Here is the follow-up on the first ZFS compression article where I've been discussing the performance gain that compression brings and the way to measure power consumption with OpenSolaris.

I checked that ZFS compression improves the IO perf when copying a large file and I now want to measure the impact on power consumption.

During the 7m19sec that it takes to copy the large file without compression, my computer consumes 3.65Wh for an average power of 29.9W. During the 5.26sec with compression=on it consumes 2.88Wh for an average power of 31.82W.

To focus even more on the energy specifically consumed by the IOs I can substract the energy  that my laptop consumes when idle, i.e. 22W (see previous article for details). This turns in 2.68Wh for 7m19sec and 1.99Wh for 5.26sec. I end up with the following numbers:

• 0.97Wh for IOs without compression
• 0.89Wh for IOS with compression=on
  

Turning compression on consumes more power but thanks to the extra speed it saves energy with a 1.09x factor - 9%.

On my laptop, compression has a triple positive impact:

• Improves IO performance by a factor of 1.35x
• Saves space by of factor of 1.77x
• Saves energy during IOs by of factor of 1.09x
  

To close this article, let's have a look at the different compression algorithms available with ZFS. For my experiment, I have been using the default by setting compression=on on the ZFS file-system. The man pages tell me that by default ZFS uses the lzjb algorithm "that is optimized for performance while providing decent data compression". Yet, the gzip algorithm is avaiable too, and I can specify the gzip level by using the value "gzip-N where N is an interget from 1 (fastest) to 9 (best compression ratio). Currently, gzip is equilvalent to gzip-6". Let's do a quick test.

First, I delete the data in the file-system where I want to do the test, then I set compression=gzip-1 on it:

# rm /rpool/compressFS/*
# zfs set compresssion=gzip-1 /rpool/compressFS/*

Then, I create a large file - mine is 4.8GBytes - to this file-system and I cp the large file into the same file-system:

# time cp OpenSolaris2009.06.vdi foo.vdi
real    5m4.266s
user    0m0.095s
sys    0m5.903s

# zfs get compressratio rpool/compressFS
NAME              PROPERTY       VALUE  SOURCE
rpool/compressFS  compressratio  2.57x  -

5m42sec for the real-time and 2.57x for the compression ratio. With compression=gzip-9 I get 12m32sec for real-time and 2.82x for the compression ratio.

These values have to be compared with the ones for compression=on (see previous article for the details): 5m26sec for real-time and 1.77x for the compression ratio.

With my laptop, the default compression provides me the best performance, while gzip-1 brings me a much better compression ratio with a minimal impact on perf. Gzip-9 slightly improves the ratio over gzip-1 but with a very negative impact on perf. For sure the differences between these algorithms depend on the disk's speed and on the data being compressed. If I were using and SSD instead of a hard-disk, algorithms with a low compression ratio - hence the onces where CPU is waiting for the disk - would probably see a boost in performance.

So, there you have it. Every CIO there is being tasked these days to potentially reduce costs and accelerate the time to market by incorporating this Mantra in their data center. Either by building in-house or by buying the service from an external vendor that offers these features to the user.

Its no news that Virtualization and Cloud Computing go together. Virtualization enables Cloud Computing, and in fact is one of the key under lying technology enabling the Cloud. However, often people are left wondering if Cloud Computing is no more than "Next-Gen Virtualization". In my mind, Yes and No, especially if you are talking about Private Clouds. I got motivated to write this piece to bring some clarity to this topic. 

A Cloud is virtualized by definition, but the degree of cloudiness depends on the level of automation and self service that is built into it. If the offering is a Public Cloud, then it must incorporate almost all principles of a Cloud Computing architecture that include multi-tenancy and pay-as-you-go model, enabled by a virtualized self-service platform with built-in services like billing, metering, charge back, along with public APIs or a Portal for public access to the Cloud. This is much more than an evolved virtualized environment, where you can quickly get a pre-deployed server or a service, ready to use, but does not necessarily have all these other built in services like self-service and pay-as-you go. In fact, a cloud service should be such that it can be abstracted for use on an as-need basis, and not just be a service that can be used over the internet.

In case of a Private Cloud, the users are more targeted and there is more leeway in how much of Cloud Computing principles are built into the infrastructure.  Infact, most enterprises are going with the incremental approach so as to leverage their existing legacy systems, yet begin to harvest the benefits of Cloud Computing. This is making Hybrid Clouds more popular where enterprises can spin off certain type of functional workloads or burst loads to an external Public Cloud. Several Public Cloud vendors are offering this feature where the customers can embed the compute nodes from the public cloud into the company VPN, hence making the public cloud nodes part of the company data center. While this architecture has some security concerns, given the public compute nodes in most cases are physically not separate from other nodes in the cloud, it being a virtualized environment, its still a good mid-way between an exclusive public cloud and a private cloud. Its worthwhile to mention here that there are companies like GoGrid and Rackspace who would offer dedicated hardware in their data centers to complement an enterprise private data center. However, the more dedicated the hardware, less are the cost and flexibility benefits of a cloud available to you.

Rack is a webserver interface for the Ruby open source programming environment.  Amanda Waite took the trouble of making this package available via the OpenSolaris packaging mechanism; ultimately, ruby-rack will be in the OpenSolaris "contrib" repository, but not until it gets a little testing and review, and some votes from the community, this functionality is now in the OpenSolaris "pending" repository.  I admit to great naivete about Ruby, so I'll refer you to this excellent explanation of what Rack does for Ruby developers.  (my summary of what I got from the explanation: Rack is a web framework for Ruby developers; in other words, if you like writing Ruby code and you want to write code to create a web-based application, Rack is a nice extension to Ruby to make doing this easy).

Anyway, been doing some testing of the package that Amanda submitted.  Here's what I did and what I observed:

First, I made sure I had access to the OpenSolaris Source Juicer "pending" repository.  As I've noted before, two steps:

1. type "pfexec pkg set-publisher -O http://jucr.opensolaris.org/pending jucr-pending"
2. type "pfexec pkg refresh"

After that, the next time you launch the Package Manager application, look on the right side of the application and you can choose "jucr-pending" from the pop-up menu.  Do that, and after a moment you'll see a list of all the packages in the "pending" repository.  You will find the package "ruby-rack" there.  Note that case matters; the package name is all lowercase.

