Efficient cache for gigabytes of data written in Go

Re:Efficient cache for gigabytes of data written in Go by Eikon on Monday December 23, 2019 @10:58AM (#21864425)

There’s no network latency involved as it’s an in memory, embedded database.

Also, max throughput here is ultimately a sum of how fast the memory allocator is and memory bandwidth which, even with a poor implementation would be orders of magnitude more than what you can do with standard NICs and Linux kernel networking stack.

Moreover, as a consequence, far less context switching is involved.

[ Parent | on HN ]

Re:Efficient cache for gigabytes of data written in Go by shepardrtc on Monday December 23, 2019 @11:05AM (#21864526)

Ahh I missed the part about the network. That makes much more sense, then.

[ Parent | on HN ]

Re:Efficient cache for gigabytes of data written in Go by Exuma on Monday December 23, 2019 @11:12PM (#21869802)

Can you explain why redis require extra network time?

[ Parent | on HN ]

Re:Efficient cache for gigabytes of data written in Go by shepardrtc on Tuesday December 24, 2019 @09:51AM (#21872382)

Its just time to transfer across the network, whereas this project is in-memory with the program that's calling it.

[ Parent | on HN ]

Re:Efficient cache for gigabytes of data written in Go by stevekemp on Monday December 23, 2019 @12:43PM (#21865350)

They do discount networked solutions, but then talk about making HTTP-calls with JSON payloads. Or was that just for testing?

[ Parent | on HN ]

Re:Efficient cache for gigabytes of data written in Go by Znafon on Monday December 23, 2019 @11:05AM (#21864523)

I thought it was possible to embed Redis.

[ Parent | on HN ]

Re:Efficient cache for gigabytes of data written in Go by Eikon on Monday December 23, 2019 @11:11AM (#21864566)

Possible don’t mean ubiquitous. You get no support and plenty headaches when you start to do that sort of weird, unsupported things with a product.

[ Parent | on HN ]