Good software stands on good data structures. We design and tune databases, analyse algorithms for runtime and memory, and develop methods for optimisation and analysis problems — from data mining to genetic algorithms and compression.
The typical data flow — from input to result. Every station is replaceable and adapted to your systems.
01Raw data
02Cleansing
03Model / algorithm
04Evaluation
05Decision
02
What we deliver
Databases, data mining and algorithms that solve hard problems.
01Database design, migration and performance tuning (MSSQL/T-SQL, MySQL, MongoDB, SQLite)
02ORM layers (PDO, Room, GreenDAO) and clean data models
03Data mining: patterns, segments and forecasts from your data
04Genetic algorithms for planning and optimisation problems
05Compression methods for storage and transfer
06Algorithm analysis, data structures and code reviews
03
A look at the code.
This is what an excerpt from a typical project looks like — readable, commented and without magic. That is exactly how we hand code over to your team.
T-SQLMSSQLMySQLMongoDBSQLitePDO
abfrage.sqlExample
-- Vorher: 38 s (Table Scan über 12 Mio. Zeilen)-- Nachher: 0,2 s (gezielter Index + umgeschriebene Abfrage)CREATEINDEX ix_auftrag_kunde_datum
ON dbo.Auftrag (KundeId, Datum DESC) INCLUDE (Betrag, Status);
SELECTTOP (50) a.Nr, a.Datum, a.Betrag
FROM dbo.Auftrag a
WHERE a.KundeId = @kunde AND a.Status <> 'storniert'ORDERBY a.Datum DESC;
04
What you get out of it.
01
Measure before optimising
We find the expensive parts instead of guessing.
02
Algorithms over hardware
Often a better idea is cheaper than a bigger server.
An optimisation method modelled on evolution: many candidate solutions are scored, the best are combined and slightly mutated. Over generations this yields very good solutions for problems that cannot be solved exactly in reasonable time — such as route or shift planning.
Can you speed up our existing database?
In most cases significantly. We measure first, find the expensive queries and optimise precisely — without rewriting your application.
SQL or NoSQL?
Relational databases are first choice for clear relationships and transactions; MongoDB suits flexible, document-like data. Often a combination makes sense.
