Introduction Background
Example MapReduce job Worker registration
Job submission
Fault tolerance
Copyright By PowCoder代写 加微信 powcoder
This site uses Just the Docs, a documentation theme for Jekyll.
Search CS 162 HW 5
Conclusion
TABLE OF CONTENTS
1 Freeing intermediate results
2 MapReduce applications
Congratulations on finishing the assignment! Here are some optional, not-for-credit things you can do to improve your MapReduce system.
These are only for fun; there is no extra credit provided.
Freeing intermediate results
At the moment, workers buffer map task outputs indefinitely. This means that they will eventually run out of memory. Implement logic to free intermediate map buffers once a job has completed (or fails).
You can add additional RPCs to handle this, or modify your existing RPCs.
MapReduce applications
Try implementing your own MapReduce application! You can get ideas for what to implement from the MapReduce paper.
Look at src/app/grep.rs and src/app/wc.rs to see how to fit a MapReduce application into the framework we provide. You¡¯ll also need to modify src/app/mod.rs to allow workers to find your new application.
Back to top
Copyright ý 2022 CS 162 staff.
Conclusion
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com