Compiere from the Source

Compiere Scaleability & Performance

These are the architectural components governing Compiere's architecture Performance & Scalability:
 
Volume:

Compiere handles transaction through its persistence layer which maps, secures and authenticates transaction consistency from dictionary meta data to database
The Compiere persistency engine is optimized for high volume OLTP environments.  For example, Compiere uses anonymous keys with the default settings of NUMBER(10) in the database and Java int. This restricts the system to 999 million records per table. If this is not sufficient, you can without code changes switch to Java Big Decimal and increase the scale of the ID column in the database to the maximum supported by the database.  Along with performance scalability, Compiere provides rigorous security and roles based access options. For example, the system can restrict users from creating queries with a defined number of records.  Long running processes can be executed in background or stared via scheduler thereby optimizing CPU performance.

Database:

Compiere uses a central database and is in production with Oracle Database Grids and database clusters.  Compiere manages transactions via a central service, eliminating the need for application developers to manage their transaction.  This increases uniformity and stability.  Compiere is available on Oracle and EnterpriseDB (PostgreSQL) with prototypes on DB/2 and Microsoft SQL Server.

Application Server:

Compiere can utilize any number of individual application servers, application server clusters and also Oracle Application Server Grids.  Each individual application server can be specialized for certain tasks or run in parallel to other application servers (e.g. multiple accounting servers). This allows for the IT infrastructure to scale up with the needs of the business application.  In addition to JBoss, Compiere was implemented with the Oracle Application Server and IBM WAS - and the web container also runs on stand alone Tomcat.

Blog Re-Org

Our blog was popular target of hackers, so we switched to a better environment with also provides Tags and a better way to find/search categories.  We plan to publish on a weekly schedule.

What would help is to get more topic suggestions.  Please feel free to add your suggestions and comments. Just sign-in - the link is on the right side.

In addition to adding the blog to your feeds, you can sign up to get blog updates via email.

Financial Consolidation

Consolidation of the financial data of companies can be a very complex process, especially if companies span national boundaries (the term company and organization here are synonyms).

Compiere makes most of the consolidation tasks surprisingly easy. Let's use a simple example: You have a production company and a distribution company. Compiere has a counter-document feature which automatically creates the document in the other company or organization. So, if you create a purchase order in the distribution company it creates a sales order in the production company. Our counter document feature is heavily used in the distribution industry between headquarter and (semi-) independent stores or outlets. With minimum overhead you can  even completely automate the supply chain tasks within your organization.

This approach also uses explicit transfer prices - eliminating the need for inter-company profit elimination. You now can create individual or consolidated transaction and financial reports, just by selecting the specific organization or some grouping of organizations or companies.

Compiere maintains company or organization hierarchies for control and reporting purposes. As you can have multiple hierarchies in parallel, you can easily simulate the effects of company reorganizations on financial statements - before or after the event. If you change the organization you can change the hierarchy directly or if you build another hierarchy of the organization, you can then just switch your primary hierarchy, and could continue reporting using any other hierarchy.

Financial Data Warehouse

Compiere is by design a financial data warehouse.

Old applications often offer a separate financial data warehouse to allow to use accounting data for decision making. The general approach is to transfer the (compressed) sub-ledger data to the general ledger and the in an additional step "enrich" this summarized GL info and store this in a financial data warehouse.

Compiere maintains all data on the lowest level of detail. This allows maintaining parallel summarization hierarchies for reporting. Example: you can maintain the old organization structure and the new one in parallel and use both of them for reporting. For your reporting, you can select from predefined accounting elements in addition to two user defined list and two user defined references.

The ready to use dimensions include owing organization, transaction organization (e.g. cost center), business partner, product, sales region, activity (for activity based costing), location (from and to - enabling regional value flow analysis), project (including phase and task), marketing campaign.  You can create multiple summarization structures for any of the accounting elements (i.e. dimensions for data warehousing).

Compiere does not require the data entry person to select/enter accounting information nor accounts. Example: the sales region can be derived from the sales rep - the marketing campaign could be derived from a coupon/voucher. As data entry personnel is motivated to finish the transactions as fast as possible (not necessarily to provide detailed information for management reporting), all information is derived from the transaction and default accounts.

Multi-Tenant Architecture (MTA)

Many Application Service Provider (ASP) offer either the Classic Hosted Model or the new Multi-Tenant Architecture (MTA).

Examples of the Classic Model are companies like Oracle who claim that the advantage is to allow unique and more extensive customizations. Proponents of a Multi Tenant Architecture are companies like SalesForce reaping the benefits of the "one instance" low overhead. For both options, the ASP is responsible for operations - but when it comes to upgrades to new versions, the classic hosted model requires the same effort as on-premise implementations.

The biggest advantage of a MTA for a user is that there is little maintenance effort on the user side - as version upgrades are supposed to have no impact on the use of the application.

How does Compiere stack up? Compiere fully supports both options and even supports traditional on-premise implementations - with the ease of operation & maintenance advantages of a Multi-Tenant Architecture.

The MTA approach is limited to support "planned" customizations (i.e. if a certain customization/functionality is not supported, there is no way to provide this in that environment). The main benefit of the MTA is that the cost of system maintenance can be shared by all tenants/users - resulting in dramatically reduced costs.

Model Driven Architecture (Part 2)

Active Dictionary or Model Driven Architecture (MDA) systems may generate executable code behind the curtains, but there is no need or no option to modify the low level code. Compiere takes it a step further and executes the model rules directly.

Without any generation or coding you can

• add a new column to a table
• add optional validation rules (list of values, range, table (incl. external), etc.)
• add the new column to a window
• entering values
• create a report to list the new column or use the new column as a query parameter for reports

and all of that within 5 minutes!

More complex scenarios will not take more than an hour - and you can do this in a production system without causing any downtime. As Compiere is designed to be easy customized, the new column is marked as user defined and treated appropriately in upgrades - so changes in the core system does not affect customizations.

If an application claims that it is model driven or based on an active dictionary, ask them to demonstrate the example above. If it takes them longer, requite to take the system down or does not allow the new column to be immediately used in reports or report parameters, it is not a "real" active dictionary based system with the consequence of higher maintenance cost and lower productivity.

Model Driven Architecture (Part 1)

Compiere employs a Model Driven Architecture (MDA) to enable broader application adaptability, faster deployments and lower cost of ownership than is possible with products utilizing earlier generations of enterprise software architectures.

This is the first of two entries on MDA and Compiere. This first article covers the historical foundations of MDA as well as the resulting advantages. The second entry will explore the specifics of how Compiere's employs an MDA. Let me start out with a bit of history.

To manage the increasing complexity of applications, Data Dictionary systems arose in the early 1970's. The early systems abstracted an application's data structure. To facilitate working with large numbers of entities and increasingly complex relationships, these tools evolved into Entity Relationship Diagramming tools.

Next, the need evolved to manage and model also business processes, so the Data Dictionary applications evolved into Application Dictionary systems. Computer Aided Software Engineering (CASE) tools emerged to enable developer productivity with the Application Dictionary. CASE introduced code generation capabilities.

Some CASE tools claimed support for "round trip" engineering. CASE systems with round trip engineering added the important innovation of ensuring that system code modifications drove updates to design documentation, ensuring that the design and system are in sync. Unfortunately, due to the inherent complexity and round-trip restrictions, bigger systems were not able to take advantage of this capability. The business consequence here is that these systems become more and more harder and costly to maintain over time.

Welcome

Welcome to my personal blog. I plan to publish my thoughts about Compiere, Open Source and other items here. Please feel free to add your comments and topic suggestions - just register! Technorati Profile