) 65536  65536

Perhaps a iteration with close on the contents of /proc/${PID}/fd would have been less resource consuming in this scenario.

All of this was done in a production system without impact to applications which is crucial, you must be able to trust that it will never bring your system down. This is something DTrace can be trusted with where some platforms lacking it but tries to provide somewhat similar observability fails, read Brendans blog: using systemtap or the older but entertaining DTrace knockoffs.

Download the DTracetoolkit here.

ZFSSA/S7000 major update

noreply@blogger.com (Henkis) — Tue, 03 Jan 2012 14:29:00 +0000

The first major software update of S7000/ZFSSA/Fishwork in over a year is now available. With the original version "2011.Q1" it seems a bit delayed, perhaps due to the departure several key persons behind the software post Oracle acquisition of Sun. New features in this release:

ZFS RAIDZ read performance improvements
Significant fairness improvements during ZFS resilver operations
Significant zpool import speed improvements
Replication enhancements - including self-replication
Seval more including bug fixes.

There are alsoa few integration features for Oracle including RMAN and Infiniband if you happen to have a Exadata around.

The improved RAIDZ performances is the hybrid raidz/mirror allocator in zpool version 29.

The ZFSSA is a fantastic product with probably the best interface and analytics available. But the development seems to have stagnated a bit the last year, so have the blog post with useful information and performance comparison by the people behind it. And I still miss one feature badly; synchronous replication of datasets, continuous replication is not always good enough.

Release Notes

Solaris 11, illumos and the source

noreply@blogger.com (Henkis) — Tue, 27 Dec 2011 19:20:00 +0000

Over a year ago Oracle closed the source for OpenSolaris, leaving contributors outside of Oracle left in the cold. That was a huge problem for the adoption of Solaris which had finally begun to rise again. Recently Solaris 11 was released, but without the source, this was likewise a huge problem, but now also for enterprise customers who are using and paying for Solaris. DTrace have lost part of it's value for Solaris 11 compared to the now dead OpenSolaris.

A perhaps even large problem is that when Oracle closed Solaris, many, many of the core developers left Oracle. Several of them now works outside of Oracle contributing to illumos, but these changes can not be ported back into Solaris 11 as long as Oracle keeps the source closed.

Solaris 11 has features not available in illlumos, but I chose to use illumos instead of Solaris 11 since I have access to the source and I am not locked to one OS-distribution. Also the licensing for Solaris 11 does not allow me to use it for small deployments without buying a whole support contract. If I "upgrade" a zpool to use new features available only in Solaris 11 I will be unable to import the pool using the free ZFS implementation that in illumos based distributions such as OpenIndiana or Nexenta or other operating systems such as FreeBSD.

I think this is a terrible move by Oracle, not only are the alienating new customers, they are also locking out great engineers who have implemented revolutionary features into Solaris. As Bryan Cantrill pointed out in his LISA '11 speech, Oracle has not made any official announcement about what they have done to OpenSolaris or what their future plans for the source are, this is very troubling and ignorant.

Solaris 11 is a great OS but it being treated terribly by Oracle, Oracle seems to think that the best way to make a profit out of Solaris is to keep it closed for everyone else, I don't agree.

Solaris history and illumos

noreply@blogger.com (Henkis) — Sun, 11 Dec 2011 01:22:00 +0000

A great presentation by Bryan Cantrill from LISA11 is now available. Besides describing the illumos project it is also a good insight in the history of Solaris and what happened after Oracle acquired Sun. A must read for anyone interested in illumos or Solaris in general.

Fork Yeah! The Rise and Development of illumos

View more presentations from bcantrill

Update: The video is now also available.

LDOM supported by Oracle

noreply@blogger.com (Henkis) — Sun, 04 Dec 2011 01:14:00 +0000

Oracle VM for SPARC is finally supported by Oracle, you can now partition a system with with logical domains and license hardware available to the domain. This is something Oracle customers have been wanting for a while now and I must say I believed it would have been solved the day Oracle acquired Sun. Nevertheless it's now finally supported at the same time the first T4 systems are being delivered to customers, which must be considered good timing.

