
Based on the wishes and needs of our customers, we have developed numerous modules over many years. This flexibility is only possible because our framework is based on a multi-layer architecture right from the start. This allows us to play one or more modules on your platform at any time.
You decide what you need.
And when.
Not just that. Our library is structured in such a way that each module individually and highly individualize them, which selectively overwrites the library at the appropriate places.
We always have our finger on the pulse, constantly refining existing modules and continuously and highly agilely expanding our range of solutions.
Each of the modules listed here is used productively in Live environments.
Modular. Individual. Scalable. Flexible.
And there are more and more...
Because digital transformation is a constantly progressing and changing process.
2 Step Verification
Industriespionage ist ein sehr lukratives Geschäft. Insbesondere, wenn es darum geht, Kundendaten und deren Umsätze auszulesen, Online-Umsätze einzusehen, Kaufverhalten, Klickverhalten uvm.

Geben Sie Hackern keine Chance!
BACKEND
Login
- Nur möglich über eine 2-Step-Verification
B2X Multi Store
Specialised trade organisations (FHOs), such as association groups or franchise companies, are often faced with the challenge of providing their affiliated dealers, craft businesses and branches with an excellent online marketing, sales and service platform at an appropriate price/performance ratio.

Since specialised trade organisations represent the link between end consumers, dealers and manufacturers, the Internet offers substantial advantages in this context, not least in terms of standardising brand management. With its eDemand platform, ICONPARC FHOs offers a cost-effective way to provide retailers with various applications, such as a web shop, via the Internet, with which they can simplify their daily administration and open up new sales channels. A participating retailer therefore needs neither hardware nor software to set up his web shop, but subscribes to it, tailored to his needs, from his FHO. The ICONPARC eDemand platform FHOs offers an attractive solution that leaves nothing to be desired in terms of scalability and flexibility and at the same time requires very little resources and infrastructure, especially with regard to the critical factors of competitive differentiation and cost reduction as well as the development of future growth potential.
Features
-
Optimum availability of the entire system
-
High level of security at significantly reduced costs thanks to central infrastructure and administration
-
Central infrastructure requires minimal effort for the individual merchant, who does not need to operate his own web, database and firewall servers, nor interfaces to payment gateways, logistics service providers, etc.
-
Convenient shop configuration and administration by participating merchants via fully browser-based access
-
The connection to the central merchandise management system of the FHO is reduced to a single interface (in comparison to a scenario with decentrally installed dealer shops).
-
High-quality editorial content and product data can be centrally defined and made available to dealers for selective use.
-
Centrally selected elements of the eDemand platform (e.g. catalogue navigation, hit lists, product details and shopping cart) are seamlessly integrated into all retailer presences.
-
Use of centrally defined editorial content, product data and processes increases the quality of dealer presences and thus promotes acceptance and customer loyalty.
-
Quicker roll-out: As exemplary content is already available, not every retailer has to worry about all the details of his appearance.
-
Cluster option for maximum scalability and resilience
(VirtualServer)
BACKEND Mobile Edition

Greifen Sie vollständig und auf alle Daten von unterwegs auf Ihr BACKEND mit Ihrem Smartphone zu. Die BACKEND mobile Edition von ICONPARC ist nahtlos und vollständig mobile optimiert. Eine 2-Step-Verification sorgt zusätzlich dabei für höchste Sicherheit. Darüber hinaus können Sie sich direkt über einen Fingerabdruck oder per Gesichtserkennung noch schneller und bequemer einloggen.
Beautiful Information

