A critical vulnerability was discovered in React Server Components (Next.js). Our systems remain protected but we advise to update packages to newest version. Learn More

Quan Mai
Jan 20, 2015
  2875
(0 votes)

Don’t let CatalogEntryFull slows you down.

Just like any EPiServer products, one of main concern of us when developing Commerce is performance. Why many parts of Commerce can be described as “Good enough”, some parts are just acceptable, and I must admit that there’s area where performance need to be improved to meet customer’s expectation.

We are under an intensive performance improvement process which we try to optimize our code to cut out any slow code we can found. And one thing come to my mind when I investigate the impact of CatalogEntryResponseGroup. Let’s see the dotTrace profile for CatalogEntryFull value:  CatalogEntryFull

You can see how insane CatalogEntryFull can do. It’s more than 4000 call to LoadEntryWorker just to load 9 Entry.

The terrible secret behind CatalogEntryFull is when you use it, Catalog system will try to load everything of the CatalogEntryDto, along with associations and children, recursively. The more complex and related your catalog entry is, the slow CatalogEntryFull becomes.

While we recommend to use new content API:s whenever possible, there are still place where Entry/Entries being used, such as when you call CartHelper.AddToCart method. Be sure to use the minimal ResponseGroup which works for you, instead of throw the full-fledge, big and slow CatalogEntryFull.

 

TL, DR; you should avoid CatalogEntryFull if you don’t have too. Use it only you absolutely need it. Use most lightweight CatalogEntryResponseGroup enum that do the work for you. You might need to use several enum combined, such as Assets | Inventory, instead of just going for the bad, ugly and slow CatalogEntryFull.

Jan 20, 2015

Comments

Arve Systad
Arve Systad Jan 20, 2015 11:36 PM

I like what I read!

Every millisecond counts, and especially in solutions with thousands of products, all performance gains are extremely welcome.

Are there plans to get rid of the CatalogEntryDto-style APIs at some point, and let the "de facto product management" go through IContentLoader and other more modern interfaces?

Quan Mai
Quan Mai Jan 21, 2015 03:40 AM

I believe that it's our vision, but it won't happen anytime soon. We would support two API:s at the same time for smoother transition.



By the way, this blog post applies for EPiServer Commerce all version, from 1 R1 to latest nuget package.

Please login to comment.
Latest blogs
Troubleshooting with Azure Application Insights Using KQL

Users at least get access to Azure Application Insights even within minimum access level if you are requesting access to DXP management portals at...

K Khan | Dec 21, 2025

Looking back at Optimizely in 2025

Explore Optimizely's architectural shift in 2025, which removed coordination cost through a unified execution loop. Learn how agentic Opal AI and...

Andy Blyth | Dec 17, 2025 |

Cleaning Up Content Graph Webhooks in PaaS CMS: Scheduled Job

The Problem Bit of a niche issue, but we are building a headless solution where the presentation layer is hosted on Netlify, when in a regular...

Minesh Shah (Netcel) | Dec 17, 2025

A day in the life of an Optimizely OMVP - OptiGraphExtensions v2.0: Enhanced Search Control with Language Support and Synonym Slots

Supercharge your Optimizely Graph search experience with powerful new features for multilingual sites and fine-grained search tuning. As search...

Graham Carr | Dec 16, 2025