For classic NUMA-aware programming I'm typically most concerned about simple cold, capacity and compulsory misses and whether we can satisfy the miss by locally connected memory or whether we have to pull the line from its home node over the coherent interconnect -- we'd like to minimize channel contention and conserve interconnect bandwidth. That is, for this style of programming we're quite aware of where memory is homed relative to the threads that will be accessing it. Ideally, a page is collocated on the node with the thread that's expected to most frequently access the page, as simple misses on the page can be satisfied without resorting to transferring the line over the interconnect. The default "first touch" NUMA page placement policy tends to work reasonable well in this regard. When a virtual page is first accessed, the operating system will attempt to provision and map that virtual page to a physical page allocated from the node where the accessing thread is running. It's worth noting that the node-level memory interleaving granularity is usually a multiple of the page size, so we can say that a given page P resides on some node N. That is, the memory underlying a page resides on just one node.

But when thinking about accesses to heavily-written communication variables we normally consider what caches the lines underlying such variables might be resident in, and in what states. We want to minimize coherence misses and cache probe activity and interconnect traffic in general. I don't usually give much thought to the location of the home NUMA node underlying such highly shared variables. On a SPARC T5440, for instance, which consists of 4 T2+ processors connected by a central coherence hub, the home node and placement of heavily accessed communication variables has very little impact on performance. The variables are frequently accessed so likely in M-state in some cache, and the location of the home node is of little consequence because a requester can use cache-to-cache transfers to get the line.

Or at least that's what I thought. Recently, though, I was exploring a simple shared memory point-to-point communication model where a client writes a request into a request mailbox and then busy-waits on a response variable. It's a simple example of delegation based on message passing. The server polls the request mailbox, and having fetched a new request value, performs some operation and then writes a reply value into the response variable. As noted above, on a T5440 performance is insensitive to the placement of the communication variables -- the request and response mailbox words. But on a Sun/Oracle X4800 I noticed that was not the case and that NUMA placement of the communication variables was actually quite important.

For background an X4800 system consists of 8 Intel X7560 Xeons . Each package (socket) has 8 cores with 2 contexts per core, so the system is 8x8x2. Each package is also a NUMA node and has locally attached memory. Every package has 3 point-to-point QPI links for cache coherence, and the system is configured with a twisted ladder "mobius" topology. The cache coherence fabric is glueless -- there's not central arbiter or coherence hub. The maximum distance between any two nodes is just 2 hops over the QPI links. For any given node, 3 other nodes are 1 hop distant and the remaining 4 nodes are 2 hops distant.

Using a single request (client) thread and a single response (server) thread, a benchmark harness explored all permutations of NUMA placement for the two threads and the two communication variables, measuring the average round-trip-time and throughput rate between the client and server. In this benchmark the server simply acts as a simple transponder, writing the request value plus 1 back into the reply field, so there's no particular computation phase and we're only measuring communication overheads. In addition to varying the placement of communication variables over pairs of nodes, we also explored variations where both variables were placed on one page (and thus on one node) -- either on the same cache line or different cache lines -- while varying the node where the variables reside along with the placement of the threads. The key observation was that if the client and server threads were on different nodes, then the best placement of variables was to have the request variable (written by the client and read by the server) reside on the same node as the client thread, and to place the response variable (written by the server and read by the client) on the same node as the server. That is, if you have a variable that's to be written by one thread and read by another, it should be homed with the writer thread. For our simple client-server model that means using split request and response communication variables with unidirectional message flow on a given page. This can yield up to twice the throughput of less favorable placement strategies.

Our X4800 uses the QPI 1.0 protocol with source-based snooping. Briefly, when node A needs to probe a cache line it fires off snoop requests to all the nodes in the system. Those recipients then forward their response not to the original requester, but to the home node H of the cache line. H waits for and collects the responses, adjudicates and resolves conflicts and ensures memory-model ordering, and then sends a definitive reply back to the original requester A. If some node B needed to transfer the line to A, it will do so by cache-to-cache transfer and let H know about the disposition of the cache line. A needs to wait for the authoritative response from H. So if a thread on node A wants to write a value to be read by a thread on node B, the latency is dependent on the distances between A, B, and H. We observe the best performance when the written-to variable is co-homed with the writer A. That is, we want H and A to be the same node, as the writer doesn't need the home to respond over the QPI link, as the writer and the home reside on the very same node. With architecturally informed placement of communication variables we eliminate at least one QPI hop from the critical path.