For each monthly statement, you will receive an attached detailed service list with references to the respective tickets as well as an insert - Beautiful Information, which analyzes data collected and summarizes it in a visually appealing PDF.
Catalog & PIM
In order to sell successfully on the Internet, attractiveness, clarity and correctness of the product data are just as important as with printed catalogues.
In contrast to the print sector, which sometimes has long catalogue runtimes, Internet customers expect product information and price quotations to always be up to date online. The provision of catalogue content as part of a website is usually associated with numerous additional challenges. Catalogue content is rarely available in a complete and structured form and cannot therefore be used directly for the Internet without additional preparation. In practice, a versatile and flexible tool is needed to provide high-quality product data for online catalogues: The Catalog & PIM module ensures that catalog content can be easily composed from various data sources. For example, basic article master data, performance characteristics and descriptions as well as price lists and illustrations can be imported step by step via different interfaces. The completely browser-based BACKEND offers all possibilities to subsequently edit, enrich or refine imported data. The image data is automatically optimized for web conformity, whether during import or direct maintenance. Product master data, descriptive features, price information, product group structures and much more can be conveniently and efficiently entered and edited. Use the easy-to-use import and administration options to ensure that your catalogue content is always up to date.

Performance characteristics
- Management of any number of catalogues and price lists including graduated prices
- Flexible country and region model for price list application
- Hierarchical merchandise category structure with any number of levels
- Support of any further classification systems
- Administration of any number of manufacturers
- Additional elements configurable for each catalog, such as
- Illustrations, descriptive texts, document attachments, PDFs and much more.
- Support of product variants such as colours, sizes, etc.
- Management of any feature groups and bars incl. product comparison view
- Individual access rights for persons and groups
BACKEND
Role
- Catalogue manager
Report
- Ratalogue orders
Searches
- Product groups
- Products
- Manufacturer
Dialogue Assistants
- Catalogues
- Product groups
- Classifications
- Products
- Characteristic bars (also for filters)
- Manufacturer
- Catalog exports
- Media Center for graphics, photos, videos, documents
The most successful way to get visitors excited about your website again and again is to continuously revise and update the content of the platform.
While at that time Google mainly read out keywords, the search engine now checks content for meaningfulness based on context and - rightly - classifies it in the ranking on this basis.
This is why good content for winning new customers is now more sustainable than expensive SEM measures.

Care via the BACKEND

With the ICONPARC CMS module there are theoretically no limits. In practice, however, it is indispensable to design structures and the Look&Feel or operating systems in a consistent way: This challenge is even greater the more editors work on the content simultaneously, nationally or internationally distributed. It is important that individual editors do not express their own design ideas. Rather, design and all design aspects in general should be implemented as they are structurally and visually prescribed by your company's brand management, taking into account the corporate identity. Otherwise you run the risk that your customers will find a colourful patchwork carpet in your external presentation - here in the e-business front-end - which will by no means do justice to your brand.
The ICONPARC CMS system automatically and conceptually ensures the uniform visualisation of all FRONTEND contents. Permissible degrees of freedom can be defined in fine granularity.
In the following we will introduce some elements of the CMS module.
Snippets
All editorial content is composed of individual content snippets. Simple snippets can represent a headline, an image, a video or a text block. More complex snippets allow for example a "text-image-link" block. In addition to the numerous snippet types included in the standard scope of delivery, any number of other snippet types can be defined within the scope of customization.