So I did this and downloaded the package.  The package notes that it depends on the Ruby package.  The Package Manager pulled it in just fine.  How do I know?  Before installation, I could not type "ruby" and get anything back.  Afterward, it worked.

Next step:

$ cd /usr/ruby/1.8/lib/ruby/gems/1.8/gems/rack-1.0.1/example
$ rackup lobster.ru


This starts a web server on port 9292 with a little "lobster" program.  I launch the web browser and go to "http://localhost:9292" and I saw a simple web app that shows a crude "picture" of a lobster, and two things I can do: flip the lobster from right to left and back again, and show a crash dump of the Ruby Rack program.  I tried both of those things; they worked just fine.

After this, I killed the "rackup" program then tried it again with the other ".ru" file in that same directory, "protectedlobster.ru".  That gives an added feature of authenticating into the web app (password is "secret"; as far as I can tell, use any username you wish).  That worked, too: if I didn't provide the right password, nothing happened.  If I provided the right password, it worked fine.

So that all worked just fine.  The next set of tests was more in-depth, running a partial test suite.  For this set of steps, I had to install both the "SUNWgcc" and "SUNWgmake" packages, again with the Package Manager.  I did these steps:

1. su - (i.e., pretend you're root; without doing this, step 3 below complained that it couldn't write to /var/ruby/1.8/gem_home/bin)
   
2. Add /var/ruby/1.8/gem_home/bin to your $PATH
3. type CXX=/usr/sfw/bin/g++ gem install memcache-client ruby-openid camping mongrel thin test-spec --no-ri --no-rdoc
4. type "gem install rake"
   
5. type "rake test".

The output showed it compiling some stuff, then it said "Started" followed by a bunch of periods showing status, then "Finish in 3.960495 seconds."  Finally, a results line said "267 tests, 928 assertions, 0 failures, 0 errors".

That sounds like good news to me!

So folks: the Ruby Rack web framework is available via OpenSolaris.  Check it out and see if it works for you, Ruby developers.  Let us know how it goes, will ya?

Powered by ScribeFire.

A big thank you to everyone who attended the Java Professional Day in Langen. Here's my JavaFX material as promised:

• The presentation in pdf format
• Access to Music Explorer FX a great JavaFX application. Click here to start it
• The path animation demo
• The video cube
• The Java Store demo video (Quicktime) 

Have a look for the JavaFX samples. Lot's of good material!

 

configure.ac files (configure.in in the old days) are used as input to autoconf which miraculously (it's all miraculous to me) expands the macros in the configure.ac file and creates a runnable configure shell script. Autoconf macros may be system dependent so if you decide to regenerate a configure script, it's the job of autoconf, in conjunction with aclocal, to locate all of the macro definitions that it's going to need to get the job done. These macro definitions may be supplied by the local installation of autoconf, by other packages via aclocal or actually supplied by the package you are building in a file named acinclude.m4

Due to some changes in the way Lighttpd is built I found that for Lighttpd 1.4.23 in WebStack we had to regenerate the configure script, I did this using the autogen.sh script that ships with the lighttpd source code. autogen.sh calls libtool, aclocal, automake, autoconf and autoheader. On Solaris 10 autoconf generated an error:

configure.ac:67: error: possibly undefined macro: AC_DEFINE
      If this token and others are legitimate, please use m4_pattern_allow.
      See the Autoconf documentation.

But AC_DEFINE is a standard autoconf macro so why can't it be found? We dug around and found mention of it being to do with pkg-config but couldn't work out why that would be. So We stripped down the lighttpd configure.ac bit by bit and found that the problem went away when we removed the lines referencing a macro named PKG_CHECK_MODULES. The file aclocal.m4 in the lighttpd build directory contains any additional macros required by autoconf to process configure.ac and is generated by aclocal. The lighttpd source distribution includes the aclocal.m4 file that was used to create the original configure script, but when you run aclocal, a new aclocal is generated from macros that have been registered with aclocal on the local system. On this system, after running aclocal, aclocal.m4 didn't contain a macro named PKG_CHECK_MODULES :o/

Further research showed that PKG_CHECK_MODULES is usually supplied in a file named pkg.m4 which is installed as part of pkg-config. It usually installs to /usr/share/aclocal which is the default location used by aclocal to locate macros. In this case though we were using a tool set specifically created by the WebStack team for building WebStack on Solaris 10. In this tool set (wstools),  aclocal lives in /export/wstools/bin and it looks for macro definitions in /export/wstools/share/aclocal. wstools doesn't include pkg-config so there's no pkg.m4. That explains why it couldn't find the PKG_CHECK_MODULES macro, but why the error about AC_DEFINE?

In configure.ac you'll find:

  if test "x$FAM_LIBS" = x; then
    PKG_CHECK_MODULES(FAM, gamin >= 0.1.0, [
      AC_DEFINE([HAVE_LIBFAM], [1], [libfam])
      AC_DEFINE([HAVE_FAM_H], [1], [fam.h])
    ])
  fi

So you can kind of see why the AC_DEFINEs passed to the PKG_CHECK_MODULES macro could be affected, but autoconf indicated that the error came from a line of code not related to PKG_CHECK_MODULES, i.e.:

AC_CHECK_MEMBER(struct tm.tm_gmtoff,[AC_DEFINE([HAVE_STRUCT_TM_GMTOFF],[1],[gmtoff
in struct tm])],,[#include <time.h>])

Here AC_DEFINE is used with another standard autoconf macro (AC_CHECK_MEMBER) and if you look at the generated configure script you find:

if test "x$ac_cv_member_struct_tm_tm_gmtoff" = x""yes; then

cat >>confdefs.h <<\_ACEOF
#define HAVE_STRUCT_TM_GMTOFF 1
_ACEOF

fi

Which shows that it didn't actually fail where it said it did. Turns out the time it failed to process AC_DEFINE were where it was used with PKG_CHECK_MODULES. The error from autoconf actually referenced the first use of AC_DEFINE and not where it failed. Best of all, rather than emit a warning about PKG_CHECK_MODULES being missing, it just adds this to the configure script:

  if test "x$FAM_LIBS" = x; then
    PKG_CHECK_MODULES(FAM, gamin >= 0.1.0,
      AC_DEFINE([HAVE_LIBFAM], [1], [libfam])
      AC_DEFINE([HAVE_FAM_H], [1], [fam.h])
    )
  fi

Which is the un-expanded macro calls from configure.ac. These aren't valid shell commands, so the generated configure script will fail. In this case the best way to fix the problem is by adding pkg-config to the WebStack build tool set.  I've also heard that this problem can occur on systems with older versions of pkg-config, I guess that they may not define a PKG_CHECK_MODULES with the signature required by the configure.ac script.