Newer Intel processors use the QPI 1.1 coherence protocol with home-based snooping. As noted above, under source-snooping a requester broadcasts snoop requests to all nodes. Those nodes send their response to the home node of the location, which provides memory ordering, reconciles conflicts, etc., and then posts a definitive reply to the requester. In home-based snooping the snoop probe goes directly to the home node and are not broadcast. The home node can consult snoop filters -- if present -- and send out requests to retrieve the line if necessary. The 3rd party owner of the line, if any, can respond either to the home or the original requester (or even to both) according to the protocol policies. There are myriad variations that have been implemented, and unfortunately vendor terminology doesn't always agree between vendors or with the academic taxonomy papers. The key is that home-snooping enables the use of a snoop filter to reduce interconnect traffic. And while home-snooping might have a longer critical path (latency) than source-based snooping, it also may require fewer messages and less overall bandwidth. It'll be interesting to reprise these experiments on a platform with home-based snooping.

While collecting data I also noticed that there are placement concerns even in the seemingly trivial case when both threads and both variables reside on a single node. Internally, the cores on each X7560 package are connected by an internal ring. (Actually there are multiple contra-rotating rings). And the last-level on-chip cache (LLC) is partitioned in banks or slices, which with each slice being associated with a core on the ring topology. A hardware hash function associates each physical address with a specific home bank. Thus we face distance and topology concerns even for intra-package communications, although the latencies are not nearly the magnitude we see inter-package. I've not seen such communication distance artifacts on the T2+, where the cache banks are connected to the cores via a high-speed crossbar instead of a ring -- communication latencies seem more regular.

Dear Partners,

We would like to inform you that the November '12 issue of our Newsletter is now available.
The issue contains information to the following topics:

News from Corp
Oracle Celebrates 25 Years of SPARC Innovation; IDC White Papers Finds Growing Customer Comfort with Oracle Solaris Operating System; Oracle Buys Instantis; Pillar Axiom OpenWorld Highlights; Announcement Oracle Solaris 11.1 Availability (data sheet, new features, FAQ's, corporate pages, internal blog, download links, Oracle shop); Announcing StorageTek VSM 6; Announcement Oracle Solaris Cluster 4.1 Availability (new features, FAQ's, cluster corp page, download site, shop for media); Announcement: Oracle Database Appliance 2.4 patch update becomes available

Technical Section
Oracle White papers on SPARC SuperCluster; Understanding Parallel Execution; With LTFS, Tape is Gaining Storage Ground with additional link to How to Create Oracle Solaris 11 Zones with Oracle Enterprise Manager Ops Center; Provisioning Capabilities of Oracle Enterprise Ops Center Manager 12c; Maximizing your SPARC T4 Oracle Solaris Application Performance with the following articles: SPARC T4 Servers Set World Record on Siebel CRM 8.1.1.4 Benchmark, SPARC T4-Based Highly Scalable Solutions Posts New World Record on SPECjEnterprise2010 Benchmark, SPARC T4 Server Delivers Outstanding Performance on Oracle Business Intelligence Enterprise Edition 11g; Oracle SUN ZFS Storage Appliance Reference Architecture for VMware vSphere4; Why 4K? - George Wilson's ZFS Day Talk; Pillar Axiom 600 with connected subjects: Oracle Introduces Pillar Axiom Release 5 Storage System Software, Driving down the high cost of Storage, This Provisioning with Pilar Axiom 600, Pillar Axiom 600- System overview and architecture; Migrate to Oracle;s SPARC Systems; Top 5 Reasons to Migrate to Oracle's SPARC Systems

Learning & Events
Recently delivered Techcasts:Learning Paths; Oracle Database 11g: Database Administration (New) - Learning Path; Webcast: Drill Down on Disaster Recovery; What are Oracle Users Doing to Improve Availability and Disaster Recovery; SAP NetWeaver and Oracle Exadata Database Machine

