8 Comments

Very interesting, and summarized to the right level of detail for someone who just wants a 5-minute overview of the subject. Thanks.

Expand full comment
author

thank you for the feedback

Expand full comment

Not helpful, just seems like 2 lines copied from the original post and pasted here.

How is this helping anyone learn any of the concept. The page has been filled with image and one liner (with no detail whatsoever) to make the blog lengthly. would be better to simply create a section of source blogs separately.

Expand full comment
author
Jan 12·edited Jan 12Author

I see. What 2 lines from which original post are you referring to?

Have you checked the references? They are from tech talks and H3 documentation.

The reading time of this post is 5 minutes. If you find this lengthy, I don't know how to help further.

Expand full comment

All are copied from the reference as such and are one liners, without any context.

Map matching transforms raw GPS signals into actual road segments.

They use Apache Cassandra to store raw locations for long-term durability.

Apache Cassandra is a distributed database. Yet Cassandra is optimized for write operations.

So they add a Redis cache layer on top of it to shed the read operations.

Expand full comment
author

I see. I didn't want to repeat the information that I included in another article.

You could read this article for more context on map matching: https://newsletter.systemdesign.one/i/139242379/map-matching

Does that help?

Expand full comment
Feb 15·edited Feb 15

Hi, its written: "They use Thrift over Remote Procedure Calls (RPC) for efficient communication.", but thrift is rpc framework, so this sentance no sense.

Expand full comment
author

I see, thanks for the feedback. It looks like I need to improve my English skills.

It was supposed to mean - they use Thrift data format on RPC channels.

Expand full comment