Sun Startup Essentials - A big hand for your business ideas from stewart townsend on Vimeo.

How can Sun Startup Essentials help you ? Watch this and see how Sun can help your small business with it's Startup Essentials programme http://sun.com/startups

India as a region that has had a higher penetration rate of mobile phones than the internet & this is certainly an area that is not only growing rapidly but has a dirth of socially impactful, easy to use consumer applications. The "me too" applications with a 100% reliance on ads has failed & continues to draw a lot of suspicion towards monetizing via advertisements. It was an interesting day kick started by Aravind Rao with a coverage of what & how startups & other mobile application developers in India can build & sell mobile applications. There is more coverage provided by Rajiv of WATBlog on Arvind Rao's talk here.

There was a lot covered about AppStores & their benefits to developers across all the discussions & talks. Appstores are certainly here to stay and are a quick application delivery & distribution channel for all developers.

Opinions - Its evident that the VAS opportunities in the country are huge. As ARPU declines, the challenge for operators is to increase revenues by differentiating their offerings and develop alternative revenue streams by offering more value added services to the existing subscribers & finding better means of consumer outreach., to stay in the game.

VAS is the way to go what with one 3G service launched & more looming. Content providers are key here to ensure that they have intelligent means to find out how to provide highly customized, relevant  & contextual content to the large & ever expanding consumer base. While UI design for mobile is important, the reality today is that more content delivered is SMS based & it needs to be provided in a manner so as to support low-end handsets as well. One of the speakers mentioned the ABCDs of most popular content in India as - Astrology, Bollywood, Cricket & Devotion ! This has been the case for the last several years & any associated content still has a large & highly interested consumer base even today. Further, social enterprises are still lacking in the region which is again a greenfield opportunity. With mobiles being ubiquitous as they are today, network services being enhanced by operators, the mobile is certainly the one quick delivery mechanism for app vendors. This is an area certainly worth exploring & investing in. I was surprised that the speakers didn't bring up too much of this aspect.

A very detailed on Mobile VAS in India is available here.

Do you want to know where I will be on November 17th, Im going to be at the O2 Indigo, not hanging out trying to get signatures of the latest bands but at the #140conf London event, more details below but with Stephen Fry on the bill its going to be a fantastic day and if you want to attend as a Sun Startup Essential member you get 40% discount. Yep 40%....

The 140 Characters Conference: London will be taking place on 17 November at the O2 Indigo.

The schedule for #140conf: London offers 50+ speakers in 27 fast sessions during the course of just one day.

At the #140conf events, we look at twitter as a platform and as a language we speak. In London we will be exploring the effects of the real-time Internet on Business. Overall we will be exploring the systemic effects the worldwide adoption twitter is having on business.

Some of the topics we will cover at #140conf London include: Racism as a Resource; Sky News: Twitter, Reputation and #iranelection; twitter and Brands, twitter and Education, twitter and Media, twitter and Small Business, twitter and Sports, twitter and The Police, twitter and TV, twitter and Music.

Please review the videos from the NYC event to get a feel of the pace of the event and the content shared

It is our intention to provide a platform for as many people as possible to share their thoughts and engage in conversation with the attending delegates. You will find individual talks set to: 10 minutes; “Featured talks” 15 and 20 minutes and the various panel discussions are set for 15 and 20 minutes.

Speakers include special guest Stephen Fry (@stephenfry) together with:

Alex Bellinger (@alexbellinger) - Journalist; Founder, SmallBizPod
Andrew Davis (@andrew_davis)
Andrew Keen (@ajkeen) - Author, Cult of The Amateur: How the Internet is Killing our Culture
Babs Rangaiah (@babs26) - Vice President, Global Comms Planning, Unilever
Becky McCray (@BeckyMcCray) - Founder, Small Biz Survival
Boyd Hilton (@boydhilton) - TV Editor for Heat Magazine
Chris Brogan (@chrisbrogan) - President, New Marketing Labs
Ciaran Norris (@ciaranj) - Director Invention/SocialMedia8
David White (@daveowhite) - Chair
Dean Landsman (@DeanLand)
Digital Maverick (@digitalmaverick) - Learning technologist, Moodle evangelist
Ebony Rhiney-James (@EbonyRJ) - Digital Marketing Manager at Warner Music Group
Federico Grosso (@federicogrosso)
Gerd Leonhard (@gleonhard) - Media Futurist, Author, Blogger
Heather Gorringe (@wiggled) - Founder, Wiggly Wigglers
James Poulter (@jamespoulter) - Digital Strategist for Ogilvy PR
Jeff Pulver (@jeffpulver), twitter investor, creator of #140conf
Jeffrey Hayzlett (@JeffreyHayzlett) - CMO, Kodak
Jonathan Almond (@jonalmond)
Josie Fraser (@josiefraser) - Social & Educational Technologist
Kathleen Hessert (@kathleenhessert) - CEO, Sports Media Challenge
Kyra Gaunt, Ph.D. (@kyraocity) - Associate Professor of Music and Anthropology, Baruch College-CUNY, writer, blogger
Lauri Stevens (@lawscomm)
Manny Norte (@mannynorte) - Kiss FM
Marsha Collier (@MarshaCollier) - Author, Ebay for Dummies; Radio Host
Maz Nadjm (@Mazi) - Responsible for social networking at Sky Television, sky.com
Nick Halstead (@nickhalstead) - CEO, TweetMeme
Nick Keane (@nickkeane)
Pamela O'Hara (@pmohara) - Co-founder, Batch Blue
Peter Gregson (@petergregson) - Cellist and Co-founder of the Production Company Coffeeloop
Ray Paul (@raypaulbbc) - Executive Producer for Weekend Programs at BBC Radio 1
Roberto Bonanzinga (@Bonanzinga) - Balderton Capital
Saul Klein (@cape) - Index Ventures
Sean Seton-Rogers (@setonrog) - PROfounders Capital
Sherry J. Bitting (@SJBRules) - Co-founder and Marketing Director, Parlour Magazine
Shirley Williams (@shirleyearley)
Stephen Fry (@stephenfry)
Steve Saul (@stephensaul) - Social media/blogger for Simon Cowell's X-Factor
Sue Black (@Dr_Black)
Tal Givoly (@givoly) - Chief Scientist, Amdocs
Tim Bradshaw (@tim) - Digital media correspondent, Financial Times
Vikki Chowney (@vikkichowney) - Editor of Reputation Online, Centau Media