References
ARTstor Selects Oracle’s Sun ZFS Storage 7420 Appliances To Support Rapidly Growing Digital Image Library, Scottish Widows Cuts Sales Administration 20%, Reduces Time to Prepare Reports by 75%, and Achieves Return on Investment in First Year,Oracle's CRM Cloud Service Powers Innovation:Applications on Demand; Technology on Demand,

How to
How to Migrate Your Data to Oracle Solaris 11 Using Shadow Migration; Using svcbundle to Create SMF Manifests and Profiles in Oracle Solaris 11; How to prepare a Sun ZFS Storage Appliance to Serve as a Storage Devise with Oracle Enterprise Manager Ops Center 12c; Command Summary: Basic Operations with the Image Packaging System In Oracle Solaris 11;How to Update to Oracle Solaris 11.1 Using the Image Packaging System,How to Migrate Oracle Database from Oracle Solaris 8 to Oracle Solaris 11; Setting Up, Configuring, and Using an Oracle WebLogic Server Cluster; Ease the Chaos with Automated Patching: Oracle Enterprise Manager Cloud Control 12c; Book excerpt: Oracle Exalogic Elastic Cloud Handbook

• Oracle Reference Architecture: Application Infrastructure Foundation One of the earliest additions to the IT Strategies from Oracle library, this paper describes the concepts and capabilities of the application infrastructure and defines the platform on which solutions are built. Read it.
• Scaling Service Oriented Architecture What is scaling, and what does it mean to a service oriented architecture? Author Philip Wik explores those issues and proposes Oracle-based solutions to SOA scaling and a SOA scaling roadmap. Read it.
• SOA, Cloud, and Service Technologies: A Conversation with Thomas Erl Thomas Erl, the world's best selling SOA author, is joined by Oracle SOA experts Tim Hall and Demed L'Her for a wide ranging four-part conversation on the evolution of SOA and the emergence of the architect in the era of cloud computing. Listen to the Podcast& Read a Transcript
  
• Cloud e-book Invite your customers to download this Cloud e-book, packed with multi-media resources to educate your customers on the value of Oracle Cloud computing.
• Assessment: Are you Leading or Lagging when it comes to SOA and BPM? Take the online SOA Assessment and BPM Assessment.
• New Collateral:
  • Whitepaper Series: The Promise of BPM Technology for Financial Services Institutions - Resource Kit
    
  • Whitepaper: Reaching Process Excellence with Process Accelerators - PDF
    
  • Demystifying Cloud Integration: Whitepapers, webcasts, and customer case studies - Resource Kit
    
  • Whitepaper: Leveraging Governance to sustain Enterprise Architecture - PDF
    
  • Article: Rethink SOA: A Recipe for Business Transformation - Article
    
  • Oracle SOA Resource Kit
    
  • Oracle SOA Governance Resource Kit
    
  • Oracle BPM Resource Kit
    

SOA & BPM Partner Community

For regular information on Oracle SOA Suite become a member in the SOA & BPM Partner Community for registration please visit  www.oracle.com/goto/emea/soa (OPN account required) If you need support with your account please contact the Oracle Partner Business Center.

BlogTwitterLinkedInMixForum

For WebLogic specialized partner we offer free google advertisement to promote your Oracle WebLogic service offerings on your website or your WebLogic events. We will host the complete campaign management.

To create your google campaign please send Jürgen Kress below:

• Your campaign text: 3 lines of text each 35 letters (NOT more letters!)
• Your campaign link: direct link to the website you want to promote

Ideas for the website which we will promote with the google ads:

• Your Oracle WebLogic Service offerings with concrete offering e.g. WebLogic Innovation Workshop
• Oracle WebLogic Specialized Logo
• Your Oracle WebLogic References
• Your WebLogic Implementation consultant with pictures
• Your WebLogic sales contact persons

Example of an WebLogic Specialization text ad:

Oracle SOA Specialized plan to become more agile?
eProseed the Oracle SOA Experts

An interview with Griffiths Waite's Business Development Director highlighting the benefits of the Oracle Specialization Programme.

