Weblog for the AMIS Technology corner
Posts tagged Database
Jouw toekomst als Oracle professional – AMIS !!EXTRA!! informatieavond over jouw ontwikkelingspad
Jan 24th
Als je dit leest en je bent Oracle professional – ontwikkelaar of administrator – is de kans groot dat je jouw carriere bent gestart in de jaren ‘90, in het tijdperk van Oracle7, Oracle Forms en Client/Server applicaties – net voor de opkomst van Java, internet, mobiel en SOA. Misschien ligt je start daar nog wel voor en misschien ben je van een tikje later. Maar in elk geval: je hebt een solide fundament en al vele jaren ervaring in Oracle technologie.
Nu is het 2012. En de afgelopen jaren is de technologie in hoog tempo ontwikkeld. Web 2.0 applicaties en multi-tier architecturen dienden zich aan en SOA, Service Bus en BPM zijn opgekomen. Java is overal – en HTML 5 staat op het punt om Flash en Silverlight te verdringen, in desktop browsers en in mobiele devices. Eindgebruikers worden mondiger en veel-eisender en real-time, push-enabled, BI-gedreven dashboards en multi-channel takenlijsten worden in snel itererende agile-projecten gerealiseerd.
Oracle 11g Database doet alle simpele DBA taken zelf – en de uitdaging voor de Database Administrator is verschoven naar groter en interessanter, zeker als 24/7 beschikbaarheid, virtualisatie en de cloud als infrastructuur opduiken. Middleware administratie en integrale security zijn zomaar twee nieuwe uitdagingen op de weg van de administrator, naast een verdergaande consolidatie en professionalisering van deployment en monitoring.
De uitdaging voor de Oracle professional lijkt duidelijk: hoe kies je je weg naar de toekomst? Om interessante klussen te blijven (gaan?) doen is het nodig om aansluiting te vinden bij de ontwikkelingen in de markt. Eerst moet je natuurlijk goed inzicht hebben in wat die ontwikkelingen zijn – in de markt in het algemeen en binnen de Oracle technology stack in het bijzonder. Dan moet je uitvinden hoe je vanuit je huidige kennis en ervaring de aansluiting vindt en jezelf prepareert voor het werken met de huidige en komende technologie. En tenslotte moet je het gaan doen – kennis vergaren, ervaring opdoen en je profileren om de passende klussen op te kunnen pakken.
Publishing to CometD Bayeux Channel from inside the Oracle Database – PL/SQL based push to CometD Web Client
Jan 19th
In recent articles, I have introduced CometD as framework for Server to Client and Client to Client Push: Running CometD 2 examples – locally on Tomcat using Maven and NetBeans http://technology.amis.nl/blog/14709/running-cometd-2-examples-locally-on-tomcat-using-maven-and-netbeans, CometD 2 Java Client Sample – open project in NetBeans based on Maven pom file, modify sources and run Java Based Comet Client (http://technology.amis.nl/blog/14720/cometd-2-java-client-sample-open-project-in-netbeans-based-on-maven-pom-file-modify-sources-and-run-java-based-comet-client) and Push based synchronized Slideshow demo application implemented using CometD and jQuery running on Tomcat (http://technology.amis.nl/blog/14870/push-based-synchronized-slideshow-demo-application-implemented-using-cometd-and-jquery-running-on-tomcat).
CometD allows web clients (or stand alone Java clients) to subscribe to Bayeux channels maintained by the CometD Server (Servlet). These clients can publish messages to the channels and CometD will deliver the messages to all or to selected clients – in a push-fashion. This makes it possible to push messages from the Java server side of a web application to web clients and indirectly to do the same between web clients. One example I have described of applying this technology is the synchronized slideshow: a web page is opened in two or more browsers. When a slide is selected in one of the browsers, the slide is selected in all browsers (because they all subscribed to the channel that a slide selection event was published to).
This article adds another push channel to the overall picture: it describes how the Oracle Database can publish directly to the Bayeux Channels (by posting to the CometD servlet over HTTP) and thereby perform push to the subscribed web clients. Database to browser push – how does that sound?
2 dagen seminar door Steven Feuerstein: Best of Oracle PL/SQL (8 en 9 december)
Oct 13th
Steven Feuerstein
In dit tweedaagse seminar neemt Steven Feuerstein je mee ver voorbij de basismogelijkheden van PL/SQL. Steven zal tijdens dit seminar de best practices behandelen die hij op tientallen plekken in de wereld heeft verzameld en die hij ook mede door zijn nauwe samenwerking met het PL/SQL product team van Oracle kan verifiëren en aanscherpen. Hij laat via code-voorbeelden en interactieve sessies zien hoe je als ontwikkelaar in staat bent om snel, hoogwaardige, goed onderhoudbare en uitbreidbare applicaties kunt maken.