The take aways from this event will provide the attending delegates knowledge, perspectives and insights to the next wave of effects twitter and the real-tme internet will have on business.

As Sun Startup Essentials is a friend of Jeffs we have a lovely discount for you, save 40% on a Delegate Ticket up until 15 November or until tickets last. 

Click here to register:

Get the tickets quickly as this will be a sell out shortly.

Did you ever have a performance problem and not know where to start looking? Well, who didn't? In such a situation, it is important to perform a proper monitoring of the system and resist the immediate urge to blame the processing power --the storage subsystem is often overlooked. Now that I spoiled the suspense giving away the answer, give me 5 minutes to illustrate this with a typical example. Why typical? Because this is a case I meet quite often with startup companies who rightly concentrate on their core business and not on their IT infrastructure --that's what we are here for.

One of our partners, Squid Solutions, recently reported a performance problem with an Oracle database; they had engaged our team as part of the Sun Startup Essentials support program. Squid Solutions makes a software called Nautilus that performs intensive analytics on database systems, whatever their kind or size. They call Nautilus an SQL Knowledge Engine because it models data and business knowledge, and then automatically generates SQL code to execute the data processing tasks. Nautilus is sold as a service performed by Squid's engineers as Customer Intelligence projects.

So, Squid Solutions had purchased a brand-new Sun Fire X2250 server --quad-core Intel Xeon with 4 GB of RAM, running Solaris 10--, and was experiencing poor performance --much lower than their old system of previous generation-- when executing a read-n-write intensive workload on the database. When I first got on the server, something jumped right to my eyes; the server used a single hard drive for all of the following: the operating system; the SWAP files; Oracle binaries; Oracle redo logs; Oracle table spaces and indexes.

This basically means that you will not be able to get more transactions from your database, than the number of IOPS the hard drive can handle, since a transaction is not fully committed until it is written in the redo log file. In the best case, we are talking about 200 IOPS for a single drive! Consider also that this number must be shared between the writes to the redo logs, and the table and index reads and writes. It therefore seems pretty safe to assume that the bottleneck is the I/O subsystem. But we still need to verify this by monitoring the system.

Monitoring the System

We ran the Squid Solutions test case on the Oracle database for 10 minutes and used the dim_STAT tool to collect and analyze all the necessary system statistics: CPU, RAM and I/O. dim_STAT is a great performance monitoring and analysis tool developed by our colleague Dimitri Kravtchuk at the Sun Solution Center in Paris: see Solaris Performance Primer: dimSTAT, dimSTAT by examples and Customizing db_STRESS for some detailed blogging on dim_STAT from our team.

The CPUs are hardly used; they are constantly over 80% idle throughout the test. Also the run queue is always equal to zero, which means that we never have a thread waiting for a CPU. This clearly shows that there is no CPU issue.

The page scan rate is constantly equal to zero: in case of memory shortage this rate can get very high --over 200 for instance. Also we see that throughout the test there is more than 700 MB of free physical memory. There is therefore no memory issue.

Finally the I/O analysis shows the problem; the disk is clearly the bottleneck here. Throughout the test, the single disk is more than 60% busy, this figure even goes up to 85% at the end of the test. Also the wait percentage that should be close to zero, is around 10% and even peaks at 20% once in a while. This test confirms our suspicion concerning the saturation of the I/O subsystem; it needs to be upgraded in order to get acceptable performance from Oracle.

Recommendation

Better performance can be achieved by isolating Oracle's redo logs on a separate drive. By doing so, we will no longer have the share the IOPS with other reads or writes. But it will not be enough; we will still be limited by the maximum number of IOPS possible per drive. The industry solution to this problem is RAID storage; by stripping the Oracle redo logs over multiple disks, you then have 200x the number of drives of possible IOPS.

A budget configuration could consist of a Sun Storage J4200, a simple JBOD array, combined to a SAS RAID HBA to your server to compensate the lack of a RAID hardware controller in the JBOD --and thus of memory cache. This adapter offers up to 256 MB DDR2 cache memory. The cache is important for performance because Oracle does not have to wait until the redo log in actually written to the hard drive. And you have a 1000x better performance writing to RAM than disk! The RAID controller will guarantee that what it has in cache will eventually be written on disk.

A better option for higher performance, reliability and flexibility is an external storage array for an integrated RAID controller such as the Sun StorageTek 2510. It is an entry-level SAS storage, that requires a SAS controller on the server that connects to it. The array can host up to 48 SAS or SATA drives, and can be controlled by one or two SAS RAID controllers. Each controller has a 512 MB cache size, 2-to-4x more cache than the previous solution.

Next Step

Would you like to monitor your own system to check where the bottleneck is? As mentioned earlier, I used a tool called dim_STAT to collect and analyze all the necessary system statistics. You can download and use it for free on Solaris SPARC, Solaris x86 and Linux. Features of dim_STAT include: history of system statistics saved in a MySQL database; simplicity to get many system statistics once, and analyze later; neat graphical output allowing a much easier analysis; capable of collecting from multiple servers, and allowing comparison; etc.