Server/Hardware Partitioning
Important change regarding OVM/SPARC and Licensing

First impressions of Solaris 11 11/11

noreply@blogger.com (Henkis) — Thu, 10 Nov 2011 22:21:00 +0000

I have had a few hours to try the final Solaris 11 release, overall I think it is far more stable and polished than the previous "Early Adaptors" release. Besides the fact that I am unable to use semi-old SPARC gear to test the release since only the latest generations of hardware are supported I have found few real problem so far.

The new packaging system finally fixes what I believe have been the biggest problem for Solaris the last five years or so, it is now repository based and it is simple to install software and dependencies are automatically solved. No more hassle of downloading and installing multiple software packages from SunFreeware to resolve dependencies. This also makes packaging faster and safer, brining the whole system to a known level and always with a safe recover option since it is used in conjunction with ZFS clones.

Zones are a fantastic tool for security/workload separation and virtualization so it's good to see that so many enhancements have been done in this area. The perhaps most noticeable is of corse that they now also use the new IPS system for packages and that makes a vanilla zones very lightweight without the hassle of a sparse zone. NFS service can now finally be provided from inside a zone. There is a tight integration with the new crossbow network virtualization making is possible to limit bandwidth to zones, use DHCP in a zone without having a separate NIC and build internal networks between zones inside a single Solaris 11 instance.

Imagine the power and flexibility of an T4-4 with 256 CPU threads and 1TB of memory running 50 zones with several high bandwidth/low latency networks inside the machine with no latency or overhead caused by virtualization.

Unfortunately the X86 version with the graphical desktop seems to be somewhat unstable compared to the Express release, I think it's related to the upgrade of the X server. I have been unable to use my laptop with to displays with the final release.

Solaris 11 is however focused on usage in servers and it seems stable for that, I have only found one disturbing problem so far, sharing ZFS filesystems does not seem to work ( zfs set -o sharenfs=on), but you can share each individual filesystem with the share command. Sadly if you are evaluating this you will probably have to wait until next year when there is a full release of Solaris since no updates are provided without a service contract. If you work for Oracle this is something you might want to fix for everyone, or tell me what I'm doing wrong.

Solaris 11 have many other new features such as per-user encryption of home directories with ZFS crypto, a new mirror/raidz hybrid bloc allocator for ZFS, numerous security enhancements among other thinks. I have only named a few of the changes I will probably keep posting Solaris 11 stuff as I find something interesting that is not directly highlighted it Oracle own what's new documents.

Update: As pointed out in the comments sharing of NFS together with ZFS works a bit differently now. If you share an existing dataset you have to set the share property. However if you set the sharenfs property when creating the dataset it works as in previous versions of Solaris 11 Express, OpenSolaris etc. Move information available in the documentation here.

Solaris 11 released

noreply@blogger.com (Henkis) — Wed, 09 Nov 2011 18:57:00 +0000

Solaris 11 is available for download "SunOS Release 5.11 Version 11.0", based on build snv_175b.

There are of course many changes since Solaris 10, most of them have been available in the latest build of OpenSolaris but there are some new that are unique to the final release of Solaris 11.

Install images are available for download and works on all current SPARC machines which is the T and M-series. There are also images available for X86-based machines which also can be used in VirtualBox. Here is a quick reference for the brand new packaging system: IPS one liners.

I will post more detailed follow-up after I've had time to test it for more than a few hours.

Oracle Solaris 11 11/11 – What’s new
Download Oracle Solaris 11
Future features of Solaris 11

Solaris 11 release and webcast

noreply@blogger.com (Henkis) — Tue, 08 Nov 2011 12:14:00 +0000

Solaris 11 will be release 2011/11/09 (2011/11/11 was not optimal for some reason).

