COMMON, Twitter and i

If you’re attending the COMMON Annual Meeting and Exposition in Orlando next week, or if you just want to stay up to date with the many things going on related to IBM i and the conference, consider using Twitter.

For the uninitiated, Twitter is used to broadcast very short messages – 140 characters or fewer. People then “follow” others to get their messages. If you want to read my messages, for example, you “follow” me using the Twitter site, and you’re shown new messages as I send them. When someone puts out a message using Twitter, it’s called a “tweet” and the person is said to be “tweeting.”  (There, now you’re initiated!)

I’ve been on Twitter for a couple of years now.  The Twitter ID I use for IBM-related tweets is Steve_Will_IBMi. Dawn May will also be tweeting; she's DawnMayiCan. We plan to tweet about events, especially if there are changes or ad hoc gatherings.

In addition to following specific people, you can also follow specific topics. When people tweet, they put topic tags in their tweets, and then people can use Twitter to keep up to date on the latest news about those topics. Two we’re already using are #ibmi and #COMMONUG. 

To set up a Twitter account:

• Go to Twitter and sign up. Choose a short, simple user name.
• Under “Profile” you can add your picture and other customizable features.

Once you’ve done that, find people to follow. Maybe the easiest way for an IBM i Twitter user to get started is to enter #ibmi in the search field, and then see who’s tweeting about IBM i. Then you can click on those people and start following them. Following is as simple as clicking the “Follow” icon near the the person’s picture on their profile page. Oh, that brings up another point – organizations can also use Twitter, so you’ll be able to follow IBM Systems Magazine’s IBM i twitter account, IBMimag, as well as other interesting organizations.

This will be our first conference trying out Twitter, so we’ll see how it goes. If you’re trying it out, too, you can give us feedback when it’s done. You can even use Twitter to do it! 

See you at the conference!

Exploiting POWER7 with Software on IBM i

Our colleagues on the IBM Rational team have been fielding questions about the new capabilities of the POWER7 processors and how IBM i software developers can take advantage of them. I’d like to thank Steve Hikida, and Wang Chen and Philip Luk on his team, for this helpful information in Q&A format. Now that IBM i 7.1 is announced, I can publish these to help all of us remember – or at least know where to find – the answers.

Q1: How do I maximize the exploitation of the POWER7 architecture?
The “Adaptive Code Generation” feature allows the ILE compilers to take advantage of architecture-specific optimization. Leveraging this technology, the ILE compilers shipped with IBM i 7.1 enable the applications to take advantage of POWER7 technology-specific optimization.

Q2: What is the Adaptive Code Generation feature?
Adaptive code generation allows improved performance without the need to hand-tune code.  Optimizations and performance tuning is performed to take advantage of new POWER7 features and the resulting generated instructions are tuned to maximize performance from the POWER7 processor.

Q3: Where can I learn more about “Adaptive Code Generation?”
Please refer to “IBM i ILE Concepts Version 7 Release 1."

Q4: What compilers support and exploit POWER7?
ILE C/C++, ILE COBOL and ILE RPG compilers are shipped with IBM i 7.1.

Q5: Which compiler do I need to exploit POWER7?
The ILE C and C++ compilers that are shipped with IBM i 7.1 have the POWER7 support. The support is only available on the 7.1 release and isn’t available on previous releases.

Q6: Do I need recompile?
You don’t need to recompile your application to exploit POWER7. 
   To take advantage of POWER7 architecture, you can use LICOPT CodeGenTarget=Power7 option to recompile your code. Alternatively, you can use the Change Program (CHGPGM) or Change Service Program (CHGSRVPGM) command with the CodeGenTarget=Power7 parameter to directly convert the existing program to take advantage of the POWER7 architecture.

Q7: If I have a mixed environment of POWER5, POWER6, POWER7 systems, what should I do?
The Adaptive Code Generation feature enables you to tune your application to many different architecture levels without the need for recompilation.

Q8: Does ILE compiler support auto-parallelization or auto-vectorization?
No. The ILE compilers don’t currently implement any auto-parallelization or auto-vectorization features.