That said, you can already get useful statistics from your Solaris system out of the box, without installing dim_STAT --dim_STAT fully relies in fact on the Solaris native tools. The three most used are vmstat for virtual memory and CPU utilization statistics, iostat for IO and CPU utilization statistics and netstat for network statistics. If you check closely the graphics in the “Test & Results” section above, you will see that for the CPU and memory statistics, vmstat was in fact used, and iostat for the disk statistics.

Using these tools --and most importantly interpreting the output-- is a skill on it own though. Better than reading through the man pages, I recommend you read the Solaris Performance Monitoring & Tuning - iostat , vmstat & netstat article if you want to quickly get started --for the French speakers, La trilogie classique : iostat, vmstat, netstat.

Oh, by the way! I almost forgot ! So why was our partner's old server doing better than the new one? Because it had two disks, and the Oracle redo logs and data files were on separate drives. 

If there is any conclusion to draw from this:

1. Don't throw all your money at CPU and RAM: the I/O subsystem may be where you can dramatically impact performance (for better or worse).
2. Analyze your system with the standard Solaris system statistics tools to get an instant high-level picture of your application's behavior --dtrace and HAR are then tools that will allow you to drill down.

Anyway, in doubt, feel free to call on our developer email hotline at isvsupport@sun.com; it's free for our ISV partners.

Today, at the PayPal conference, Sun and PayPal announced that PayPal will now be a payment option for the Java Store. To read about the announcement and all the details, check out this article.   Additionally, the Java Store and Warehouse are now live.

 To get the client (i.e. store): http://store.java.com/developerrelease

Developer Portal (i.e. warehouse to upload client): http://java.sun.com/warehouse/ 

If you are using Java SE 5.0, you need to know that the EOSL for Java SE 5.0 occurred on Oct 30th.  Version 5.0u22 is the last publicly available update of 5.0.

 http://java.sun.com/javase/downloads/index_jdk5.jsp

US AMI Details

AMI ID :	ami-f2bf5c9b
AMI Manifest :	sun-opensolaris-2009-06/drupal_amp_stack_hardened_opensolaris_2009.06_32_1.0.img.manifest.xml
AKI / ARI ID:	aki-1783627e / ari-9d6889f4
License :	Public

SmallBizPod is a podcast, blog, news and community site for small businesses, start-ups and entrepreneurs. This is how it all happened.

Back in March 2005, Alex Bellinger became one of only a handful of people in the UK to start podcasting.

Looking around for a subject he realised there were no British-based podcasts on business. Since he had just set up his own business and had spent many years focusing on the small business sector, a podcast on entrepreneurship seemed like a good way to go.

SmallBizPod was born. The podcast, blog and website now attract tens of thousands of listeners and readers a month from all over the world.

Inspire, Inform, Entertain

From the start SmallBizPod has been there to inspire, to inform, and to be fun to listen to and read as well. I think thanks to some wonderful contributors and interviewees, it's achieved just that.

Over time, the site has grown to become one of the most visited sites for SMEs in the UK. It's also developed into a wonderful free audio resource to provide inspiration and advice from business people with real experience and real insight.

Take the time to listen to some inspiring interviews from some renowned people on subjects such as marketing, business advice, funding and many more, the beauty is that these are free and really insighful and great to hear that not just you in that Startup boat alone.

Visit http://www.smallbizpod.co.uk/  now and keep an update on the latest podcasts.

Sun Startup Essentials helps small business reduce carbon emissions by 85%

La deuxième édition du BarCampSophiaAntipolis se tiendra le jeudi 5 novembre à l’Agora Einstein - Sophia Antipolis, de 16h30 à minuit. Les participants ont opté pour le thème du Ecommerce. L’OpenCoffee Club Sophia sera partenaire de cet événement majeur des nouvelles technologies de la Côte d’Azur, a tel point qu’il aura lieu au sein même du Barcamp spécialement à cette occasion.

Le thème du Ecommerce, présenté aujourd’hui comme le nouvel eldorado des commerçants, sera débattu sous plusieurs angles de réflexions, aussi bien techniques que commerciaux. Il y a d’ailleurs déjà plusieurs sujets très pragmatiques proposés par les futurs participants :

• Le Ecommerce est-il toujours un eldorado ou une affaire de professionnels ?
• Quelle solution pour se lancer dans la vente en ligne ?
• Les différentes méthodes de communication et leurs retombées en terme d’audience.
• La place du mobile dans le Ecommerce de demain.
• Les outils communautaires : Facebook, Twitter, …
• Les moyens de paiement et la difficulté de convaincre les banques.

L’entrée est gratuite, mais l’inscription est obligatoire afin de faciliter l’organisation de cet événement. Pour ce faire, il suffit simplement de se rendre sur le wiki et d’éditer la page pour rajouter son nom. Il est également possible de contacter Sandra Bayer par email (à sbayer@cari.fr) si vous rencontrez des difficultés pour vous enregistrer.

Vous pouvez commencer à discuter avec les autres participants ou suivre le déroulement du BarCampSophiaAntipolis grâce aux sites suivants :

• Groupe Facebook (plus de 120 membres)
• Groupe Google (pour les propositions de thèmes)
• Twitter (pour être tenu au courant de tout)

US AMI Details:

AMI ID :	ami-e0b05389
AMI Manifest :	sun-opensolaris-2009-06/amp_stack_hardened_opensolaris_2009.06_32_1.1.img.manifest.xml
AKI / ARI ID:	aki-1783627e / ari-9d6889f4
License :	Public

bash # export EC2_URL="https://eu-west-1.ec2.amazonaws.com"
bash # export LOCATION="EU"

Congratulations Micello team for winning the Runner-up award at the AT&T Platinum Fast-Pitch contest!!! Wow!

The event took place at CTIA, held on October 7-8, 2009 in San Diego, CA where AT&T reviewed over 60 entries. The winners received a share in $40,000 of prize money and opportunities for inclusion in an AT&T distribution channel.

Micello has built a database of maps of places people go to shopping malls, college campuses, convention centers etc, and allows users to access, navigate and explore those places using their mobile phones. Unlike, its competitors, Micello keeps its own database of place of interest rather than relying on satellite-based map systems. In addition, by integrating social networking services into the application, Micello brings together all the activity happening inside a building where people gather. 