Oracle will host a launch event in New York and you can register to attend to the live webcast.

Even if I have abandoned Solaris 11 for OpenIndiana for storage related installations, Solaris 11 have it's obvious place on bigger iron in the datacenter or for any mission critical workload that needs enterprise support. I would gladly have continued to use Solaris 11 for storage but the change made by Oracle to ditch the community and move to closed source and stricter licensing prevents that.

This will make fantastic features such as crossbow, IPS, Native CIFS and COMSTAR available for use in production environments. Many enhancements for zones have also been made, they can for example be NFS servers in Solaris 11.

Also if you want to make the most use of the new SPARC T4, Solaris 11 is the best choice since not every change usable to the T4 have been ported back to Solaris 10 8/11.

If you pay for support of Solaris 11, please demand that Oracle gives you access to the source, DTrace will loose it's value otherwise and I think Oracle needs to hear that.

Oracle Solaris 11 Launch webcast

ZFS is 10 years!

noreply@blogger.com (Henkis) — Wed, 02 Nov 2011 10:48:00 +0000

It's been 10 years (31 Oct) since the first ZFS prototype was run inside of Sun Microsystems, 4 years later it was integrated in OpenSolaris and available to everyone to test. I've been using ZFS since 2004 and I must say that while I have always been impressed by ZFS it took until 2010 before the last bits where in place to make it really usable in demanding environments without problems. It takes time for a complex and critical system to grow mature, that's why I think ZFS will be the best option for a few years to come even if other systems could provide the same functionality. Today ZFS is available in both commercial and free operating systems as well as the foundation of several commercial storage products such as the Sun ZFS storage appliance and NexentaStor.

Some of the latest ZFS features are only available in the closed version of ZFS owned by Oracle and distributed with Solaris but the free version is also going forward. It 's being developed by the illumos community with companies such as Nexenta, Joeyent and Delphix. The illumos branch of ZFS is also stable and used by the companies developing it in their own commercial products.

ZFS 10 year anniversary
ZFS: Live after Oracle
ZFS working group

Fast upgrades, with zones

noreply@blogger.com (Henkis) — Thu, 20 Oct 2011 14:59:00 +0000

I've been working with a sub-optimal solution with zones for a few years. The zones are located on SAN and movable between hosts which is good, but they are also located on an a filesystem which is not ZFS and not even UFS. This have made upgrades and patching terribly hard and slow, especially with more than 20 zones per host. All local zones earlier had to be upgraded at the same time as a global zone and all they had to be down for the entire operation (sometimes over 8 hours).

After a testing upgrade-on-attach for a while and combining it with Turbocharged SVr4 packages I now have a solution which brings the downtime to under one hour for an entire upgrade including local zones.

Everything would have been even easier if we had all root filesystems on ZFS. Since migrating several terabytes of data at the same time as the upgrade was not an option this was a good solution:

Detach all local zones
Add Live upgrade and Turbocharge patches
(119254-70,121428-13,121430-40,124630-28 or later, included in S10u8)
lucreate to a ZFS rpool
luupgrade to S10U10 and add additional patches
luactivate and reboot
$(zoneadm attach -U) all zones in parallel

This solution is dependent on good I/O and perhaps even a separate disk/LUN for every zone root plus sufficient CPU resources on the system.

There you go, Solaris 10 8/11 is ready for the SPARC T4!

Patches for "Turbo-Charging SVR4 Package Install" are now available

Data corruption

noreply@blogger.com (Henkis) — Tue, 18 Oct 2011 08:01:00 +0000

Today I had an interesting experience on how bad things are on platforms that lacks ZFS. For an upgrade I was dumb enough to store a ISO image on a non-ZFS filesystem for a while. Later when performing upgrade tests I begun to get all sorts of strange errors from live upgrade.

The image had been corrupted on the disks or somewhere between the plates and the memory, before I verified it with md5 it was not found by any layer of the storage system.

