Raphael S. Carvalho b4cf429aab sstables/LCS: Fix increased write amplification due to incorrect SSTable demotion
LCS demotes a SSTable from a given level when it thinks that level is inactive.
Inactive level means N rounds (compaction attempt) without any activity in it,
in other words, no SSTable has been promoted to it.
The problem happens because the metadata that tracks inactiveness of each level
can be incorrectly updated when there's an ongoing compaction. LCS has parallel
compaction disabled. So if a table finds itself running a long operation like
cleanup that blocks minor compaction, LCS could incorrectly think that many
levels need demotion, and by the time cleanup finishes, some demotions would
incorrectly take place.
This problem is fixed by only updating the counter that tracks inactiveness
when compaction completes, so it's not incorrectly updated when there's an
ongoing compaction for the table.

Fixes #4919.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20190917235708.8131-1-raphaelsc@scylladb.com>
2019-09-22 10:46:38 +03:00
2018-11-21 00:01:44 +02:00
2018-12-03 11:18:02 +02:00
2019-08-05 14:55:53 +02:00
2019-09-16 19:39:14 +03:00
2018-11-27 13:01:02 +02:00
2019-09-03 07:43:35 -07:00
2019-08-05 14:55:53 +02:00
2018-06-19 16:26:51 +03:00
2019-08-05 14:55:53 +02:00
2018-11-21 00:01:44 +02:00
2018-11-21 00:01:44 +02:00
2018-11-21 00:01:44 +02:00
2018-11-21 00:01:44 +02:00
2019-02-20 08:03:46 -08:00
2018-11-21 00:01:44 +02:00
2019-07-29 17:15:51 +02:00
2018-11-01 13:16:17 +00:00
2019-03-28 14:21:10 +02:00
2019-05-07 09:55:53 +03:00
2019-09-11 18:01:58 +03:00
2019-08-14 10:02:00 -07:00
2018-11-28 23:54:03 +01:00
2019-09-07 10:07:33 -07:00
2018-11-21 00:01:44 +02:00
2018-11-21 00:01:44 +02:00
2018-11-21 00:01:44 +02:00
2018-11-21 00:01:44 +02:00
2019-06-17 13:09:54 +01:00
2019-08-05 14:55:53 +02:00

Scylla

Quick-start

To get the build going quickly, Scylla offers a frozen toolchain which would build and run Scylla using a pre-configured Docker image. Using the frozen toolchain will also isolate all of the installed dependencies in a Docker container. Assuming you have met the toolchain prerequisites, which is running Docker in user mode, building and running is as easy as:

$ ./tools/toolchain/dbuild ./configure.py
$ ./tools/toolchain/dbuild ninja build/release/scylla
$ ./tools/toolchain/dbuild ./build/release/scylla --developer-mode 1

Please see HACKING.md for detailed information on building and developing Scylla.

Note: GCC >= 8.1.1 is required to compile Scylla.

Running Scylla

  • Run Scylla
./build/release/scylla

  • run Scylla with one CPU and ./tmp as data directory
./build/release/scylla --datadir tmp --commitlog-directory tmp --smp 1
  • For more run options:
./build/release/scylla --help

Scylla APIs and compatibility

By default, Scylla is compatible with Apache Cassandra and its APIs - CQL and Thrift. There is also experimental support for the API of Amazon DynamoDB, but being experimental it needs to be explicitly enabled to be used. For more information on how to enable the experimental DynamoDB compatibility in Scylla, and the current limitations of this feature, see Alternator and Getting started with Alternator.

Building Fedora RPM

As a pre-requisite, you need to install Mock on your machine:

# Install mock:
sudo yum install mock

# Add user to the "mock" group:
usermod -a -G mock $USER && newgrp mock

Then, to build an RPM, run:

./dist/redhat/build_rpm.sh

The built RPM is stored in /var/lib/mock/<configuration>/result directory. For example, on Fedora 21 mock reports the following:

INFO: Done(scylla-server-0.00-1.fc21.src.rpm) Config(default) 20 minutes 7 seconds
INFO: Results and/or logs in: /var/lib/mock/fedora-21-x86_64/result

Building Fedora-based Docker image

Build a Docker image with:

cd dist/docker
docker build -t <image-name> .

Run the image with:

docker run -p $(hostname -i):9042:9042 -i -t <image name>

Contributing to Scylla

Hacking howto Guidelines for contributing

Description
No description provided
Readme 523 MiB
Languages
C++ 72.1%
Python 26.7%
CMake 0.3%
GAP 0.3%
Shell 0.3%