Find out more at micello.com. 

I don't have anything unique of my own to offer today, but I wanted to pass on a really nice piece of writing that I read this morning.

Jeff Bonwick writes in his blog today about a new feature in ZFS, called deduplication.  There's no way I can explain it better than he, so I'll just leave you to it.  It really lays out the problem of deduplication and ZFS's solution to it nicely.

Powered by ScribeFire.

Sun Startup Essentials est l'un des partenaires mondiaux de Startup Weekend et s'associe à l'organisation de la première session à Paris, du 4 au 6 décembre 2009 à La Cantine.

Il s'agit d'une nouvelle formule venue tout droit des US : vous avez un week-end, c'est-à-dire 54h, pour monter votre startup, de l'idée initiale à la création de la société ou au développement du produit ! C'est possible ? On verra bien... En tous cas, ce sera l'occasion de rencontrer des gens intéressants, de former des équipes dynamiques, de partager des expériences, de s'amuser et de créer des liens pour l'avenir. On ne sait jamais ce qui peut se passer dans ces moments-là. La prochaine star française du web sortira peut-être de Startup Weekend Paris. Vous pouvez vous inscrire dès maintenant sur le site.

A rather lovely Wordle of the SSE blog - rather funky don't you think.

Free and open source content management systems (CMS) like Drupal and Joomla have given entrepreneurs everywhere an opportunity to bootstrap their concepts quickly and effectively. What's the difference between them, and which one is right for you? Let's take a look.

Joomla - This CMS is very popular and freely available as an open source software package. It can be used to design Web sites, community portals, online magazines, corporate intranets, and even e-commerce sites. Out of the box, users can build data reporting tools, inventory control systems, product catalogs, business directories, and more. You don't need to be a technology whiz, however, to use Joomla. To get a feel for how easy it is, try this interactive online demo.

Mambo - Another open source option on the playing field is Mambo, a CMS with Web- and server-based installation options. Its template system means there's no complicated coding required but, because it's open source software, you can customize, tweak, and get as technical as you want. Mambo handles media content like images and video exceptionally well, and has a robust content publication scheduling feature to help keep your site fresh and updated around the clock.

Drupal - This open source content management system made headlines recently when it was announced that the White House uses Drupal to power WhiteHouse.gov. The basic package is loaded with tools to create all kinds of terrific Web sites and portals, but when you include the free add-on modules things really get interesting. Use Drupal to create e-commerce sites, podcasts, newsletters, picture galleries, forums, message boards, and more.

concrete5 - For many startups, an online presence is mainly about branding and selling your company's message. When that's the case, many entrepreneurs wish they could find a CMS that's meant for marketing. Concrete5 may be just the ticket. Designed for "regular people," not developers, concrete5 makes setting up a Web site as easy as sending an email. Down-to-earth editing tools make it possible for someone with no developer experience to build and edit a quality Web site in minutes. That's a terrific bonus, since most startups don't have extra cash lying around to spring for a professional Web designer.

Most open source content management systems have the same basic capabilities and features, and mainly vary in level of difficulty to set up and use. The easiest way to determine which CMS is best for you is to try their online demos and play around a little bit to get a feel for which one best matches your computer skill level. No matter which one you choose, each project has a healthy community of developers and users at the ready to help you get set up and troubleshoot along the way.

Flickr image courtesy of Marco Belluci.

Originally posted at http://blogs.sun.com/sun4startups/entry/cms_startup_in_a_box1

...The CRASH you haven't seen yet...

This time we go behind the scenes of the worst financial calamity that flooded our TV screens last Christmas...

What lessons have we learned? 

What do you do when your whole world has SUDDENLY collapsed, the rug has been pulled out from under you, the regulators are circling and the media spotlight is focused on you?

We are doing business in unpredictable times - impacted by the collapse of the financial systems.  And experts are predicting another down surge...

How do you lead your team, your employees, shareholders and family who are all looking to you for inspiration?

As we get closer to the 1st anniversary of one of the darkest winters the finance markets had ever seen, we bring to you our next breakfast...

Leading through a Catastrophe!

 

Mark Sismey-Durrant is the Chief Executive of Heritable Bank. He knows what he is talking about - having been at the heart of some of the most dramatic events before Christmas last year...(Iceland rings a bell?)

He has been there, done it - and is coming out on the other side.   He has never spoken at an interactive public event - this is the first time Mark will share his insider secrets of leading in these uncertain times.

So, here is your invitation to a breakfast full of shock, surprises, inspiration and intrigue.

Insider Secrets - Behind the Scenes with Mark Sismey-Durrant

Date: Monday, November 23, 2009 at 8:00 AM (GMT)

Location: London Customer Briefing Centre Sun Microsystems Ltd. Regis House 45 King William Street London, London EC4R 9AN
United Kingdom

Click here to book: http://marksismeydurrant.eventbrite.com

Sun Startup Essential Members check your November newsletter for a Discount code and for 3 FREE PLACES to attend.

On october 29 the dutch Unix users group (NLUUG) organized the fall conference in congress center De Reehorst in Ede, Netherlands. This conference topic was “The Open Web”. It was to celebrate 20 years of the World Wide Web. It not only focussed on threads and challenges of today and tomorrow, but also on opportunities like integration and securing your web server with the Yubikey in combination with digital certificates.

Sun Microsystems was present with the NLOSUG; the dutch OpenSolaris User Group. The stand was manned by Kristian, Matt, Jan and myself. Kristian did an excellent job showing the advantages of Dtrace during web development. Matt showed the Snowman and Darkstar demos using the project Darkstar OpenSource gaming backend.

Many questions we got at the desk were about OpenSolaris and it's features. Much questions about ZFS, backups, DTrace and virtualization; but also much interest in how to setup a Sun Ray deployment properly, VDI and how to develop web applications using NetBeans.

On May 6 2010 the NLUUG will organize the spring conference. The title is “System Administration” and aims to extend and deepen the knowledge of system administrators. Interested in attending this conference or interested in participating? Please see the NLUUG website for more information.

A balance between processor performance, memory bandwidth, and IO performance all contribute to system performance. Applications like business intelligence and data warehousing (BIDW), and Rich Media are particularly sensitive to overall performance. 