I would normally never store anything important on a filesystem which is unable to protect data, and this time I was lucky since I found the corruption after a day instead of after a few months.

If feels good to know that even my data at home is protected by ZFS.

SPARC presentations from OOW

noreply@blogger.com (Henkis) — Wed, 12 Oct 2011 15:58:00 +0000

The presentations from OOW are now online, two are very interesting from a SPARC perspective. One of them is a somewhat deeper technical review of the new SPARC T4. The second states that the next generation M-systems are currently being tested together with the 28nm 16-core T5 that testing should start this October and they have marked that processor as being delivered "early".

Since there should be few differences between a T4 and T5 besides the 28nm shrink, double the cores and possibly a frequency bump it seems like it could be possible squeeze the time to market for that design. (With few differences I mean that the core itself is ready)

Next Generation SPARC Processor, An In-Depth Technical Review
SPARC Strategy

A new dawn for SPARC

noreply@blogger.com (Henkis) — Thu, 06 Oct 2011 19:22:00 +0000

I think latest news and release surrounding SPARC changes the economy quite a bit. You will now get Virtualization with LDOM and Zones, Solaris and management with OpCenter included in your support contract.

This comes as the same time as the SPARC T4 is released with is suitable for a wide range of workloads contrary to the previous generations on T-series. The new T4 systems are also much cheeper compared to M-series with the same amount of compute power.

The included encryption also seems to be the fastest currently available in a common processor, so you can get the benefit of encrypting your filesystem/database/network with a very small performance penalty.

Compared to a few month ago the list of good news is quite impressive:

SPARC T4 with 5x the single thread performance of SPARC T3
2-3x times faster encryption compared to T3 (which already was fast)
Live migration of LDOMs
OpCenter included in support contact
Live migration between T2,T3 and T4 in the near future
T4 supports 256GB memory per socket (1TB for the 5U T4-4)
A engineered system is available, the SPARC SuperCluster
Solaris 11 will be released shortly

Compared to other virtualization technologies Solaris Zones has no overhead and VM for SPARC (LDOM) have no overhead for CPU or memory. Some presentations at OOW claimed that zones had 4x less latency than KVM and 15x lower overhead compared to VMWare.

The next generation of T-processor is also being developed and should arrive next year. They will double the number of cores to 16 per CPU. A possible T5-4 would have 64 cores and 512 threads in one 5U enclosure.

I know there is a lot of talk about supporting the whole stack, but you will have one single vendor to blame for problems with the server, virtualization and management. Perhaps more importantly it will not cost more than the hardware itself and the support contract.

More news from OOW 2011

noreply@blogger.com (Henkis) — Thu, 06 Oct 2011 18:47:00 +0000

Another quick update of things I've found out so far, I will write more when I find some spare time:

There will be a Solaris 11 release event the 9th of November
The final Solaris 11 release should be based on build snv_175
The next version of OpCenter (12C) will be able to manage existing zones (brown field)
The next update of Solaris 10 (in 2012) will be the last Solaris 10 update
Solaris developers claimed that part of the reason for not supporting older sun4u machines was due to the way cache was handled which was not optimal.
While Solaris 10 supports the SPARC T4 you will get optimal performance using Solaris 11 since some changes could not be backported.

Solaris 11 Launch Event Webcast

First days of OOW

noreply@blogger.com (Henkis) — Wed, 05 Oct 2011 16:49:00 +0000

A quick summary of he most interesting news from the first days of Oracle Open World that are related to Solaris/SPARC.

Solaris 11 is slightly ahead of schedule, it was supposed to be released by the end of the year.

OpCenter is now included in your support contract for no additional cost, including virtualization!

A new Oracle VM for SPARC will be release (3.0) that supports live migration between processors with diffent clock frequency and even between T2 and T3.

Future releases of Oracle VM for SPARC will try to remove limitations, for example enable live migration while using dedicated PCI hardware in guest domans.

