Slashdot Mirror

The Java System Messaging Server is a key component of the Java System Communications software portfolio, which also includes the Java System Calendar Server and Java System Instant Messaging plus the Java System Synchronization tool and the Java System Connector for Microsoft Outlook. Integration across these products enables features such as calendar event and task deadline reminders as well as offline forwarding of alerts.

Just use that in conjunction with the SJS Directory Server's identity synchronization with Windows AD environments and you're money.

The Java System Messaging Server is a key component of the Java System Communications software portfolio, which also includes the Java System Calendar Server and Java System Instant Messaging plus the Java System Synchronization tool and the Java System Connector for Microsoft Outlook. Integration across these products enables features such as calendar event and task deadline reminders as well as offline forwarding of alerts.

Just use that in conjunction with the SJS Directory Server's identity synchronization with Windows AD environments and you're money.

Oh, I know someone who does. Well, at times, anyway. They seem to suffer from some serious versioning schizophrenia.

The new push will be to have 8 very simple cores (albeit with advanced SSE4 units with even wider vector instructions such as 256 or 512 bits) and allow each core to run 2 or 4 threads. This won't be hyperthreading as hyperthreading is a form of SMT (although Intel may reuse the name). It will be a form of fine-grained multithreading that allows context switches on L1 or L2 cache misses, as well as other latent operations. Of course their will also be logic to allow all the threads to run equally.

Apart from SSE4, you've just described Suns' Niagara processor. See this blog entry of Jonathan Schwartz's for some nice pictures.

--paulj

The new push will be to have 8 very simple cores (albeit with advanced SSE4 units with even wider vector instructions such as 256 or 512 bits) and allow each core to run 2 or 4 threads. This won't be hyperthreading as hyperthreading is a form of SMT (although Intel may reuse the name). It will be a form of fine-grained multithreading that allows context switches on L1 or L2 cache misses, as well as other latent operations. Of course their will also be logic to allow all the threads to run equally.

Apart from SSE4, you've just described Suns' Niagara processor. See this blog entry of Jonathan Schwartz's for some nice pictures.

--paulj

Since IBM and Sun don't write word processing software anymore

Both are current products even if they don't get much shelf space or press attention.

$349 is way too expensive.

I can use a $199 Linux PC and boot ThinStation and get the same results, minus the smart-card. Granted, it will take up more space, but for $150 savings....

No, what I want is a minimal-cost, dirt-cheap thin client, one that ONLY contains a "just fast enough" processor and graphics chip and "just enough" i/o to get the job done.

Surely such a thing can be made to retail for under $40 each if produced in the millions, not counting monitor, keyboard, and mouse. Add $5 if it's integrated into a cheap keyboard.

Since IBM and Sun don't write word processing software anymore (AFAIK)

Really? *cough*SmartSuite*cough*StarOffice*cough*

I think its more likely that the patent is targeted at the various free word processing programs.

I don't think it's actually *targeted* at anything. Targeting something with a patent would imply that the feature exists. (Which would invalidate the patent.) Instead, Microsoft is simply building a large portfolio. The idea is that if they cast a large enough net, they can eventually threaten any would-be attacker with hundreds of vague claims. While none of them would probably hold up in court, the claims would tie things up for long enough to bankrupt or entirely block the attacker.

Its there in the text if you RTFA.

"Early this year, however, the agency reviewed its IT infrastructure as part of an effort to meet performance targets, comply with Scotland's Freedom of Information Act and work more closely with other law enforcement groups. Following the review, and a follow-up study in March, it decided to switch back to Microsoft."

"Help from Microsoft in other areas may have influenced the decision. The company plans to work with Scottish police to develop an electronic document management system to help it comply with requests made under the 2002 Freedom of Information Act, and a document sharing system for police staff, Microsoft said."

So its a showcase, they get their freedom of information system (but only if built around MS stuff), Microsoft gets a switcher-backer to advertise with. Overall the Police force saves, and MS gets less money, but MS gains a nice PR show.

And why was this obscure department one chosen:

"The agency's decision is something of an embarrassment for Sun, which played up the deal with Central Scotland Police as a StarOffice success story. It devoted a Web page to it at http://uk.sun.com/sunnews/success/public_sector/10 2.html"

So its no different than the other convoluted deals. Do you recall when Windows 2003 suddenly spiked in success as domains switched from being hosting on Linux to Windows? But it was just a large ISP switching domain names on hold, once the deal expired, they switched it back.

Plenty of times. Off the top of my head:

1) The ability to find the size of a filesystem without resorting to JNI to execute stat_vfs(). http://bugs.sun.com/bugdatabase/view_bug.do?bug_id =4057701 Claims to be fixed but fix is irrelevant until our J2EE app server upgrades to that version -- Sun's fix is scheduled for release in 2Q 2006.