Er is tijdens dit seminar voldoende ruimte om vragen te stellen en ideeën met Steven Feurestein uit te wisselen.
Bij dit seminar zijn de volgende zaken inbegrepen:
- Een exemplaar van het boek Oracle PL/SQL Programming (5th edition) van Steven Feuerstein (twv €71,-).
- Een PL/SQL quiz onder leiding van Steven Feuerstein met leuke prijzen.
- Ook ontvang je een gratis jaarabonnement op het PL/SQL Channel. Hiermee heb je toegang tot 27 uur videotraining (normaal $395).
Dit is de kans om in Nederland in een select gezelschap met één van de experts op PL/SQL gebied kennis te maken en je expertise in Oracle PL/SQL verder te vergroten. Meer informatie of inschrijven kan via deze link.
(NB: Deelname voor deze twee dagen kost 1050 euro)
(The slides from) Don’t Call us, We’ll Push you – on Cross Tier Push Architecture (JavaOne 2011)
Oct 5th
I just completed my Birds of a Feather session at JavaOne 2011 about Push in an enterprise application architecture. Over 150 attendees were in the audience, many of them asking questions afterwards. It felt like it went well.
The slides for this session are below. I will post source code for the demos at some later stage.
OOW 2011 – What’s New, Improved and Coming in Oracle Application Development
Oct 3rd
Currently at Tom Kyte’s session regarding topics new, improved or coming in Oracle Application Development. Tom told about the history APEX has gone thru and the current setup with the APEX Listener and even the “PL/SQL Gateway” was mentioned. I always have to laugh a bit because this last one touches the XDB Protocol Server which can do way more then only this PL/SQL extension for APEX which has been embedded in this XDB Protocol Server framework. There is a APEX book out there that touches a bit the possibilities of the framework, one of the reasons I will promote and explain it a bit more during conferences like UKOUG, because I think its a shame that people don’t know its full potential. Anyway. Tom stressed once again that APEX is a serious environment regarding the huge websites out there based on APEX supporting 1000th of user sessions. Also it nowadays had a better debug support (4.1) and use / support for ROWID, improved data upload and calendar wizard support and redesigned websheets in APEX 4.1.
Data Integriteit anno 2011 – Hands-on met Toon Koppelaars en RuleGen 3.0 bij AMIS op dinsdag 27 september
Sep 22nd
datum: dinsdag 27 september; 16.30 uur
locatie: AMIS, Edisonbaan 15, Nieuwegein
sprekers: Toon Koppelaars en Lucas Jellema
Het belang van data integriteit behoeft geen toelichting – als het goed is. Vrijwel iedere enterprise applicatie is gebouwd op het fundament van een database. Robuustheid van die database en absolute betrouwbaarheid van de gegevens daarin zijn. essentieel. Het garanderen van die integriteit en robuustheid is lang niet altijd goed geregeld – en is ook geen triviale uitdaging.
De veelheid van toegangskanalen tot de database – batch jobs, web services, web applicaties in Java en .NET, PL/SQL modules en natuurlijk TOAD voor de applicatie beheerder – maakt wel duidelijk dat de integriteitsregels voor de data ofwel de business rules niet in de applicaties of in een service bus (alleen) kunnen worden geïmplementeerd. Integriteit in moderne applicatie-landschappen kan alleen worden gegarandeerd met implementatie van de business rules in de database zelf.
De Oracle database biedt naast de integrity constraints voor primary, unique en foreign keys weliswaar een veelheid aan mechanismen en constructies maar geen concrete, eenduidige, robuuste en efficiënte manier om business rules te implementeren. RuleGen is een framework dat optimaal gebruik maakt van wat Oracle 10g en 11g aan mechanismes biedt om te komen tot consistente en waterdichte afdwinging van de business rules. De uitvoering van de rules is efficiënt. Op run time kunnen rules gemonitord worden, in ‘log only mode’ draaien en zowel op statement- als transactie-level worden afgedwongen. De analyse en ontwikkeling van business rules wordt door RuleGen met een to-the-point user interface ondersteund die ontwikkelaars laat concentreren op de essentie ipv op het loodgieterswerk.
“RuleGen zou je kunnen beschouwen als de natuurlijke opvolger van CDM RuleFrame dat eind vorige eeuw door Oracle Nederland werd gelanceerd – een verbeterde uitvoering die de database slimmer benut” – aldus Lucas Jellema, CTO van AMIS en ooit de grondlegger van CDM RuleFrame.