Some work on Solaris 12 have already been started

Solaris 11 will have more focus on virtualization/cloud

SPARC super cluster will support exadata instances (one per T4-4)

Now i'm of to a new session, a deep technical review of the SPARC T4.

OpenWorld and Solaris reunion

noreply@blogger.com (Henkis) — Fri, 30 Sep 2011 19:03:00 +0000

I'm going to San Francisco and Oracle OpenWorld now. In the coming week will dig into the details of Solaris 11 and the new SPARC T4 processor among other things. Mark Hurd will probably deliver an updated roadmap for Solaris and SPARC, perhaps we will even get some kind of confirmation on the release date of Solaris 11.

Joyent will also host a Solaris family reunion event on tuesday:

"There’s been a lot of news in the past few months about the progeny of Solaris: Illumos, SmartOS, OpenIndiana, and the forthcoming Oracle Solaris 11. Since many of our old friends/colleagues will be in town for Oracle Open World, we of the Illumos / SmartOS / OI community thought it would be a nice opportunity to get together, hoist a few beers, talk about old times, and maybe share our visions of the future."

Solaris family reunion
Oracle OpenWorld

The SPARC T4

noreply@blogger.com (Henkis) — Tue, 27 Sep 2011 23:18:00 +0000

Besides the new core and cache layout the SPARC T4 is very similar to the SPARC T3 in I/O and memory design. It also supports 16 DDR3 DIMMs with two controllers, built on 40nm technology, uses 6 * 9.6GB/s coherency links, Dual PCIe Gen2 interface and dual on-chip 10GbE interfaces.

The new S3 core is whats sets the SPARC T4 apart from previous T-processors. It has several features that greatly improves the performance compared to the T2/T3:

Out-oforder execution
Dual instruction issue
Data/instruction prefetch
Deeper pipeline
MMU Page size up to 2GB
Level 3 cache

"All of these characteristics in the SPARC T4 have yielded improvements in single-thread performance by 5X while retaining networking and throughput performance equal to that of previous multicore processors from Sun/Oracle."

The crypto graphic units (SPU, previously MAU) have also been moved into the pipeline so there is no longer any need to managed crypto units be individually. They provide high performance crypto acceleration for the supported algorithms:

"The SPU is designed to achieve wire-speed encryption and decryption on the processor’s 10 GbE ports. "

"These accelerators support 16 industry standard security ciphers and enable high speed encryption at rates 3 to 5 times that of competing processors."

The T4 has the ability to execute critical threads exclusively on a core. This is done by issuing a system call but it can also be handled from the command line by raising the thread priority to above 60. This means that existing applications can take advantage of this feature without rewrite. Applications that depend on a single high performance thread this thread can be declared as critical while other threads can take advantage of the highly threaded design of the T4 allowing great throughput while still providing the needed single-thread performance.

(I think what they should have been more specific in the first sentence by writing Oracle Solaris 10 and 11)
"Oracle 10 now and 11 (initial release) will have the ability to permit either a user or programmer to allow the Oracle Solaris Scheduler to recognize a 'critical thread' by means of raising its priority to 60 or above through the use of either the Command Line Interface or system calls to a function. If this is done, that thread will run by itself on a single core, garnering all resources of that core for itself. The one condition that would prevent this single thread from executing on a single core is when there are more runnable threads than available CPUs. This limit was put into place to prevent resource starvation to other threads. There will be further enhancements to Critical Thread Optimization done for Oracle Solaris 11 initial release)."

For many organization the high single thread performance of the T4 will enable the T-series to be used as a general platform for SPARC virtualization. Previously you could mix zones or dynamic domains on the M-series with LDOM on the T-series but there was not good solution for general workloads due to the weaker performance of the T2/T3 cores. Solaris zones are still useful tool for virtualization that has it advantages but the built-in virtualization in the T-series can provide better separation and live migration between hosts (which in turn can contain zones).

