Do I need an ORM ?

Written on May 2, 2015

Estimated reading time : 2 mins

Tags : | database | orm |

Problem Statement

• Many popular database products such as structured query language database management systems (SQL DBMS) can only store and manipulate scalar values such as integers and strings organized within tables.
• The programmer must either :
  1. Convert the object values into groups of simpler values for storage in the database (and convert them back upon retrieval) (ORM !!!)
     • Only use simple scalar values within the program :
     • Via Procedural Code
     • Via OO Code with DAO Design Pattern - Use an Object Oriented Database (subset of NoSQL)

About

• Object-relational mapping is a programming technique for converting data between incompatible type systems in object-oriented programming languages.
• This creates, in effect, a “virtual object database” that can be used from within the programming language.

Advantages / Pros

• Reduction in code

    /* Eg in Normal Code */
    String sql = "SELECT * FROM persons WHERE id = 10";
    DbCommand cmd = new DbCommand(connection, sql);
    Result res = cmd.Execute();
    String name = res[0]["FIRST_NAME"];
  

    /* Eg in ORM Code */
    Person p = Person.Get(10); 
    String name = p.getFirstName();
  

• Easy development in most cases

Dis-advantages / Cons

• Exponential Complexity
  • ORM tends to be useful for 70-80% of requirements which are generally simple but the rest requires you to hack around some limitations of the ORM or use raw queries instead (not cacheable !)
• Performance
  • It is due to the fact that they use wildcard * for selecting data.
  • Compiling queries from ORM code is slow (eg : Doctrine ORM compiler is written in PHP)
  • Slower than traditional queries though it can be offset to a large extent by using opcode caching.
• High level of abstraction obscuring what is actually happening in the implementation code.
• Heavy reliance on ORM software has been cited as a major factor in producing poorly designed databases

Recommendation

• A good ORM can definitely reduce development time & encourage use of OOP concepts.
• Usually 70% of the requirements tend to be simple & can easily be implemented via ORM within a short time (relative to a non ORM solution).
• For the rest, a good ORM provides for means to execute complex queries directly.
• For a good performance, you will have to use an opcode cache layer + a data cache layer (memcache, varnish etc) so as to offset the speed issue. If you don’t want to do this, then ORM solution is not for you.

References

• http://martinfowler.com/bliki/OrmHate.html
• http://blog.codinghorror.com/object-relational-mapping-is-the-vietnam-of-computer-science/
• http://stackoverflow.com/questions/108699/good-php-orm-library
• http://stackoverflow.com/questions/7765070/redbean-orm-performance
• https://philsturgeon.uk/blog/2011/06/misconceptions-about-orms/