Templates
The combination of desired snippet types is configured to create templates for the actual editorial content, which can be easily managed in the backend of the solution with the CMS template assistant. For example, it can be specified that certain snippet types must always be present in content derived from a certain template. The order of the applicable snippet types can also be specified, as well as the permissibility of multiple occurrences of individual snippet types within the template.
Contributions
Editorial contributions can be entered and updated in the CMS Content Dialogue Assistant: These page components can be placed - personalized if required - at any number of locations within a web presence. If a contribution is modified, this automatically affects all places where this contribution appears. As part of the personalization process, visibility can be limited, for example, by time or by assigning it to certain visitor or customer groups.
Documents
The CMS Content Dialogue Assistant can also be used to create complete editorial pages: These documents can be used in particular as link targets, for example, for elements of a navigation menu as well as for landing pages or newsletters. Editorial contributions can be placed on suitable page areas.
Example: Definition and design of a simple article template
The following example should give an impression of how a coherent template definition - which can be combined with other elements - is created from many degrees of freedom.
First, the order of the assigned (and thus allowed) snippet types has been set. Each selected snippet can be a mandatory field and may not be left empty.
Headings must be written in capitals, in the color petrol and with bold formatted 32 point font in the house font, left-aligned.
For pictures, only the size ratios 2:1 or 1:3 are permitted. To ensure that the resolution is sufficient for all end devices, images must have a width of at least 2,000 pixels. After saving, images are automatically calculated in all required sizes and optimized for display on desktop PCs, tablets and smartphones.
Buttons always have a caption in the house font in capitals and with a 20 point size. The frame is 1 pixel thick in black. With MouseOver the button is coloured in dark blue with white lettering.
The clou: No editor can get past the template definition, i.e. it is not possible to change or overwrite the design. Your brand manager will take this with a sigh of relief, because in this way the promising combination of diverse content with uniform presentation is guaranteed.
We know from several decades of practical experience: It is perfectly normal for your company's CI to change over the years. Whenever this happens, it is not necessary to revise every single document or article: Rather, it is often sufficient to adapt the design definitions created for the FRONTEND. Afterwards, all content appears in the new Look&Feel! This is an invaluable advantage, not least for platforms with several 10,000 content pages, because: A completely new design can go live with just one click.
Seamless integration
We follow the principle of seamless integration for the modules from the ICONPARC E-Business Suite. But what exactly does this mean and what are the advantages for you? We have already presented how editorial content is basically managed in the form of articles and documents. Through seamless integration with other modules, you can also add content from the CMS module to pages from other modules. This includes, in particular, hit lists, product group pages and product detail views, as well as all process pages, for example the full shopping cart display and the pages of the checkout or registration process. The more you enrich your pages with successful editorial content, the more acceptance you gain from visitors and customers - and Google rewards your efforts with a better ranking.
At the same time, editorial content can be personalized to display appropriate content to different visitors. Personalization results in pages with slightly to highly individualized content. Conventional caching methods on page level fail at this point. In order to deliver pages with personalised content with best performance, ICONPARC has developed complex, multi-level caching systems on the page component level. The invalidation of partial contents is carried out on the same level, so that changes are also only read from the database once and are then available in an optimised cached form.
or via the FRONTEND Cockpit