Besides the new processor the SPARC T4 systems comes pretty much the same chassis as the T3, they do however support the double amount of memory using 16GB DIMMs.

Oracle's SPARC T4-1, SPARC T4-2, SPARC T4-4, and SPARC T4-1B Server Architecture

Oracle announces SPARC T4 servers

noreply@blogger.com (Henkis) — Tue, 27 Sep 2011 14:53:00 +0000

Oracle have now announced the new line of servers based on the T4 processor. They come in three configuration ranging from 1 CPU/8cores with up to 256GB memory to 4 processors and up to 1TB of memory. There is also a blade version which has one processor and up to 256GB memory. The T4-1 and T4-2 processors are running at 2.85GHz and processors in the T4-4 are running at 3.0GHz.

Oracle claims a 5 times increase in single thread performance compared to previous T-processors while keeping the same high thread count and cryptographic acceleration. All the new systems support Oracle VM for SPARC, previously known as Logical domains (LDOMs). LDOMs will finally be an option for general workloads, and the supported version 2.1 also support live migration.

Supported Solaris releases are Solaris 10 8/11, Solaris 11 or Solaris 10 9/10,10/09 with 10 8/11 Patch Bundle.

SPARC T-Series
Oracle's SPARC T4-1, SPARC T4-2, SPARC T4-4, and SPARC T4-1B Server Architecture
SPARC T4 announcement

T4 focus in new S10 8/11 KUP

noreply@blogger.com (Henkis) — Wed, 21 Sep 2011 15:11:00 +0000

As expected Oracle is focusing on the yet to be released SPARC T4 processor in maintenance patches for Solaris 10 8/11. The recently released kernel patch has several changes specific for the new platform. All we need now is the hardware itself.

From patch 147440-02:

6994535: watchdog timeouts seen on small subset of T3/T4 systems
7014100: Solaris cpu-mondo queue filling up on T3/T4 systems
7040407: T4 should not advertise cspare hwcap
7045829: libc_psr memmove needs to avoid use of block init store for T4
7045836: memcpy on T4 could run substantially faster
7047568: perf counter changes for T4 1.2
7079983: T4 memcpy triggers compiler CR 7076485

Patch details 147440-02 - SunOS 5.10: Solaris kernel patch

Solaris 10 8/11 released

noreply@blogger.com (Henkis) — Fri, 16 Sep 2011 12:30:00 +0000

Solaris 10 8/11 (update 10) has now been released and is available for download.

A short summary of some of the changes since last update besides support for the SPARC T4:

Support for Two-Terabyte Memory Systems
ZFS Enhancements such as ZFS diff, ZIL synchronicity, RADIZ/mirror hybrid allocator and read-only import.
diskinfo Utility
Shared Memory Enhancements
SPARC: Support for Fast Reboot
libmtmalloc Improvements
LDAP Name Service
Samba Upgrade to Version 3.5.8
Samba, which provides file and print services to SMB/CIFS (Server Message Block/Common Internet File System) clients, has been upgraded to version 3.5.8.

Update: This release could provide better ZFS performance besides the new features, around 7-20% for reads and 10-15% for writes. The tests where done with a few simple dd(1) instances but interesting anyway, read more here.

What's New in the Oracle Solaris 10 8/11 Release
Oracle Solaris Downloads
Solaris 10 update 10 ZFS refresh
Solaris 10 update 10 update
Solaris 10 update 10

Solaris 11 EA

noreply@blogger.com (Henkis) — Thu, 15 Sep 2011 10:39:00 +0000

A much updated pre-release of Solaris 11 is now available on OTN, it is supposed to contain all the final functionality of Solaris 11. Text installer for SPARC/X64 is available. live-CD for X64 as well as repository images.

This release should work all the same (64-bit) X86 systems as Solaris 11 Express, but on for SPARC you must have a T or M-series class machine, support for older UltraSPARC based system has been removed.