The Sun Fire X4275 is designed for the ground up for powering both rich media streaming applications, and business intelligence and data warehousing applications.

In this webinar, you'll find out:

• The design principles of the Sun Fire X4275 (Intel Xeon 5500 series processors, 64 GB of memory, 12 TB of storage, 8xSSD or 12x15K drives for IOPS performance all in the same 2U chassis, and up to 24 GB of network bandwidth)
• How flash technology can multiply IO performance offering significant savings in cost/stream, cost/IOPS and cost/GB as well as savings in energy and space
• How the Sun Fire X4275 is the building block of the powerful Oracle Exadata V2 server recently announced by Oracle.

Date: November 4, 2009
Time: 8:00 am PDT / 11:00 am EDT / 4:00 pm UK / 5:00 pm Central Europe / 8:30 pm India (New Delhi) (check timezone)
Speakers: Tushar Katarki, Product Manager, Angelo Rajadurai, Sr. Technical Evangelist, Sun Startup Essentials

Register Now

Archived Webinars

Its being a while since my last blog. And since many projects still in progress I will share my impression about recent build of OpenSolaris b126.

I am long time user of OpenSolaris. This is my tool for everyday work and I have it on my notebook. But still I am just a user, a little of admin, totally not a developer of this system. OpenSolaris 2009.06 was a major milestone with most every features polished to that date. But some was not as good as I wanted them to be. So I came a long road of updating to dev builds. I had all of them starting from 116 till 126. In every build there were improvements as well as new bugs. Only starting with 126 I can say that I feel good about it.

So here is the list why go dev and what you should be ready to. I am not writing in which release bug was fixed. Consider this as a "diff 111a 126" and I am not pretend to show all diff, this is only what is sensitive to me:

• Firefox was beta in 2009.06 and bunch of needed addons does not work in this beta. That was first reason why I started to go dev. Probably I can go updating applications only, but I decided to have whole system recent. Firefox is 3.5.3 now.
  
• System monitor applet had disk activity broken, but now it is fine.
• Overwall system performance (Compiz at maximum visual, Firefox) performs much better. This is purely subjective.
• Sound subsystem has changed. Keyboard beep annoying my coworkers. Did not get how to mute it or lower volume permanently.
• Pidgin's yahoo icon problem is fixed. It was showing offline even when you connected. Pidgin is 2.6.2
• In some builds printer monitor applet gone wild eating cpu. But now its fine.
• I noticed that my /var was growing. This is due to each 'pkg update' is storing downloaded packages. Can be fixed with

$ pfexec pkg set-property flush-content-cache-on-success true

• No more Ctrl-D in Terminal, have to use "exit". Didn't look at it yet. Typing "exit" may be annoying, but it saves from accidental closing of terminals. Right now I have just 5 opened sessions in terminal, usually more.
• I have strange problem with Marvell Yukon Ethernet. Before 125 I had a driver from the vendor. But it does not work for build 125 and 126. So I decided to remove YUKONXsolx and give a try to yge. Something strange happening here. I have external USB disk drive attached. I use this drive periodically as rpool mirror to have a spare. As soon as I have this drive attached before the system booted my ethernet refuses to work. So I need to have USB disk drive detached when system boots. Details are here.
  
• Now waiting for zfs dedup
• Other frequently used applications: Rhythmbox with Audio.fm, Thunderbird, OpenOffice, QCAD, Gimp, Stellarium (Blastwave please update your repository with recent versions! and why it so rare online?), Adobe Reader.
  

Bonjour Séverin, peux-tu te présenter ?

Bonjour Laurent ! Entrepreneur du web depuis 2006, je m'appelle Séverin Benizri et je suis fondateur de la société Paris Labs qui édite des plateformes de réseaux sociaux, notamment le service Helia.fr. J'ai un parcours un peu hybride entre le commercial et la technique, comme beaucoup de monde dans ce secteur.

Que fait Helia ?

Helia est un réseau social professionnel qui encourage le partage d'expérience entre des internautes qui ont des affinités pour le même métier ou les mêmes entreprises. Helia propose également des outils de communication aux entreprises qui souhaitent maîtriser leur réputation sur le Web. D'ici à la fin de l'année, il y aura des informations à jour sur plus de 500.000 entreprises sur Helia, des actualités et des offres d'emploi.

Pourquoi as-tu adhéré au programme Sun Startup Essentials ?

Nous étions très impliqués dans le développement Java. Il nous a paru naturel de nous associer à la maison mère... Le programme présentait de belles opportunités aussi bien dans le domaine du conseil que dans l'acquisition de matériel informatique.

Quelles technologies Sun utilises-tu ?

Aujourd'hui, nous travaillons avec des ingénieurs Sun sur une migration vers OpenSolaris afin d'utiliser ZFS, la virtualisation avec les zones Solaris et GlassFish.

Pourquoi as-tu choisi d'installer OpenSolaris sur une Dedibox ?

Aujourd'hui, Dedibox propose un bon rapport qualité/prix qui correspond à nos besoins actuels d'hébergement. Nous voulions explorer les possibilités d'OpenSolaris sans pour autant changer de fournisseur de services dans un premier temps. En revanche, nous nous intéressons de près aux serveurs Sun, notamment aux processeurs SPARC.

Comment s'est passée l'installation ?

Après avoir envoyé le DVD d'OpenSolaris aux équipes Dedibox, nous avons obtenu un accès VNC (partage d'écran) sur notre serveur pour procéder à l'installation. L'installation s'est déroulée sans problème, tout le matériel ayant été reconnu avec des pilotes adaptés dans la distribution Sun !
Ensuite, avec l'aide d'un ingénieur de Sun, Amir Javanshir, nous avons pu configurer le serveur pour qu'il héberge notre service. Après quelques adaptations et configurations des principaux outils, notre service était dupliqué sur l'environnement OpenSolaris. 

Quelles sont les prochaines étapes pour toi ?

Nous sommes en train de d'effectuer des tests fonctionnels et de performance sur les différentes architectures. Parallèlement, nous faisons évoluer notre technologie vers une architecture plus modulaire (OSGi). Notre objectif est d'héberger la nouvelle version de la technologie sur un serveur Sun avec l'environnement OpenSolaris à T1-2010.