With the ICONPARC CMS module there are theoretically no limits. In practice, however, it is indispensable to design structures and the Look&Feel or operating systems in a consistent way: This challenge is even greater the more editors work on the content simultaneously, nationally or internationally distributed. It is important that individual editors do not express their own design ideas. Rather, design and all design aspects in general should be implemented as they are structurally and visually prescribed by your company's brand management, taking into account the corporate identity. Otherwise you run the risk that your customers will find a colourful patchwork carpet in your external presentation - here in the e-business front-end - which will by no means do justice to your brand.
The ICONPARC CMS system automatically and conceptually ensures the uniform visualisation of all FRONTEND contents. Permissible degrees of freedom can be defined in fine granularity.
In the following we will introduce some elements of the CMS module.
Snippets
All editorial content is composed of individual content snippets. Simple snippets can represent a headline, an image, a video or a text block. More complex snippets allow for example a "text-image-link" block. In addition to the numerous snippet types included in the standard scope of delivery, any number of other snippet types can be defined within the scope of customization.
Templates
The combination of desired snippet types is configured to create templates for the actual editorial content, which can be easily managed in the backend of the solution with the CMS template assistant. For example, it can be specified that certain snippet types must always be present in content derived from a certain template. The order of the applicable snippet types can also be specified, as well as the permissibility of multiple occurrences of individual snippet types within the template.
Contributions
Editorial contributions can be entered and updated in the CMS Content Assistant: These page components can be placed - personalized if required - at any number of locations within a web presence. If a contribution is modified, this automatically affects all places where this contribution appears. As part of the personalization process, visibility can be limited, for example, by time or by assigning it to certain visitor or customer groups.
Documents
The CMS Content Assistant can also be used to create complete editorial pages: These documents can be used in particular as link targets, for example, for elements of a navigation menu as well as for landing pages or newsletters. Editorial contributions can be placed on suitable page areas.
Example: Definition and design of a simple article template
The following example should give an impression of how a coherent template definition - which can be combined with other elements - is created from many degrees of freedom.
First, the order of the assigned (and thus allowed) snippet types has been set. Each selected snippet can be a mandatory field and may not be left empty.
Headings must be written in capitals, in the color petrol and with bold formatted 32 point font in the house font, left-aligned.
For pictures, only the size ratios 2:1 or 1:3 are permitted. To ensure that the resolution is sufficient for all end devices, images must have a width of at least 2,000 pixels. After saving, images are automatically calculated in all required sizes and optimized for display on desktop PCs, tablets and smartphones.
Buttons always have a caption in the house font in capitals and with a 20 point size. The frame is 1 pixel thick in black. With MouseOver the button is coloured in dark blue with white lettering.
The clou: No editor can get past the template definition, i.e. it is not possible to change or overwrite the design. Your brand manager will take this with a sigh of relief, because in this way the promising combination of diverse content with uniform presentation is guaranteed.
We know from several decades of practical experience: It is perfectly normal for your company's CI to change over the years. Whenever this happens, it is not necessary to revise every single document or article: Rather, it is often sufficient to adapt the design definitions created for the FRONTEND. Afterwards, all content appears in the new Look&Feel! This is an invaluable advantage, not least for platforms with several 10,000 content pages, because: A completely new design can go live with just one click.
Always Connected
We follow the principle of seamless integration for the modules from the ICONPARC E-Business Suite. But what exactly does this mean and what are the advantages for you? We have already presented how editorial content is basically managed in the form of articles and documents. Through seamless integration with other modules, you can also add content from the CMS module to pages from other modules. This includes, in particular, hit lists, product group pages and product detail views, as well as all process pages, for example the full shopping cart display and the pages of the checkout or registration process. The more you enrich your pages with successful editorial content, the more acceptance you gain from visitors and customers - and Google rewards your efforts with a better ranking.
At the same time, editorial content can be personalized to display appropriate content to different visitors. Personalization results in pages with slightly to highly individualized content. Conventional caching methods on page level fail at this point. In order to deliver pages with personalised content with best performance, ICONPARC has developed complex, multi-level caching systems on the page component level. The invalidation of partial contents is carried out on the same level, so that changes are also only read from the database once and are then available in an optimised cached form.