Q9: What is PASE and how does it relate to POWER7?
The Portable Application Solutions Environment (PASE) allows the AIX compiler to be run on IBM i, so all of the POWER7 features within the new AIX compilers can be exploited under PASE.

Q10: When PASE might not be the best solution?
PASE is generally not a good choice for code that provides a large number of callable interfaces that must be called from ILE or has any of the following characteristics:
   • Code that needs higher performance call and return than provided by either starting or ending PASE on each call or by calling a PASE procedure in an already-active PASE program.
   • Code that needs to share memory or namespace between an ILE caller and the library code. A PASE program doesn’t implicitly share memory or namespace with ILE code that called it. (However, ILE code that’s called from PASE can share or use PASE memory.)

Check out the entire 7.1 Information Center, which went live on April 23, when IBM I 7.1 became generally available.
 

IBM PowerHA SystemMirror for i

As promised in last week’s article introducing IBM i 7.1, I’ll be publishing a series of articles that give more details about each of the major features I described on announce day. Today’s article was written by Jenny Dervin, the architect for PowerHA on IBM i.  Thanks, Jenny, for the great information.  Oh, and readers, be sure to catch the note at the bottom about a webcast where you can hear more about this product.

Each IBM i release introduces major enhancements, and PowerHA is no exception. For starters, its name has been updated in 7.1 to be PowerHA SystemMirror for i. This name better reflects the underlying technology behind the product. It has also been split into two separate editions. PowerHA SystemMirror for i Standard Edition includes technology designed for a data center’s high-availability solution, such as synchronous geographic mirroring, FlashCopy, and IBM’s new LUN-level switching technology. PowerHA SystemMirror for i Enterprise Edition includes technology suitable for longer-distance disaster recovery and larger-scale configuration – Metro Mirror, Global Mirror and asynchronous geographic mirroring.

A major enhancement in this release is support for asynchronous geographic mirroring. This support will allow greater distance between the production and backup systems.  With synchronous geographic mirroring, disk writes aren’t considered complete until the write arrives on the backup system. This can cause longer end-user response times as the production and backup systems are separated by a greater distance. In 7.1, with asynchronous geographic mirroring, the end-user response time isn’t affected by the distance between production and backup. IBM i asynchronously sends the writes to the backup, allowing the local write to complete without waiting.

LUN-level switching is a new option available for customers using the System Storage DS6000 or DS8000. The iASP is stored on a set of LUNs on the storage server, and that set of LUNs can be switched between partitions or systems. Customers can use LUN-level switching for local high availability in conjunction with either Metro Mirror or Global Mirror, which provides coverage for storage failures and disaster recovery.

Space efficient FlashCopy is another technology supported in PowerHA SystemMirror 7.1 for DS8000 customers. Space-efficient FlashCopy allows a point-in-time copy with potentially less storage used for the copy.  

In 7.1, PowerHA SystemMirror provides the capability to register with the Hardware Management Console (HMC) or Integrated Virtualization Manager (IVM) for partition and CEC failures. This allows PowerHA to better detect node failures and provide automatic failover when desired.

Additional commands are now available within SystemMirror to retrieve information about the cluster, cluster resource groups, sessions and copy descriptions. This set of commands can be used to automate cluster management.

Other SystemMirror enhancements include complete support for IPv6 addresses throughout the product, and improved performance for the full synchronization process within geographic mirroring.

For more information on any of these 7.1 enhancements, see the IBM i Information Center High Availability.  The Information Center will be available when 7.1 is generally available on April 23.

This set of enhancements will provide additional replication options for our customers, increased reliability and enhanced usability. If you have not yet investigated the use of PowerHA SystemMirror within your business, now is the time to do so.   

Webcast Note:
If you want more information about High Availability solutions, attend the April 27th webcast, "Resiliency without Downtime: High Availability solutions for IBM i and AIX.” Understand the value of the latest PowerHA releases, explained by IBM product manager Steve Finnes, and supported with proof points provided by Jim Horn, IT Consultant to CATCO, a leading U.S. trucking industry parts and service supplier, and PowerHA customer.  

