Revision 0bb555630f5e85a1471843f8dc0dabec297c1c49 authored by Siying Dong on 08 April 2019, 20:24:29 UTC, committed by Facebook Github Bot on 08 April 2019, 20:32:06 UTC
Summary: Create new function NPHash64() and GetSliceNPHash64(), which are currently implemented using murmurhash. Replace the current direct call of murmurhash() to use the new functions if the hash results are not used in on-disk format. This will make it easier to try out or switch to alternative functions in the uses where data format compatibility doesn't need to be considered. This part shouldn't have any performance impact. Also, the sharded cache hash function is changed to the new format, because it falls into this categoery. It doesn't show visible performance impact in db_bench results. CPU showed by perf is increased from about 0.2% to 0.4% in an extreme benchmark setting (4KB blocks, no-compression, everything cached in block cache). We've known that the current hash function used, our own Hash() has serious hash quality problem. It can generate a lots of conflicts with similar input. In this use case, it means extra lock contention for reads from the same file. This slight CPU regression is worthy to me to counter the potential bad performance with hot keys. And hopefully this will get further improved in the future with a better hash function. cache_test's condition is relaxed a little bit to. The new hash is slightly more skewed in this use case, but I manually checked the data and see the hash results are still in a reasonable range. Pull Request resolved: https://github.com/facebook/rocksdb/pull/5155 Differential Revision: D14834821 Pulled By: siying fbshipit-source-id: ec9a2c0a2f8ae4b54d08b13a5c2e9cc97aa80cb5
1 parent de00f28
DUMP_FORMAT.md
## RocksDB dump format
The version 1 RocksDB dump format is fairly simple:
1) The dump starts with the magic 8 byte identifier "ROCKDUMP"
2) The magic is followed by an 8 byte big-endian version which is 0x00000001.
3) Next are arbitrarily sized chunks of bytes prepended by 4 byte little endian number indicating how large each chunk is.
4) The first chunk is special and is a json string indicating some things about the creation of this dump. It contains the following keys:
* database-path: The path of the database this dump was created from.
* hostname: The hostname of the machine where the dump was created.
* creation-time: Unix seconds since epoc when this dump was created.
5) Following the info dump the slices paired into are key/value pairs.
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...