Our CMS is seamlessly integrated into all modules.

Our CMS Caching Strategies

But what exactly does that mean and what are the benefits for you?
You want to place editorial content such as documents and articles anywhere you want, for example on product group, product and process pages such as in the check-out process, etc., in order to enrich visitors and customers with more content. If you do not do this in this way, you will only have the catalog and product data for your texts. In large companies, however, this procedure is difficult to implement or is not planned.
As a result, it is not possible to cache editorial pages completely as independent pages or to deliver them via Content Delivery Network (CDN), because the system then displays different content, such as editorial content, product data, and not to forget customer-specific prices.
In theory, this means that every content must be retrieved dynamically from the database during page generation. This would burden the database too much with reading processes and significantly slow down the speed of your platform with every click.
This is why we have developed and implemented extensive, multi-level and highly complex cache strategies.
Here is an example
Assuming there are 300 accessible pages and as personalization five web presences (= multi-store), two languages, two SessionStates (= visitor, known customer), four Areas (= where the content is placed on the page), 16 product areas (= e.g. product groups).
Then there are 300 x 5 x 2 x 2 x 4 x 16 = 384,000 theoretically possible cache objects for all combinations of personalization.
If all result sets are empty (= 2KB), this results in a memory capacity of 750 MB. If half of the result set is filled (= 20KB), the result is 3.75 GB of memory.
Our methodology reduces the number of cache objects to a fraction and avoids many cache invalidations. After each invalidation, the data has to be requested from the database again at a high price.
In the example above, we need a maximum of 600 cache objects. If all result sets are empty (= 2KB), this results in a memory size of 1.2 MB. If all result sets are filled (20+30 = 50KB), this results in a memory capacity of 30 MB.
In practice, our methodology results in a speed advantage in the order of 1,000!
This makes us extremely fast in the delivery of content. Your visitors are happy about this and Google rewards you with a better ranking. A further advantage is: As soon as you activate a change, it is immediately visible to everyone - without delay and internationally.
Technical specifications
SEO features
- Configuration and validation per document from
- Talking URL
- Title
- Header
- META description
- META keywords
Performance characteristics
-
Fully browser-based content editor
-
Fast, fluid editing of editorial content. Add, edit, sort, delete items using the keyboard or a context menu using the right mouse button.
-
No reload of the page, only of change data
-
Clear, practical editor interface (convenient operation, quick familiarization)
-
Personal folder for the organization of self-created content
-
Preview of the FRONTEND before activation for customers
-
Code separation of content and design
-
Support for multiple designs despite identical content
-
Ensure link consistency. If a document is deleted in BACKEND, the system informs you that a navigation entry, for example, must first be deleted because it points to the document. This means that there are no data corpses (404 pages) that Google would rate negatively.
-
Document templates including template editor
-
Post templates including template editor
-
Layout templates including template editor
-
Linking content and products
-
Integration of content and business processes
-
Time-controlled contribution placement
-
Additional dynamics through automatically rotating content and sliders
-
Automatic generation of sitemaps and menu systems
-
Fully dynamic page creation
-
Multidomain capabl
-
Easy integration of external databases
-
Automatic archiving of editorial content
-
Support for any file format, such as images, text files, PDFs, videos, sound files and much more.
-
Picture gallery function
-
Multiple uploads of multiple images
-
Sites are always created in real time and therefore do not need to be generated.
-
Caching ensures high performance
Workflow
-
Definable process steps for the creation of editorial content
-
Role-based release process
-
Several editors can work independently on one website
-
Direct sending of documents and contributions (ad hoc workflow)
-
Status management of editorial content
FRONTEND
Variations of the page contents depending on ...
- from the visitor
- ... the state of the visitor's session, unknown visitor (guest), Identified user - recognized via a cookie, authenticated user - with password verified and logged in
- ... of the customer group
- ... the virtual page. If, for example, you operate several Web sites in different CIs, but with structurally identical content, these are displayed in different design configurations (separation of logic and content).
BACKEND
Role
- Chief editor
- Editor
Reports
- Activity
- Page views
- Events
Searches
- Editorial contents
- Navigation elements
Dialogue Assistants
- Documents
- Contributions
- Navigation
- SEO
- Events
E-Commerce
In order to use the Internet as a profitable sales channel, it is necessary to address interested parties and customers in a targeted manner, to make the selection of products as easy as possible for them, to present catalogue contents in a clear (and up-to-date!) manner, and finally to make the shopping process pleasant.

With the help of the versatile and individually supplementable comfort functionalities of the ICONPARC E-Commerce Module, you can turn occasional visitors into regular customers. Buyers are automatically kept up to date by e-mail on progress in order processing. Articles already purchased can be conveniently reordered from the order history of the respective customer. Via newsletter you draw the attention of customers and interested parties to special offers, services and promotions. These and other performance features contribute to the fact that your B2B e-business platform meets with a high response.
So that you always have a full overview of your customers and their orders, extensive possibilities for the evaluation of all processes are integrated into the BACKEND.
FRONTEND
-
Check-out processes freely configurable in different steps
-
Changes in the shopping cart, such as changing order quantities, are made without reloading the page. The newly calculated subtotals, totals and shipping costs are also displayed directly without reloading.
BACKEND
- Role
- E-Commerce manager
- E-Commerce manager
- Search
- Clients
- Customer employees
- Orders
- Returns
- Turnover
- Reports
- Commissions
- Distribution of orders
- Bestseller
- Returns
- Sessions
- Turnover
- Dialogue Assistants
- Clients
- Customer employees
- Mail notifications
- Orders
- Vouchers
- Views
- Customer master overview
- Order overview
- Time controlled processes
- Monitoring of the order transfer to the ERP system
- Mails to customers
Middleware
If you introduce a standard product for your digitization, you are purchasing an isolated solution. This stands alone, is not integrated into your mostly complex system landscape and does not fulfil any process controls.