2) Erroneous OutOfMemoryErrors being thrown under Solaris on Sun JDK (1.2.x, 1.3.x) for our J2EE app (lots of EJBs): http://bugs.sun.com/bugdatabase/view_bug.do?bug_id =4390238 Had to use the workaround described in the bug. We encountered the bug and could faithfully reproduce it two years after it was closed as "not reproducible".

3) JDK 1.2.2: HttpURLConnection.getErrorStream() returned null because the Sun developer never implemented the method. I had to re-implement an HTTP client just to have this function (and to avoid other problems in the API such as throwing FileNotFoundException for 404's).

4) JVM for our app server (1.2.2, 1.3.1) on linux cored when DISPLAY was exported over ssh.

5) JVM 1.3.x cored on Solaris when using DBCS Japanese locale. 1.2.x cored when DBCS X11 fonts were missing.

We tended to see new reproducible JVM crash cases (server side only, J2EE, 1.2.x, 1.3.x, IBM and Sun JVMs, AIX Linux Solaris Windows) about once a month. Sometimes it was the mix in with native database drivers and app server plugins, other times DBCS locales, many times just J2EE long-lived server having to do weird desktop-isms (chart generation was a huge pain before Headless AWT).

YMMV. Reliability has improved over time.

Plenty of times. Off the top of my head:

1) The ability to find the size of a filesystem without resorting to JNI to execute stat_vfs(). http://bugs.sun.com/bugdatabase/view_bug.do?bug_id =4057701 Claims to be fixed but fix is irrelevant until our J2EE app server upgrades to that version -- Sun's fix is scheduled for release in 2Q 2006.

2) Erroneous OutOfMemoryErrors being thrown under Solaris on Sun JDK (1.2.x, 1.3.x) for our J2EE app (lots of EJBs): http://bugs.sun.com/bugdatabase/view_bug.do?bug_id =4390238 Had to use the workaround described in the bug. We encountered the bug and could faithfully reproduce it two years after it was closed as "not reproducible".

3) JDK 1.2.2: HttpURLConnection.getErrorStream() returned null because the Sun developer never implemented the method. I had to re-implement an HTTP client just to have this function (and to avoid other problems in the API such as throwing FileNotFoundException for 404's).

4) JVM for our app server (1.2.2, 1.3.1) on linux cored when DISPLAY was exported over ssh.

5) JVM 1.3.x cored on Solaris when using DBCS Japanese locale. 1.2.x cored when DBCS X11 fonts were missing.

We tended to see new reproducible JVM crash cases (server side only, J2EE, 1.2.x, 1.3.x, IBM and Sun JVMs, AIX Linux Solaris Windows) about once a month. Sometimes it was the mix in with native database drivers and app server plugins, other times DBCS locales, many times just J2EE long-lived server having to do weird desktop-isms (chart generation was a huge pain before Headless AWT).

YMMV. Reliability has improved over time.

Schwartz, is that you?

<:import foo:>
<table>
    <:for `foo.Users.getSome()` u:>
        <tr><td><:val `u.username`:></td></tr>
    <:/for:>
</table>

On the subject of using MySQL with J2EE, Sun recently submitted some SPECjAppServer2004 benchmarks using MySQL and Sun Java system App Server 8.1. http://www.theserverside.com/news/thread.tss?threa d_id=35375 Before anyone starts criticizing benchmarks - the submissions demonstrate a couple of things : 1. You can get some great performance with FREE 2. MySQL has past the strict J2EE Compatability Test Suite - demonstrating that it is more than capable of supporting J2EE (wrt. portability) 3. That MySQL has it's place, as does Oracle - in terms of price / performance / features - though I think the feature argument gets less important with every release. Rich Sharples Sun Microsystems http://blogs.sun.com/roller/page/sharps/

Presentation and blog posting from Bryan Cantrill - here

I love reading these junk posts from self declared experts in CPU architecture, if only there were in charge the world would be perfect ...

He's right. Go look at the power requirements and rack footprints of Sun and IBM's latest.

Compare this 2U quad Dell that draws 2400 watts with this Sun 2U 4-way that with 4 dual cores has twice as many CPUs but still only draws 760W.

Gee, twice as many CPU cores and less than 1/3 the power (and therefore heat).

Don't you feel fucking stupid now?

1) Sounds more like a criticism of Swing, but maybe I misunderstood you

2) You really want every class to include a description of Objects methods? I think it's better to use @inheritDoc where appropriate, but otherwise not to display inherited comments.

3) it's -private, not -showprivate

1) Sounds more like a criticism of Swing, but maybe I misunderstood you

2) You really want every class to include a description of Objects methods? I think it's better to use @inheritDoc where appropriate, but otherwise not to display inherited comments.