WebLogic Partner Community

For regular information become a member in the WebLogic Partner Community please visit: http://www.oracle.com/partners/goto/wls-emea ( OPN account required). If you need support with your account please contact the Oracle Partner Business Center.

BlogTwitterLinkedInMixForumWiki

欢迎阅读Oracle数据库支持通讯11月版。本期的内容包括:

最佳实践：针对性能问题的主动型数据收集
Oracle数据库技术更新
支持提示
网上研讨会
补丁更新

下载地址：Oracle数据库支持通讯2012年11月版

Hi all! We are here again to intorduce you some of our new NetBeans 7.3 features. Today we'll show you some another Twig improvements. So let's start!

Code Templates

First feature is about Code Templates. We added some basic templates to improve your Editor experience. You will be really fast with it!

If someone don't know what Code Templates are, they are piece of code (snippet) which is inserted into editor after typing its abbreviation and pressing Tab key (or another one which you define in Tools -> Options -> Editor -> Code Templates -> Expand Template on) to epxand it.

All default Twig Code Templates can be found in Tools -> Options -> Editor -> Code Templates -> Twig Markup. You can add your custom templates there as well.

Note:
Twig Markup code templates have to be expanded inside Twig delimiters (i.e. { and }). If you try to expand them outside of delimiters, it will not work, because then you are in HTML context. If you want to add a template which will contain Twig delimiter too, you have to add it directly into Tools -> Options -> Editor -> Code Templates -> HTML/XHTML. Don't add them into Twig File, it will not work.

Interpolation Coloring

The second, minor, feature is, that we know how to colorize Twig Interpolation. It's a small feature, but usefull :-)

And that's all for today and as usual, please test it and if you find something strange, don't hesitate to file a new issue (product php, component Twig). Thanks a lot!

Effective November 22, the Oracle Endeca Community has been split into separate communities representing individual Oracle Endeca products. The Oracle Endeca Information Discovery Community will fall under the Business Intelligence (BI) category, and can be found here:

https://communities.oracle.com/portal/server.pt/community/oracle_endeca_information_discovery/551.

This community will focus on the Oracle Endeca Information Discovery (OEID) product, formerly known as Endeca Latitude and Endeca Discovery Framework.

The previous Oracle Endeca Community has been renamed to Oracle Endeca Guided Search Community and will focus on discussions around the Oracle Endeca Guided Search product, formerly known as Endeca Infront and Endeca IAP.

The Guided Search Community will continue to be located under the Oracle Commerce Category. Forum threads in the previous Oracle Endeca Community related to Oracle Endeca Information Discovery product have been moved to the new Oracle Endeca Information Discovery Community. We look forward to your continued involvement.

• Ist die Entwicklung mit Oracle ADF schnellerund einfacher als mit Forms, Apex oder .NET?
• Mobile Application Development mit ADF Mobile
• Oracle ADF Entwicklung mit Eclipse
• Oracle WebCenter Portal und ADF Development
• Building Process Centric Anwendungen mit ADF und BPM
• Oracle Business Intelligence und ADF Integration
• Live Q&A Chat mit Oracle Mitarbeitern

The video explores how Oracle solutions are impacting the retail, automotive, pharmaceutical, medical and travel industries.Through the central character, Sam, this video highlights how cross-industry solutions like collaboration software, data processing and business intelligence platforms can improve business performance, reliability and flexibility; lower the cost and complexity of IT implementation and management; and deliver greater productivity, agility and better business intelligence.

To find out more about how Oracle’s products and services can help you to deliver better results, visit www.oracle.com/retail

Independent Software Vendors (ISVs) who develop applications for Solaris 11 can exploit a number of interesting services as long as they are OPN Members with a Gold (or above) status and a Solaris Knowledge specialization:

• Free access to a Solaris development cloud with preconfigured Solaris developer zones through the
• apply for the: Oracle Exastack Remote Labs to get free access to Solaris development environments for SPARC and x86.
• Free access to patches and support information through MOS for Oracle Solaris, Oracle Solaris Studio, Oracle Solaris Cluster including updates for development systems 
• apply for the Oracle Solaris Development Initiative.
• Free email developer support for all questions around Oracle Solaris, Oracle Solaris Studio, Oracle Solaris Cluster and Oracle technologies integrating with Solaris 11
• apply for the Solaris Adoption Technical Assistance.