You have all the master data in your ERP: Customers, customer employees, invoice and delivery addresses, payment terms, delivery options, etc. These must somehow flow into the new system. And not only that: If customers change their data online, they have to be automatically mirrored back into the ERP system.
Catalogues, product groups and article data: This data is also already available in ERP and PIM. They must also be automatically integrated into the platform. After all, this is the only way for products to enjoy the fastest market launch - via the digital sales channel Internet.
Customer-specific prices, discounts, delivery times and availability. Without this real-time information, a B2B platform has no added value for your customers.
What about orders, delivery notes, invoices?
Would you like to transfer these manually via mail to your ERP system?

Calculation example for customer-specific prices, conditions and discounts.
This data is stored in your ERP system for each customer master. The B2B e-business platform could determine the data in real time via web services. In theory. In practice, several hundred, if not a thousand customers are online at the same time. So if only 200 customers call up a hit list with 40 products at the same time, that is already 8,000 prices that the ERP must determine and deliver immediately. As a rule, an ERP system cannot do this, and if it does, it does so with too long a delay.
BACKEND
Role
- Administrator
Report
- Interfaces
Search
- Interfaces
Dialogue Assistant
- Interfaces-Configuratios
Time-controlled processes
- Reconciliation of customer bases
- Progress in processing orders
Event-driven processes
- Customized prices
- Article availabilities
- Quantities in the warehouse and with suppliers
- Order data transfer
Technologies (examples)
- Transference
- Webservices
- HTTPS
- (S)FTP
Search & Filter
The largest product range is of no use if customers cannot find the articles they need. The Search & Filter module, on the other hand, makes it easy for your customers: with a full text search of all product data, hit sorting according to highest relevance and sales figures, feature filtering, grouping...

The Search & Filter module makes searches findable
-
Search via a quick search field across all product data or expert search with individual search fields for name, description, manufacturer, price and other features
-
Autocompletion / Search as you type: During the search input, articles, categories or manufacturers are already suggested.
-
Post-filtering / after search navigation on the basis of any article characteristics or article categories and product groups
-
Synonym support and typo correction so that your customers will find items even if they have mistyped or use words that do not appear in the item description.
-
So that your customers do not lose track of many hits, the search results are automatically sorted so that the most relevant ones for the customers are displayed first. The system considers various factors such as sales figures, click rates, product groups, keywords, etc.
-
High-performance: Even with huge product ranges, the search function delivers suitable results in less than 200ms. For example, one of our customers has over 40 million searchable articles in his assortment! This does not interrupt the performance, the search results appear lightning fast, i.e. significantly less than half a second.
The technical basis for the module is a so-called search index. This is a kind of table of contents for your product data, which makes it possible to determine products with certain properties and search words within milliseconds. To generate this, we use the standard software Apache Lucene™, which we have extended with various functions for the ICONPARC platform.
Through this architecture we relieve your database (which can therefore process other queries faster) and deliver highly relevant search results without noticeable delay. Speed is the trump card in two ways:
Customers love fast page access, and that's why Google rewards you with a much better ranking.
FRONTEND
-
Lightning-fast display of search results according to design specifications for hit lists and Auto Suggest
-
Post-filtering / after search navigation on the basis of any article characteristics or article categories and product groups
BACKEND
Role
- Search optimizer
Report
- Search analyses
Search
- Editorial contents
- Navigation elements
Dialogue Assistants
Index configurations. Configuration and definition of columns which are included in the search index, e.g. title, i.e. the name, description, all characteristics (such as examples 1-2-3 wavy, size, color dimensions, etc.).
- Splitting exactly these columns into individual TOKENs using an analyzer. Example: This is the title. Each word is broken down individually, that is, this, is, the, title within a configured set of rules, how special characters such as -/&%$§"!.... etc. are handled.
- Search rules, e.g. exact search & wildcards ...
- Ranking rules: Implementation of the rules, such as searching in n columns, such as title, description, product properties ... under the influence of the configured weights per column with AND and/or OR, as well as revaluations, such as devaluing accessories or cross-selling, revaluing up-selling, or revaluing or devaluing depending on prices, or ranking by viewed items, or by sold items, or if searching for an explicit brand, all other products that do not belong to this brand are hidden, because the brand owner wants it so, all other products that do not belong to this brand are because the brand owner wants it so, are hidden.
- Synonyms
- Stopwords
- Similarity search
Time-controlled processes
- Index creations
System Update Info

