The Cassandra Archiver for Control System Studio is a set of plugins for CSS / EPICS, that use an Apache Cassandra database to archive control system values, providing outstanding performance compared to traditional archiving to a relational-database management-system.
An extensive manual, explaining the concepts of the Cassandra Archiver, a configuration guide and a step-by-step installation guide is included in the binary distribution and also available for online view.
This program and the accompanying materials are made available under the terms of the Eclipse Public License v1.0. By downloading the software you agree to be bound the the terms of this license. In particular, you have to agree to sections five (“NO WARRANTY”) and six (“DISCLAIMER OF LIABILITY”) before downloading the software.
Some files in the binary distribution have different copyrights and licenses. Refer to the information provided in the package for details about those copyrights and licenses.
The Cassandra Archiver for CSS has been replaced by the Cassandra PV Archiver. New users should use the Cassandra PV Archiver. Existing users should consider migrating to the Cassandra PV Archiver. This page is kept online so that existing users can access the documentation and old releases of the Cassandra Archiver for CSS.
Release 2.3.1 (June 6th, 2016):
Changes:
Fixed a bug in the reader part of the archiver. This bug could cause
many samples before the actually queried period to be returned instead of
only returning a single sample like intended. As a side effect, the
handling of exceptions that might occur while reading samples from the
archive was improved slightly. This release only makes changes to the
reader part. The archive engine, configuration utility, and clean-up tool
remain unchanged.
Release 2.3.0 (October 20th, 2014):
Changes:
This release brings a number of minor improvements: It bundles a patched
version of the CSS archive engine that correctly reconnects after the
connection to the database has been lost. It also has improvements in
the clean-up tool, that make the clean-up process less likely to fail if
interrupted. Even if the clean-up process is interrupted, it is applied at
least partially. Finally, there are a number of improvements in the sample
retention process and the archive reader, that help avoid burdening the
Cassandra database with many tombstones for deleted samples.
Release 2.2.0 (July 28th, 2013):
Changes:
This release brings performance optimizations in the sample compressor.
When using compressed samples, the new code will reduce the number of
database requests needed significantly, thus reducing the CPU and I/O
load. The format of the stored data has not changed with this release,
thus it is fully compatible with the 2.1.0 release.
Release 2.1.0 (March 13th, 2013):
Changes:
This release brings support for Apache Cassandra 1.2, which offers some
very useful new features (e.g. virtual nodes). This release also introduces
a revised serialization format for sample values, that is more space
efficient for integer values. As a consequence older releases cannot read
data written by this release of the Cassandra Archiver, however this
version can still read the old format, allowing for an upgrade without any
conversion needed.
Release 2.0.1 (March 10th, 2013):
Changes:
This release fixes a bug that appeared when the interval configured for
samples was very long (approximately more than 10 hours). Usually, such
an interval is only used with compressed samples, which limited the scope
of this bug.
Release 2.0.0 (January 7th, 2013):
Changes:
This release uses a new structure for the Cassandra database, bringing a
major boost in performance. It also fixes some bugs and brings general
stability improvements. This release of the Cassandra Archiver needs
Apache Cassandra 1.1.x.
Release 1.2.0 (September 17th, 2012):
Changes:
This release brings support for multiple compression threads. This
can help if the compression does not "catch up" with the new data.
It also improves the queuing mechanism used for compression to be
more memory efficient.
Release 1.1.0 (January 28th, 2012):
Changes:
This release uses an improved compression algorithm for calculating
compressed samples. If a suitable compression level with a shorter
compression period is available, the samples from this compression
level are used instead of using raw samples. This increases the
throughput of the samples compressor significantly.
Release 1.0.1 (January 24th, 2012):
Changes:
Bugfix release: Fixed Cassandra authentication problem in archive
engine. Fixed infinite loop in archive reader (also affected
sample compression). Included plugin_customization.ini in
archive engine product. Fixed a bug in the sample cache used by
the sample compressor.
Release 1.0.0 (January 20th, 2012):
Changes:
First release of the Cassandra Archiver.
You can contact the author of this software, Sebastian Marsching, at cassandra-archiver-maintainer <at> oss <dot> aquenos <dot> com.
Visit other projects on oss.aquenos.com.
© 2012-2014 aquenos GmbH