<aside> 💡 Total Cost of Ownership (TCO).

This cost includes both the hardware procurement and software license, as well as the labor overhead of deploying the DBMS and the energy needed to run the machines.

</aside>

<aside> 💡 Parallel vs. Distributed

I. Definition

Database is spread out across multiple resources to improve different aspects of the DBMS.

Appears as a single logical database instance to the application, regardless of physical organization.

II. Difference

• Parallel DBMSs
  • Resources are physically close to each other.
  • Resources communicate over high-speed interconnect.
  • Communication is assumed to be cheap and reliable.
• Distributed DBMSs
  • Resources can be far from each other.
  • Resources communicate using slow(er) interconnect.
  • Communication cost and problems cannot be ignored. </aside>

1. Process Models

<aside> 💡 Process Model → defines how the system is architected to support concurrent requests from a multi-user application.

Worker → the DBMS component that is responsible for executing tasks on behalf of the client and returning the results.

</aside>

1.1 Process per DBMS Worker

Each worker is a separate OS process.

• Relies on OS scheduler.
• Use shared-memory for global data structures.
• A process crash doesn’t take down entire system.

1.2 Process Pool

A worker uses any free process from the pool.

• Still relies on OS scheduler and shared memory.
• Bad for CPU cache locality.

1.3 Thread per DBMS Worker

Single process with multiple worker threads.