You and i 7.1

Well, it’s here!  IBM i 7.1 is announced today. Over the next few months, we’re going to be blogging, writing articles, conducting Webcasts, and presenting in person on the great things contained in IBM i 7.1.

My job today is to hit the highlights of the release. These are the big-ticket functions that will be the focus of the marketing.  But, as with every release, there are many functions that we include that don’t make the “highlights” list, but which might be a highlight for you. To get an idea of what those items are, be sure to visit Dawn May’s “i Can” blog today. 

So, without further ado, here they are:

The IBM i 7.1 Highlights

•    DB2 for i: XML datatype, column-level encryption and more 

•    PowerHA SystemMirror for i: asynchronous geographic mirroring, LUN-level switching and more

•    Solid state drives: automatic data movement

•    Rational Open Access RPG edition : New, open, modernization options for RPG

•    Zend Server Community Edition: For IBM i, shipped with IBM i.

•    IBM Systems Director Navigator for i: New Web-based management capabilities.


Now, each of these topics is going to be covered in much greater detail, either in this blog, or in articles to which I’ll be able to point, but here are a few words on each to introduce the topic.

DB2 for i: XML datatype, column-level encryption and more.  Modern applications, especially those that have to exchange data across the Internet, often must consume data described with XML.  In 7.1, DB2 for i adds support for XML with a new datatype, xml parsing, annotation (converting xml to traditional database data) and schema validation, allowing application writers to integrate XML and DB2 in a simple and powerful way. Additionally, DB2 for i, using the Omnifind product, can perform xml-aware data searches in an XML column.  DB2 for i can also create XML directly from the data it contains.  How does all of this work?  Well, we’ll get you a good description of that in a few weeks. We’ll also show you how you can add the capability to encrypt the data stored in a particular column.  Oh, and there’s a performance improvement awaiting many of you because we’re extending our high-performance query engine to cover logical files.  This is good stuff!  Stay tuned.

PowerHA SystemMirror for i: asynchronous geographic mirroring, LUN-level switching and more.  PowerHA SystemMirror for i is the hardware replication high availability product we introduced first with IBM i 6.1. The new release enhances our geographic mirroring support by the introduction of an asynchronous protocol, which allows the mirrored and mirroring sites to be much farther away from each other. The 7.1 release also works with the IBM DS8000 and IBM DS6000 to allow data stored in IASPs controlled by these storage systems to switch from one IBM i to another, allowing data to remain available even when the primary IBM i system becomes unavailable. Again, we have the experts for this technology writing detailed articles on this function.

Solid state drives: automatic data movement. For a year IBM i has supported solid state drives.  I blogged about that support, and the architectural advantage that IBM i has in making good use of SSDs, and now we have a chance to incorporate that support in a major release, and enhance it with the ability of the operating system to move user data to SSDs automatically when the integrated storage management determines your system performance will benefit from the movement.

Rational Open Access RPG edition: New, open, modernization options for RPG. Here’s an enhancement that’s really an architectural change that opens RPG up to whole new uses.  And it’s being brought to the community by IBM i, our friends in IBM Rational, and several ISVs who’ve been working with IBM to hit the ground running. The idea behind the new capability is to remove the necessity for all output from RPG programs to be forced into the limited 5250 data stream. Instead, it can be passed to handlers, which can transform it in any way the handler wants.  Wrap it in XML; put it in the format required by a pervasive device; make a Web page that doesn’t look like a 5250 screen; whatever you want. And, while this is being released in conjunction with IBM i 7.1, the capability will work with the 6.1 release as well.

Zend Server Community Edition: For IBM i, shipped with IBM i. The new Zend Server Community Edition can do everything Zend Core could do, and more. Now it’s loaded with every 5.4, 6.1 or 7.1 release of IBM i you receive, at no additional charge. If you haven’t started using PHP yet, well, go ahead and get started.  It’s in there and better than ever!

