QuestionQ25

VerifiedDesigning integration solutions to meet performance requirements

Refer to the exhibit.

Question Image

A manufacturing company is creating a new set of APIs for its retail business. One API, the Master Look Up API, is a System API that uses a persistent object store. Nearly all other APIs will use this API to obtain master lookup data.

The Master Look Up API is deployed on two CloudHub workers of 0.1 vCore each because a large amount of master data must be cached. Most master lookup data is stored as key-value pairs. The cache is refreshed when a key is not found in the cache.

Performance testing determined that the Master Look Up API has a high response time because of the latency of database queries used to retrieve the master lookup data.

What two methods can be used to address these performance issues?

Choose two
Explanation

Synchronizing access to the object store prevents concurrent cache misses on separate CloudHub workers from triggering duplicate database lookups for the same key. Applying HTTP caching to GET responses lets consumers reuse master-lookup results and reduces calls that would otherwise reach the API and its database. MuleSoft documents that multi-worker CloudHub applications require coordinated locking to avoid uncoordinated object-store access, and that Object Store v2 performs I/O for every read and write.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!