In deze workshop wordt RuleGen geïntroduceerd. Read the rest of this entry »
RuleGen 3.0 – the latest, leanest and most robust solution for complex data constraints in an Oracle Database
Jul 6th
No matter how complex the enterprise and application architectures become, no matter the number of tiers, services, devices and user interfaces – at the heart of most enterprises will be a relational database.
And no matter how hard we try to implement a fully service based architecture or a multi-purpose business tier (for example using EJBs) – we will have multiple routes to the database and the data in it. Data will be manipulated through web applications, web services, client/server applications, batch database jobs, application managers working directly against the database from the command line or TOAD-like tools. If for no other reason – that by itself is an overridding motivation for enforcing every data constraint at the lowest possible level – the one level that none of these channels can avoid: the database itself. In addition to the fact that only enforcement inside the database can provide real integrity (and what the hack would be semi-real integrity?), for most data rules (especially complex ones) implementation inside the database is the most performant, scalable, maintainable, managable, agile, elegant and easy to implement as well.
Note: I am not advocating that data rules should not be implemented in other tiers as well. For the purpose of user friendliness through quick feedback and preventing incorrect data from ever getting near the database, it is of course perfectly fine to specify required fields, fields with a certainly length or a domain with allowable values in user interfaces and/or describe and enforce data constraints in the POJO Domain Model and the XSDs for the canonical data model of the ESB. However: all of these are only supplemental. They do not provide robustness.
Implementing data rules inside the database is not trivial, despite all the clear advantages outlined above.
Choosing the best way for SOA Suite and Oracle Service Bus to interact with the Oracle Database
Jun 21st
In the past few weeks, I have encountered a similar discussion in various organisations. Each organisation uses either SOA Suite (11g) or Oracle Service Bus (11g) at the core of their SOA infrastructure – either for integration purposes, for workflow and process orchestration or for both. In each organisation, the role of one or more databases is crucial and interaction between the SOA environment and the database is one of the most common and therefore important functions in their enterprise IT environment.
Given the importance, it is only logical that the way(s) chosen for linking the SOA component to the database is carefully selected – and evaluated every now and again. Technology may have progressed, experience may have taught us a lesson or two, the knowledge and skills may have evolved.
Interestingly enough, the number of ways for SOA Suite and OSB to communicate with the database is quite high. This article includes an illustration that shows over 20 different interaction channels that we can choose from with a fairly wild variation of attributes, required skills, productivity and performance characteristics.
Among the choices we face is the question of the communication protocol to use for accessing the database. Roughly speaking, we can go through JDBC (via a JDBC Data Source on the WebLogic Server), via HTTP or using ‘another’ channel such as File System, FTP, Email, XMPP (Chat) etc. The communication protocol may in part determine another choice that awaits us: what will be the format and nature of the messages that we exchange: text based (XML, JSON, CSV) or strongly (Oracle-ly) typed. We know that on the one hand we have the world of XML in the SOA middleware components, and on the other side the relation world of the Oracle database – at least relational and strongly typed at the core.
So when choosing the method for having the SOA components interact with the database, we have to decide on the communication protocol and we have to determine where in the route from table to SOA component (and vv.) the conversion from Relation to XML or from XML to Relational will take place.
Considerations when making that decision include the question who has to do the work of making the XML <=> Relational conversion at design time and at run time – what is the productivity of an approach and what the performance impact. Other considerations include robustness and agility or ability to maintain and evolve the implementation. Of course the required skills factor into this decision as well as the complexity of the architecture, the design, development, test and deployment process and the nature of the interaction: read only, data manipulation, small or large documents, synchronous or asynchronous (or even fire and forget) etc.
How to use an outerjoin to synchronise data
Apr 11th
There is often a need to synchronize data from a query. For example reading external data and update the tables. Sometimes it can be achieved by a single merge statement, but this is not always possible. Mind, for example, the need to update more than one table, or updating the data using an api.
As an example we want to grant and revoke privileges on database objects. Tables needing select, insert, update and delete privileges, packages execute and views select privileges. The table and packages also get the grant option for the select and execute privilege.
Read the rest of this entry »
My First Steps with RAC One Node 11gR2 (11.2.0.2)
Apr 5th
I’ve been using RAC (10gR2) for years now and I was wondering how RAC One Node differs from RAC. That’s why I used a few ‘old’ RAC servers that we had ‘laying’ around to install a RAC One Node Cluster. I also used that cluster to get some hands on experience with ASM but that is not the topic for this post, so I will not go into that further.
I assume that installing the software is a breeze for everyone who has some experience with RAC and who has read the manual so I will not go into that as well. Once you have fulfilled the prerequisites it’s a matter of next, next, finish. Read the rest of this entry »