IBM Systems Director Navigator for i: New Web-based management capabilities. Dawn covered the various management tools for IBM i in her blog recently. In particular, the IBM Systems Director Navigator for i, the Web-based interface to manage IBM i, is enhanced with additional capabilities for journal, database, performance-data investigator and tape as well as considerably more capability for plug-in products BRMS, advanced job scheduler (JS1) and omnifind (OMF). Brand-new capabilities such as set target system and an integrated command (5250) emulator round out a rich delivery of new Web function. Some of these functions were previously only available in the client-based (Windows OS-based or Linux technology-based) Navigator product and several are brand new. As Dawn mentioned, the strategic direction for management tools is browser-based, so it was very important to our customers that we add these critical functions to the tools that will be the focus of future development.

As I said before, these are the highlights, but they’re not the entire release – not at all!  And I’ve only been able to scratch the surface of each of the highlights. Over the next few months, Dawn and I will bring you descriptions of each of these, or we’ll point you to other places where our technical experts are writing articles describing them – or both!

These are exciting times. Stay tuned!

What do You and i have in COMMON?

As I mentioned a few weeks back, I’ve been getting ready for COMMON 2010. We’re less than a month away from the event now, and my presentations are starting to coalesce. As I tweeted a few days ago, the closer COMMON gets, and the more I see I have scheduled for that week – May 1-2 with the COMMON Americas Advisory Council and then May 3-6 at the conference itself – the more excited I get.

I get to present four sessions on my own, and one in combination with other IBM i experts. 

“What sessions are those?” you may ask.  Well, thanks for asking!

First of all, you may have caught wind of a new release of IBM i due out soon. I don’t know where in the world you would’ve gotten that idea, but it just might be true. And, if it’s indeed true, I would probably be a good person to have a session covering that new release. In fact, it’s likely that I would be asked to give that session twice, just to ensure people had a chance to hear about it.

Second, I wrote in an earlier blog this year that Software as a Service (SaaS) is a big deal in the industry, and that IBM i is a great platform for SaaS. To put my money where my mouth is (or maybe I’m just putting my mouth where my mouth is?) I have a session titled “Software as a Service (SaaS) on IBM i -- Architecture and ISV Solution Examples.” We’ve been talking to several ISVs about SaaS this year, validating our belief about the suitability of IBM i for that kind of workload, and learning things from them about specifically what they’re doing. I’ll cover what SaaS is, what its general architectures are, how the architecture of IBM i can be used to do SaaS, what our ISVs are doing, and how this relates to the technologies typically labeled “cloud.”

Third, I have what is probably my favorite presentation to give: “Why i?” It’s a common occurrence for those of us who use this platform – we get asked to defend our choice. I have built this presentation to help IBM i advocates provide answers to that question, with different kinds of answers depending on the kind of person asking. Is it a technical person? I’ll show the basic architectural reasons. Is it a business person? There are good business reasons to select i. Is it someone who thought IBM i is no longer relevant? I have answers there, too.

Also, I will be part of an “IBM i Q&A Session,” along with other IBM i experts. This is a session where we take questions about anything IBM I platform-related and provide answers on the spot. We get stumped sometimes, but if we do, we follow up later. Its format is very open, and we appreciate the chance to interact with others in the community.

Speaking of interacting with the community, as you know if you’ve ever been to COMMON, there are chances to “interact” every evening, too. While not one of my scheduled sessions, I have definitely been known to show up at these.

And finally, though it is not specifically assigned to me, there is the annual session for the “Young i Professionals (YiPs),” and since being invited to that session a few COMMON meetings back, I can’t resist attending to meet the young and young-at-heart in our IBM i community.

For anyone who’s still on the fence about whether to attend, I want to encourage you to attend. Even if none of my sessions sound interesting, I guarantee that there are plenty of great speakers giving excellent presentations on topics from low level programming to negotiating to humor in the workplace.

I hope to see you there!   It wouldn’t be the same without You (and i).

(P.S. I mentioned “tweeting.”  If you want to follow my IBM twitter account, I am Steve_Will_IBMi.  In addition to blogging, I intend to tweet a bit.)