OLTP systems are heavyly writefocused on fast inserts and updates of records.
OLAP systems are read focused.

Transactional workloads, those ones that require fast inserts and updates of records.
The reason why we treat these use cases differently is that transactional systems are heavily right focused.

Example

• he trade-off is that queries tend to be relatively simple, and tend to affect only a few records.

For example, take a banking system. Deposit your salary from your company to your account, that's a transaction. It updates your balance field. The bank is using an online transaction processing or OLTP.

In analytics workload on the other hand tends to read the entire dataset, and it is often used for planning our decision support.
Analytical systems on the other hand can be periodically populated from the entirety of the operational systems.

Example

For example, a bank regulator might require us to provide a report of every customer who transferred more than $10,000 to an overseas account. They might ask the bank to include costumers who tried to transfer the 10,000 in smaller chunks over a period of a week. A report like this will require significant scanning of that large dataset, and require a complex query that includes aggregating over moving time Windows in SQL. This is an example of an online analytical processing or OLAP workload.

	transactional	analytical
source of data:	operational data; OLTPs are the original source of the data	consolidation data; OLAP data comes from the various OLTP database
purpose of data:	control ans run fundamental business tasks	help with planning, problem solving and decision support
what the data shows:	reveals snapshot of ongoing business processes	multidimensional views of various kinds of business activities
inserts and updates:	short and fast inserts and updates initiated by end users	periodic long-running batch jobs refresh the data
queries:	relatively standardized and simple queries returning relativly few records	often complex queries invlovling aggregations.
processing speed:	typically very fast	depends on amount of data involved; improve query speed with indexes.
space requirements:	can be ralatively small if historical data is archived	larger, more indexes than OLTP