Oracle Fusion Middleware PaaS and Oracle Java Cloud Service

  Roberto Monteiro, Senior Sales Consultant, Oracle
In this session, learn how Oracle Fusion Middleware platform as a service (PaaS) can supercharge productivity with instant access to a platform for developing and deploying business applications in the cloud, complete with integrated security and database access. See how these capabilities are used by Oracle Java Cloud Service.

 Dec 4 - 17:15 - Mezzanine: Room 7

Last week I was over in Melbourne and Sydney to present at an Oracle Technology Network System Administration Day. During a set of morning sessions, I presented on some of the new things that we've done in Oracle Solaris 11 and our focus for the future. One of the sessions I presented was giving a quick technical overview of what a typical application deployment scenario would look like using technologies like the Image Packaging System (IPS), Service Management Facility (SMF), Oracle Solaris Zones, and network virtualization. While it's just a simple example, it provides the basic building blocks for a more advanced configuration that a data center would typically deploy. Given these would likely be of general interest, I thought I'd upload the slides for others to view - enjoy!

You can download the original source PPTX without some of the formatting errors in the above.

Today I started some experiments with Oracle guru Steven Davelaar, who lives about 20 minutes away from my place in Amsterdam by underground. Very convenient. He showed me a bunch of things in JDeveloper, while I showed him a bunch of things in NetBeans IDE.

He managed to deploy an ADF application to GlassFish in JDeveloper. And, so far, I failed to do the same thing in NetBeans IDE. Quite a few (around 100) JARs are needed, aside from the question of correctly setting up or importing an ADF application, and we're still figuring out which and who and when and where. And how. And if. And why.

Nonetheless, I did manage to get Oracle DB set up in NetBeans IDE, after downloading it from here:

http://www.oracle.com/technetwork/products/express-edition/downloads/index.html

Here's what it looks like when registered in NetBeans IDE, i.e., notice that I have a cool sample database available:

Data from the above database I managed to display very easily via the various NetBeans code generators in a PrimeFaces application, exactly as has been done many times in demonstrations and tutorials everywhere, i.e., generate JPA entities, then create an EJB, then inject the EJB into a PrimeFaces data table:

The next step is to somehow do the same with ADF in NetBeans IDE.

I had some trouble with passwords for Oracle DB, the command line (with Steven's help) proved helpful:

Wish us luck as we continue our ADF-inspired journey. This blog entry by Shay is also relevant:

Deploying Oracle ADF Essentials Applications to Glassfish

1. Install OAM, Webtier (OHS) and WebGate as per the standard installation steps.
2. Create a WebGate instance (i.e deploy WebGate)

A WebGate instance must be created that will copy required bits of agent from WEBGATE_HOME to WebGate instance location that shares the same INSTANCE_HOME with OHS

3. Configure WebGate

In the webgate configuration the EditHttpdConf utility will copy OUI instantiated apache_webgate.template from WEBGATE_HOME to webgate instance location (renamed to webgate.conf), and update httpd.conf with one additional line to include webgate.conf.

4. Register WebGate

Use RREG tool to register the OAM 11G WebGate

After performing the above steps, there will be two artifcats created under Oracle/Middleware/Oracle_IDM1/oam/server/rreg/output, namely ObAccessClient.xml (Stroing webgate config parameters) and cwallet.sso (storing the agent key). These files must be copied to WebGate instance config folder (/Oracle/Middleware/Oracle_WT1/instances/instance1/config/ohs1/webgate/config)

Restart OHS

5. Deploy the web application (myApp) in WebLogic application server
6. Proxy Configuration in OHS

The mod_wl_ohs module enables requests to be proxied from Oracle HTTP Server 11g to Oracle WebLogic Server.

Restart OHS. Now if we access the web application URL with OHS host and port

(Ex: http://OHS_HOST:<OHS_PORT>/myApp) so that the requests will be proxied to WebLogic server.

7. Create a new application domain

8. Test the Web Application Integration.