There is a project called vega that's a Spark implementation in Rust: https://github.com/rajasekarv/vega. It looks promising.
Databricks rewrote the Spark engine in C++ (called the Delta Engine, see here: https://databricks.com/blog/2020/06/24/introducing-delta-eng...).
Hopefully there will be some good alternatives that don't consume so much memory soon ;)
There is also https://github.com/ballista-compute/ballista (shameless plug for my own project)
Hopefully they have all measures in place to prevent data to be tainted by C and C++ "features".
It is the bytecode compiled language like any other that uses LLVM as their backend.
The alternative to Spark will be Spark, as JVM slowly improves their memory APIs from release to release.
People tend to hand wave how much it costs to reboot an ecosystem.
Rust, to me, seems like a natural enough choice. It is easy to mate it to other languages, including all the major data science ones, so it would theoretically work well as the basis for a distributed compute engine that has good support for all of them as client languages. Would the same work for Julia? IIRC, it's a bytecode compiled language, which I imagine would make it difficult to link Julia libraries from other technology stacks.