Direct naar content

Who is it?

We were recently approached by an intermediary asking us to provide database consultants with knowledge of DB2, Oracle, MariaDB, PostgreSQL, MongoDB, MS SQL, CockroachDB and MySQL for “a client. The client’s name remained shrouded in mystery. For us then, it is always a sport to find out from his databases who actually is the customer.

Gerard Zuidweg

Managing Partner
Gerard Zuidweg - Managing Partner

Clue 1: The age

Using the aforementioned database technologies and a little knowledge of the world around us, we often come a long way. You are either a database specialist, or you are not. In this case, for example, we see multiple database management systems (DBMS): SQL/RDBM, NoSQL AND New SQL DBMS applied in the technology stack.

De oudste en nog steeds meest gebruikte DBMS is het Relational Database Management Systeem, gebaseerd op de in de jaren 70 ontwikkelde Structured Query Language (SQL). De meeste bekende systemen in deze categorie zijn bijvoorbeeld OracleMicrosoft SQLMySQL en PostgreSQL. In een RDBMS worden gegevens opgeslagen in een gestructureerde tabelindeling van rijen en kolommen met als doel te voldoen aan ACID-eigenschappen voor datamanagement (Atomicity, Consistentie, Isolatie, Duurzaamheid). RDBMS / SQL konden ACID-eigenschappen bereiken omdat de gegevens in één databaseserver bleven, zodat consistentie geen probleem is.

The oldest and still most widely used DBMS is the Relational Database Management System, based on the Structured Query Language (SQL) developed in the 1970s. Most well-known systems in this category include  OracleMicrosoft SQLMySQL and  PostgreSQL.. In an RDBMS, data is stored in a structured table arrangement of rows and columns with the goal of satisfying ACID properties for data management (Atomicity, Consistency, Isolation, Durability). RDBMS / SQL could achieve ACID properties because the data remained in one database server, so consistency is not an issue.

For our question “Who is it?” it means that it is probably a large and somewhat older pre-1970 organization that started automating with IBM Mainframes and DB2 and later included Oracle in the stack. So we would start looking in the corner of, say, banking or insurance or government.

Clue 2: NoSQL

With the global explosion of the Internet in the 1990s, the amount of data generated grew exponentially. RDBMS were actually not suitable for these huge amounts of complex data, nor could they meet the new demands that came with global availability and scalability.

This led in the late 1990s to the emergence of NoSQL databases that offered a solution for scaling in a distributed environment. The well-known names in these include MongoDB, Cassandra and Redis.

Based on this, the organization we are looking for is apparently an internationally operating organization that makes large amounts of data available, managed and/or processed via the Internet. We then infer this from the fact that the organization uses MySQL and MariaDB, which is often used as a technology among Internet applications. Furthermore, MongoDB is used for large data warehouses, so it is probably an organization where a lot of data and Online Analysis Processes (OLAP) are involved.

Clue 3: NewSQL

However, the distributed NoSQL variants have two major drawbacks. First, they do not use SQL. This creates a lot of extra work to get the NoSQL database to interact with applications that generally can easily handle SQL. Second, because of their design, they have difficulty ensuring data consistency.

To address these limitations, a new generation of database management systems has emerged: NewSQL. NewSQL claims to combine the advantages of RDBMS (strong consistency) with advantages of NoSQL (scalability). In other words, NewSQL databases are relational database systems that match or exceed high performance and scalability of NoSQL while maintaining the ACID guarantees of traditional DBMS.

For example, Google has relied heavily on this technology with the development of its NewSQL variant Spanner. Some other well-known examples are ClustrixDB, NuoDB, CockroachDB (spin-off of Spanner), Altibase, MemDB, VoltDB, c-treeACE, Percona Apache, TIBCO ActiveSpaces and ActorDB.

So we also dare to conclude that the organization we are looking for is in the process of digital transformation. After all, a NewSQL DBMS helps organizations process Internet transactions and make data available globally, consistently and error-free.

Dus: Who is that?

In general, all variants of SQL, NoSQL and NewSQL can occur in an organization. One does not exclude the other. After all, as we always state, the choice of a DBMS is determined by the use case. We will gladly return to that in a later blog.

In answer to our question “Who is it?”, we arrive at the profile of a long-standing organization in the banking or insurance industry or an international government organization based in the Netherlands that operates globally, manages large amounts of data, makes it available and increasingly wants to enable its customers to interact independently on its applications. That leaves a lot of organizations, but we dared to take the gamble. It turned out to be a long shot.

Want to know more?

Want to know which organization it was or want to learn more about which DBMS is the best fit for your organization? If so, feel free to contact us.