3) it's -private, not -showprivate

• Sun's Java JRE's license specifies that redistribution of this software is only allowed if it is "bundled as part of, and for the sole purpose of running, your Programs". Making a copy of the JRE as part of an entire copy of a GNU/Linux distribution seems to serve a different purpose.
• Acrobat Reader's license allows you to redistribute the software but requires you to keep no copies: "You may not, rent, lease, sublicense, assign or transfer your rights in the Software, or authorize all or any portion of the Software to be copied onto another user's computer except as may be expressly permitted herein. You may, however, transfer all your rights to Use the Software to another person or legal entity provided that: [...] (b) you retain no copies, including backups and copies stored on a computer [...]".

Just switched to Sun from Dell for some low end rack mount x86 servers. Check these and see if they work out for you. Did for me...

For details on the network stack improvements, start here:
http://www.sun.com/bigadmin/content/networkperf/

For details on the network stack improvements, start here:
http://www.sun.com/bigadmin/content/networkperf/

Much to my surprise, Sun has become the preferred source for servers in our data center. The reason is the availablity of AMD Opteron servers from Sun. These are replacing Intel Zeon based servers from HP and IBM. They're running Linux, of course.

You'll have to deal with priort art

That's just trademark infringement.

You'll have to deal with priort art

how about the term SunRay (I am posting this from a SunRay. They are kinda cool. Quiet, too.)

Absolutely! I was all hot to get a Mini, but figured I'd have to spend close to US$800 to get one I consider "usable" (512+MB, WiFi). Ultimately I just ordered one of these, but I still might get one for my wife.

Ahem.

Open Firmware!!!!!!

Thank you.

The "J2EE 5" link in TFA links to OpenSolaris, not to http://java.sun.com/j2ee/5.0/index.jsp or http://www.jcp.org/en/jsr/detail?id=244 .

Sun actually supports Harmony and said so the day it was announced - VNU has it wrong (as usual). Congratulations, you found the only report that neglected to mention that both the senior Sun engineer responsible for Java strategy and the person responsible for future open source strategy at Sun both expressed support for harmony at a level pretty similar to IBM's support ("jolly good, and we may contribute at some point in the future"). While Gosling's misreported off-the-cuff comments make for good tabloid sensationalism they don't represent Sun's official view.

OOP there it is.
OOP there it is....
OOP
OOP

Actually it now called Java 2 version 5.0 and sometimes version 1.5.0 See http://java.sun.com/j2se/1.5.0/docs/relnotes/versi on-5.0.html for the official word. Not that this makes it any less confusing.

or could be a Java user. :)
http://java.sun.com/j2se/naming_versioning_5_0.htm l

Then why are you wrong... Lets start.....

First of all, the cost of virtual memory. As we all know, it is not free. Google "virtual memory cost TLB percent" and hit "I feel lucky", you should come to this page...

http://developers.sun.com/solaris/articles/optimiz ing_apps.html

This is talking about a common case where an app causes a lot of TLB misses, using a substantial proportion of the execution time doing nothing but VM translation. Notice that the performance penalty is (in their example) roughly 30%, due to TLB misses causing floating point pipeline stalls. This is a commonly accepted figure. EVERY TLB miss will cause a few hundred clock cycles. A TLB miss rate of even 1% will have a SUBSTANTIAL effect on performance. Running without Virtual Memory (in system mode) completely eliminates the TLB and all of this performance is reclaimed. This is possible with any truly typesafe system, such as JAVA.

Secondly, compilers..... Yes, some of the compiler/verifier is on the client, however you could just as easily say that the hardware underneath is a compiler itself (translating CISC x86 instructions to RISK opcodes), so this terminology isn't particularly meaningful. It becomes totally misleading when you consider that there exist hardware platforms that can execute java bytecode directly. My meaning should have been clear from context to someone of your credentials. In any case, most of the checks are done within the compiler (Hotspot), though that compiler happens to reside on the client machine. If you dislike this being referred to as a compiler, then that is your business.

Third, you include some extremely bad examples of runtime checks. Modern Virtual Machines already eliminate a lot of array bounds checks with compiler passes (though granted, this particular compiler pass is performed on the client machine), and null pointer checks are handled by a hardware interrupt, just like an integer divide by zero. Both of the examples you gave are very unfortunate, as they are two checks that rarely actually must be performed by software.

As for your question of whether or not the software checks in java are more expensive than the hardware means, there is very little direct research available, but Java is commony accepted to be around 95% of the performance of straight C, and Virtual Memory is commonly accepted to be about 70% of the performance of System mode, so you tell me. For instance, see...
http://www.idiom.com/~zilla/Computer/javaCbenchmar k.html

I wouldn't have assumed you were junior if you didn't make such classic mistakes.