Thanks to Craig S. Bell for pointing this out in previous comments, I did not expect any new release before Oracle World.

Update: The build of this release is snv_171, the Solaris 11 Express release was based on snv_151a.

Solaris 11 Early Adopter Program

OpenIndiana 151a released

noreply@blogger.com (Henkis) — Wed, 14 Sep 2011 13:16:00 +0000

The free successor to the OpenSolaris distribution have released OpenIndiana 151a, exactly one year after their first release. The new release is based on illumos and has the new KVM, several ZFS enhancements and support for new hardware.

It's sill considered a development release, I have however found it quite stable for my private storage node which has been running build of 151a for well over a month. A stable released based on 151a is expected later this year.

openindiana.org

Solaris 10 8/11 KUP

noreply@blogger.com (Henkis) — Mon, 12 Sep 2011 14:24:00 +0000

The first sustaining kernel patch for Solaris 10 8/11 (update 10) is now available: 147440-01. But Solaris 8/11 is yet to be released, they could be waiting until Oracle world for the release or they are having trouble with the release. The patch does not contain many changes but another interesting patch that requires the update 10 kernel patch was also just released:

147159-01: T4 crypto performance patch

7030953: AES CBC mode code path spends 50% time in non-encryption operations
7032670: AES-128-ECB 16KB encryption on T4 can be significantly better
7032698: AES-128-CFB128 16KB encryption (via ucrypto API) on T4 can be significantly better
7033814: assertion failure in OpenSSL speed RSA-4096 test via pkcs11 engine on T4
7036405: heavy mutex contention in libumem results in negative scaling of multi-threaded RSA test
7045416: PROFILE TRIGGER VALUE DROP-DOWN DOESN'T POPULATE FOR FIELDS BASED ON TABLES
7048794: 64-bit libsoftcrypto not enabled for T4
7069494: AGILE00205948:CAN'T DELETE STEP IN WORKFLOW EASILY

This is is the second proof that Solaris 10 will indeed support the new SPARC T4 processor.

Update: The final release of Solaris 10 8/11 have the the first sustaining kernel patch already installed (147440-01/147441-01).

IDG SPARC T4 article

noreply@blogger.com (Henkis) — Mon, 29 Aug 2011 18:42:00 +0000

A interesting article about the upcoming SPARC T4 is available from IDG but only with an subscription. The article is also available in text form on a Danish Oracle blog.

Excerpts from the article:
"The new T4 processor, running at 3GHz or more, has features that will also allow T4-based systems to take on some workloads that today are going to Intel Xeon processors, which today perform faster on single-thread workloads than do the T3- series of SPARC processors."

"At the Hot Chips 2011 conference, an IEEE technical conference held at Stanford University from August 17-19, 2011, Oracle systems engineers described the top features of the new T4 processors, including a 16-stage integer instruction pipeline and enhanced cryptographic performance. Among the business benefits associated with the new design will be: double the amount of per-thread throughput performance, compared to T3 – and a range of 2 to 7 times more single-thread performance for business workloads than T3 processors. Given the binary-compatibility of T3 and T4, this means that the same Oracle Solaris applications that have been running on T3 will see considerable speedup on T4, without recompilation."

Nyt om næste generation SPARC T-series
(Only a short introduction in Danish, article in English)

Oracle OpenWorld 2011

noreply@blogger.com (Henkis) — Wed, 24 Aug 2011 16:47:00 +0000

I will try to attend Oracle OpenWorld 2011, it will my first time and I think this year will have a lot of information regarding the SPARC T4 and Solaris 11. It has been almost a year since we saw the last bits of the Solaris 11 development so I expect there have been substantial changes. Maybe the Solaris 10 Update 10 release will be kept unannounced until the conference, it's already been several weeks since the KUP was released on MOS.

Expect regular updates if i manage to attend the event.

Solaris 10 update in 2012
Solaris release dates reminder
SPARC T4 information and beta program
Oracle OpenWorld 2011