Sync updates from HyperPebblesDB to PebblesDB

[xxks-kkk]

This pull request contains the changes from HyperPebblesDB 1.0 that can be sync into PebblesDB

Features include:

• Continuous code integration using Travis-CI
• New emptyGuards option in db_bench that gives detailed SSTables distribution under each guard across all levels
• db_bench enables "fillseq" and "fillrandom" even when using existing database
• db_bench support "ycsb" option by reading a trace file generated from the YCSB benchmark (check "scripts/trace_gen.sh" for details)
• Support guard-based parallel compaction, which can be turn on/off with options.parallel_guard_compaction
• Support Succinct Range Filter (SuRF) as an alternative to bloom filter. Check SuRF paper for details
• Better internal documentation
• Verified portability on both Mac & Linux using g++-5

HyperPebblesDB 1.0

[vijay03]

Hi Zeyuan,

Thanks for these contributions! This must have been a lot of work.

However, I want to ensure that these changes don't break PebblesDB accidentally. Could you refactor these into smaller pull requests where @pandian4github, @jayashreemohan29 or I could look at the code to make sure it is right? I would also encourage you to mention any testing you have performed in the pull request. Finally, with each pull request, could you also modify the README file? Add a new heading "Improvements made after the SOSP paper" and list your contributions :)

[xxks-kkk]

Hello Vijay,

Thanks for the comment. I'll refactor this pull request and incorporate your comments.