All customers receive an overview of all installed updates including graphically displayed time histories via the BACKEND. In addition, numerous metrics for the respective Live System are visualized. In addition, each customer can configure automated mail notifications for himself. Updates and enhancements can be viewed at any time and from anywhere via our BACKEND.
Metrics for the FRONTEND
- Number of customers, customer employees, inquiries, offers, orders, purchase orders, watch lists, design configurations, roles, searches, report jobs, views, dialogue assistants, notification e-mails, SEO URLs
Metrics for BACKEND
- Number of users, roles, searches, reports, views, tools, dialogue assistants, scheduled processes, notification emails

Ticket System

We involve customers closely in the development process of their solution. To ensure that nothing is lost from the constant exchange between you and us, we use a ticket system that is structured like a discussion forum or chat program: Each aspect of the current project phase receives one or more threads that reflect the entire process of coordination between all parties involved on your and our side. The ICONPARC ticket system has its own FRONTEND, but can also be operated entirely via e-mail.
Virtual Data Modeling Layer
Datenmodelle haben eine wesentlich längere Lebensdauer als Software, welche Funktionen und Prozesse umfasst. Daher haben wir von Anfang an unseren Fokus auf die Datenqualität gelegt. Nach über 15 Mannjahren Entwicklung haben wir ein Tool entwickelt, welches den Bauplan einer Datenbank beschreibt, um sich damit stabil, konsistent, optimal und extrem schnell fehlerfrei zu replizieren.

Um zu verstehen, was das genau in der Praxis heißt, beginnen wir mit einer kleinen Einführung in die Welt der Datenbanken...
Datenbankmodell
Grundlage für die Strukturierung der Daten und ihrer Beziehungen zueinander ist das Datenbankmodell, das durch den DBMS-Hersteller festgelegt wird. Je nach Datenbankmodell muss das Datenbankschema an bestimmte Strukturierungsmöglichkeiten angepasst werden:
- hierarchisch: Die Datenobjekte können ausschließlich in einer Eltern-Kind-Beziehung zueinander stehen.
- netzwerkartig: Die Datenobjekte werden miteinander in Netzen verbunden.
- relational: Die Daten werden zeilenweise in Tabellen verwaltet. Es kann beliebige Beziehungen zwischen Daten geben. Sie werden durch Werte bestimmter Tabellenspalten festgelegt.
- objektorientiert: Die Beziehungen zwischen Datenobjekten werden vom Datenbanksystem selbst verwaltet. Objekte können Eigenschaften und Daten von anderen Objekten erben.
- dokumentenorientiert: Die zu speichernden Objekte werden als Dokumente mit möglicherweise verschiedenen Attributen, d.h. ohne die Voraussetzung der Strukturgleichheit, gespeichert
Vorteile von relationalen Datenbanken
- einfaches Datenmodell
- Datenkonsistenz (bei normalisierten Modellen), da Änderungen zentral und nur an einer Stelle erfolgen
- Verknüpfung über Inhalte
- Durch gleichberechtigte Relationen sind Einstiegspunkte beliebig
- Durch die Zerlegung aller Objekt in Einzelteile werden die Daten mengenorientiert verarbeitet. Damit können komplizierte und komplexe Abfragen auf große Datenmengen und mächtige Operationen erfolgen
- Hohes Maß an Datenunabhängigkeit: Die Navigation obliegt völlig dem Datenbanksystem und ist damit einfach für den Nutzer
- Standardisierte einheitliche Datenbanksprache für mehrere Anbieter
Darum arbeiten wir ausschließlich mit relationalen Datenbanken wie MSSQL, PostgreSQL sowie ORACLE.
Jede Anwendungssoftware benötigt drei Systeme
- Entwicklungssystem
Entwicklungs-Plattform in der DMZ und nur intern erreichbar - Testsystem
Plattformen außerhalb der DMZ und damit nur für Tester erreichbar. Nach Freigabe der Tester (in unserem Fall der Kunde) wird es auf das - Produktivsystem
migriert und ist damit für alle sichtbar und produktiv
Die Herausforderung bei mehreren Systemen: Die Migration der Modelle und Daten
Für neue Prozesse und Funktionen muss das Datenmodell verändert bzw. erweitert werden. Dies betrifft z.B. Tabellen, Fremdschlüssel, Indizes sowie deren Dateninhalte. Diese müssen im laufenden Betrieb auf das Testsystem und später auf das Produktivsystem migriert werden, möglichst effizient, also ohne Downtime der Systeme.