Merci Séverin et bon courage pour la suite !

Pour en savoir plus, vous pouvez envoyer un email à severin.benizri@parislabs.net

The other day I was one of the lucky few with the opportunity to attend the Media140 Branding event over at RIBA in London.  The event was aimed at media professionals and the majority of the audience were PR agencies and branding professionals.


Having spoken to a number of the attendees I would have to say there were as many experts in the audience with interesting things to say on Branding, building a brand and representing the brand on Twitter and related tools as there were on stage.   The content that was provided on the stage was only a small part of what was actually shared and the #media140 twitter stream provided a raft of additional conversation and content. 

So what came up at the event... And what did not... Here's a few of my notes:

Community

Rather than engaging directly with communities the general consensus for agencies represented by the speakers and panels was that you should seek out those that influence rather than those that look after, or run communities.  This was an interesting perspective from my point of view as a community owner through Girl Geek Dinners and as a brand advocate with TweetMeme.  I have to say that in some ways the community leaders are also to some degree influencers but not in the traditional agency approach.  It's interesting to hear how the agencies see influencers in terms of followers and retweet reach rather than people who can direct a mass of people towards a brand.  I'm sure I'll be writing more on this in the future!

Acceptance

The acceptance of a brand seems to be directly related to respecting the privacy of individuals and engaging at a relevant point, not at random or at a point suited to the brand owner.  Brands don't seem to be accepted unless they are honest and open... Transparency is key.  The issue of transparency came up many times across all the panel discussions.

Trust

Transparency of the conversation became really important+, who are you talking with if you are talking to a brand... Brands being personal came up a lot and the conversations flowed around being a brand on Twitter versus being an individual on twitter that happened to work for a brand.  Consumers were less rude with individuals who worked for a company than they did with a company that was represented on twitter.

"richard baker of Virgin Trains gets fewer abusive comments than @virgintrains - are people rude to brands over people?"

Engagement

How do you converse or shout at people who you want to engage and purchase your brand products?  After all that really is what you are after right?  Is it all about conversation or is it about shouting out to get your brand heard and noticed.  The Habitat example got a fair bit of time at today's event.  Along with Ryanair.  Both were seen as examples of brand failure today due to negative press but as the stats show negative press can result in the case of Ryanair as positive sales as people hear the brand name and recognize it.  So does that still mean that all press is good press.  It certainly suggested that TV and newspapers still have the greater traction and action as a result of them.  But could over time Twitter, Facebook etc actually be better and provide a faster route to a sale... The answer still hangs in the air on that one at the moment with a lack of stats on conversion rates.

Identity

There was a lot of conversation around the brand identity getting lost on Twitter with too many Twitter accounts and people not knowing which account or even if they should be tweeting to an account to get a response or go through formal methods such as a call centre to get a response.  Today's wifi and subsequent BT wireless failure was a classic example of this.  The issue was tweeted and a promise of it being rectified was made via twitter but the Wifi never did come back up properly.  The issue was said to be with BT for the whole of the local area.  So would they have been better off calling in the issue... I guess we will never know and this is the same sort of experience most customers have with brands on Twitter.  Then you have the situation where brands are represented by individuals such as @kerryfromDell compared to @innocentdrinks where lots of people are under one brand name.  Personally I went looking for @innocentsmoothies and guess what... I probably wasn't the only one... Identifying the brand and finding them can be as hard as finding out their rules of engagement.  These should be clear and concise... Do you just follow people back or only follow those that you engage with.

Values

One thing that came up today was that older brands struggle to find out what their true brand values actually are today.  These were lost when the original founders left the company.  They are also split and reduced as companies merge or are acquired.  So then companies have to look deep within themselves to find their true brand values and show these through their brand and it's representations online.  And for those companies where this is proving to be hard we are seeing individuals with personality within an organization being brought forward to represent the brand with their own personal flair.

Mission

What is the mission of the brand... Is it to make reading cool.... Or to provide healthy drinks to all!  What is your mission and also what do you want to get out of your social media strategy!

Tools

There were a whole range of social media tools that were mentioned in the talks today, granted most were from just one talk but across the talks here's the tools that were recommended by the speakers at Media 140!

TweetMeme (analytics, custom channels, interesting content, retweet buttons & search)

TweetTabs (Multi tabbed real time search)

Cymfony

ViralTracker

Delicious

TweetDeck

TweetFunnel

TwitterFall

Trendistic

Audience is a conference aimed at those who recognise the need to reach engage and influence audiences of all kinds, an investigation into how this is changing, and a look at how technology has in the past and is now, through new media tools and the social web, changing audience participation and interaction.

The  speakers range from large traditional media organisations with Dan Farber of CBS, though the recording industry with Warner Bros. Music CTO Ethan Kaplan, advertising with Crayon’s Joe Jaffe, HR and internal audiences with Frank Roche of iFractal and new media players like Mike Arrington of TechCrunch and Jason Calacanis of Mahalo, not to mention Musicians like Adam & Mia and the inimitable Andrew Keen of “Cult of the Amateur” fame with more being added all the time.

Engaging on November 6th in the Hudson Theatre, one of New York’s oldest playhouses. We felt it was appropriate to hold a conference that was about Audience in a place that was specifically designed for one.

If you’re interested in meeting with a diverse and eclectic crowd and have a goal of understanding how the methods used to reach engage and influence audiences have changed and are changing as we speak and the possible future direction of these things then please do head along. They’d be delighted to have you.

More Details below and also a SPECIAL OFFER for SSE members - $50 discount - use the code  sunstartup and go here to get your tickets now

GlassFish Enterprise Server v2.1.1 has been released.   This version provides infrastructure support for the SailFin Application Server (Telco), contains about 170 bug fixes, and includes updates to several component packages including JSF, Grizzly, Jersey, and Open MQ.  You can refer to the GlassFish v2.1.1 page for details on the improvements that went into this release.  The bits for this release can be accessed as follows:

Community: https://glassfish.dev.java.net/downloads/v2.1.1-final.html

Product:  http://www.sun.com/software/products/appsrvr/get_it.jsp