Once again, it is very foolish to make claims about my chip designing skills. Everyone who has investigated the issue accepts that modern CPUs do the best they can GIVEN that they are running unsafe languages. It is also commonly accepted that dramatic efficiencies could be realized (I just gave one example, no TLBs needed, saving around 30%) if the software itself could be proven safe. It would be another mistake expected of a novice confuse this rather clear cut situation and claim that I was calling the Intel designers idiots.

Who knows, perhaps in your zeal to call me out as a fool you skimmed over some implications and said things that didn't sound particularly well founded.

However, since you asked... I'm a BA in Physics from an Ivy League institution. I currently work in computational finance. I spend a great deal of time working with java, and occasionally native code. We are largly performance bound, so I have studied this issue extensively.

I meant to say "install and use", but you are correct. Also, given your suggestion, Solaris 10 offers all of what you ask, only it does not [yet] come preloaded on super low-ball hardware....although there is the Sun Ultra 20 for a great price.

Still, I wish that somebody would build new workstation-quality computers that had an elegant 64-bit RISC architecture, kind of like the Power Mac G5...

...Remember Alpha, SPARC, PowerPC, Motorola 68k, and PA-RISC? I wish that we had this diversity in chipsets again.

Ummm... I take it you've never played with one of these.

I am sad however that I missed out on the Enterprise 450 because I didn't hear what they said it was in time. It went for ..........$85 arrrrrgh

You can get still a Sun UltraSPARC for US$1395. The elegance will be present unless you're programming in assembly. The processor will be completely abstracted from you and the only difference will be that the computers are faster, especially the laptops. Just pretend it's a better chip.

...would be quite comfortable with reformatting.

Oh, and he appears to be an exec at Sun. Maybe he's one of those academics who just doesn't know his way around a computer, no matter what his education is.

It all depends on the application, of course, but I apologize for ranting there.

Sun's opteron offerings look promising, and so do their new lines that are slated to come out in the future (Niagara, for instance) http://blogs.sun.com/roller/page/jonathan/20040910 #the_difference_between_humans_and

Way to go Sun. Looking at the pictures, it seems that it has dual opterons. Awesome.

Here is more info on the v20z server:
http://www.sun.com/servers/entry/v20z/index.jsp/

all except the Alpha were RISC chips

So what do you consider Alpha to have been? (I consider it to have been a RISC processor.)

The PowerPC will disappear from computing in 2007.

So much for the pSeries and iSeries servers, I guess....

(Presumably by "computing" you're referring to personal computers, although I'm not sure why you're including SPARC in that category; from your reference to Sun workstations, perhaps you mean "desktop computers", or perhaps, at least with low-end Sun workstations, "computers ordinary people might be able to buy".)

Of course, there are always IBM workstations if you want a POWER-family processor (a family that includes PowerPC). They're a lot more expensive than the cheapest Sun SPARC workstation, however.

His blog is empty.

assuming everything is setup properly and has been designed for integrating in this manner

but even if it's not setup correctly, it's so easy to install you can fix the problem in a jiffy!

thanks sun!

I have been looking around for info to build such a thing. I'd like to have 1TB of raid-protected storage for a digital video recorder. Performance is not an issue, but I would like it to be quiet, lowpower, and not too bi

Perhaps this is what you are looking for ?

I'm going to go out on a limb an guess that this means that Apple isn't going to take Jonathan Schwartz up on his offer that they adopt Solaris 10 for the underpinnings of the Mac OS and adopt NetBeans as their development environment...

Shocking, I know.

Oye vey...

--
http://joshstaiger.org/

SPARC extinct ... really ?

http://www.sun.com/servers/ultrasparciv/

And perhaps more importantly:

http://blogs.sun.com/roller/page/jonathan?entry=th e_difference_between_humans_and

SPARC extinct ... really ?

http://www.sun.com/servers/ultrasparciv/

And perhaps more importantly:

http://blogs.sun.com/roller/page/jonathan?entry=th e_difference_between_humans_and

First check this graphics: http://blogs.sun.com/roller/resources/JeffV/itaniu m_rev_ext.gif Itanium processor was pushed so badly in the past, there are 9 IDC estimates for the processor reenues. All those 9 estimates were terribly wrong. how come IDC makes 9 consecutive wrong estimates for a product? Itanium deserves al kind of jokes and bad remarks IMHO. - it is still an expensive product. Maybe CPU's are relatively cheap now but systems designed for it are at least double the price of an equally performing Opteron systems. Motherboards, chipsets are expensive. - PErformance numbers you presented from super computers does not reflect realities. For super computers performance metrics are depends on so many parameters (inter connection, application type etc). Nobody can claim itanium is the best bang for buck. - it is a power sucker. - Still no good native software support, no good OS support, people are reluctant to use Virtual machines either (Java etc.)

Yeah cause SWT is part of the java standard library. Perhaps you were thinking of AWT or Swing?