Und genau da kommen die Nachteile von relationalen Datenbanken zum Tragen
- Entstehung semantischer Lücken: Entity-Typen werden in Tabellen abgelegt. Diese semantische Beziehungen zwischen Entitäten, wie sie im Entity-Relationship-Modell definiert werden, können in einem Relationenmodell nicht explizit ausgedrückt werden
- Is-a- und Part-of-Beziehungen lassen sich ebenfalls nicht ausdrücken, da es im klassischen Relationenmodell nicht möglich ist, den Tupeln wiederum Sub-Tupel zuzuordnen.
- Keine Unterscheidung von Identitäts- und Kompatibilitätsspalten
- Deklaration von nur wenigen Datentypen möglich
- Die bei der Normalisierung erforderliche Zerlegung der Daten in Einzelteile erschwert inhaltlich und unter Performance-Gesichtspunkten die Suche nach komplexeren Zusammenhängen: Daten, die getrennt voneinander gespeichert werden, müssen auf der Anwendungsebene in Form von Joins wieder zusammengeführt werden.
- Es lässt sich wenig Wissen über die Regeln und Funktionen speichern, die mit den Daten verbunden sind.
Fazit
Datenmodell-Migrationen sind aufwendig.

Die Lösung um nur Vorteile zu gewinnen: Unsere selbstbeschreibende Datenbanken mittels Metamodellen
Um all diese Nachteile in der Praxis auszuräumen haben wir ein Metamodell entwickelt, welches sich selbst beschreibt und mit virtuellen Datentypen angereichert ist. Hier einige Beispiele:
Virtueller Datentyp bei ICONPARC
- eindeutiger Schlüssel (unique_key & als Identitätspalte gekennzeichnet)
- Boolsches Flag (boolean_flag)
- Objektname (name)
- Primärschlüssel (objid)
- Beschreibung (description)
- Ersteller des Objects (creator)
- Bearbeiter des Objects (updater) als
... entspricht dem realen Datentyp bei MSSQL
- varchar(50), not null + unique index
- integer, not null
- varchar(50), not null
- integer, not null + primary index
- text, nullable
- integer, not null + foreign key to users
- integer, not null + foreign key to users
Damit erfolgt die Migration extrem schnell, zielsicher und fehlerfrei. Nicht nur das, sondern wir können auch die Inhalte problemlos systemübergreifend überführen, da wir Identitäts- sowie Kompatibilitätsdaten kennen und diese heranziehen können. Und wir können größere Migrationen auch automatisiert zu einem bestimmten System ausführen lassen, zum Beispiel in der Nacht, damit unter Tags die Performance nicht beeinträchtigt wird.

Um Fehler zu vermeiden, werden alle Datenbanken auf allen Systemen stündlich gecrawlt und so das Metamodell mit den real existierenden Datenbanken gegengecheckt. Was Sie davon haben? Maximale Stabilität, Agilität, Flexibilität, Effizienz für extrem schnelles Wachstum.
WebAuthn
Secure passwords are long and complex. Unfortunately, this also makes them difficult to remember and, especially on smartphones and tablets, very awkward to enter.

With WebAuthn you can log in completely without a password. This can be a smartphone, a current Windows 10 device or a small USB/NFC stick. On the smartphone, this works just like unlocking, for example via fingerprint, facial recognition or a swipe gesture. And on desktop computers, for example, with a USB token. If you get a USB hardware token like a Yubikey, you can use this one key both on the desktop (via USB) and on the smartphone (via NFC) to log in.
This is much faster than typing in a password. And it's even more secure.