Performance improvements #64
Conversation
…ameter to prevent build error
|
I already tried to test it I deployed previous version and this one. I've found previous version is way faster I think could be related to hosted service, because it doesn't make any sense. Both started almost at the same time
This version:
Those are the numbers I calculated, Can't find a way to determine how faster is this version on indexing. |
There was a problem hiding this comment.
- have you found the reason why the indexing time was actually much worse with the improvements? is there a way to compare the two instances in a more deterministic way? Or basically any test depends on the hosted service hardware they randomly assign to you?
- have you found any difference on the data?
I couldn't find any deterministic way of report the performance improvement.
No I haven't. There are a few id's changed, but this shouldn't represent any conflict on your side. |
Based on this article from Edge & Node people I worked on this for improving the performance of our subgraph.
This includes immutable entities and using Bytes as the id of entities. The two features are independent of each other; immutable entities only require a small change to the subgraph schema, while using byte strings as the id also requires small changes to the mappings.
Both of these enhancements will improve indexing speed, reduce the amount of data that needs to be stored for a subgraph, and also speed up some queries.
Due to the changes I already made I deployed this here and tested the affected entities to check if the information is still right. The information seems to be ok.
I need to perform an indexing speed test comparing it with the previous version. To perform this test I will remove a prop in both subgraphs (not crucial, just for making it index again) and I will query how many blocks are indexed in 10 minutes on different parts of the indexing progress for being able to decide if the change makes sense. Besides that, there should be improvements on some queries.
Close #54