[CASSANDRA-2398] Type specific compression

• Type: New Feature
• Status: Resolved
• Priority: Normal
• Resolution: Later
• Labels: compression
• Assignee: Unassigned
• Reporter: Stu Hood

Reviewers

Sylvain Lebresne

• Created: 28/Mar/11 19:31
• Updated: 16/Apr/19 09:33
• Resolved: 24/Jun/14 22:21

Description

Cassandra has a lot of locations that are ripe for type specific compression. A short list:

Indexes

• Keys compressed as BytesType, which could default to LZO/LZMA
• Offsets (delta and varint encoding)
• Column names added by 2319

Data

• Keys, columns, timestamps: see http://wiki.apache.org/cassandra/FileFormatDesignDoc

A basic interface for type specific compression could be as simple as:

public void compress(int version, final List<ByteBuffer> from, DataOutput to) throws IOException
public void decompress(int version, DataInput from, List<ByteBuffer> to) throws IOException
public void skip(int version, DataInput from) throws IOException

Issue Links

• blocks: CASSANDRA-674 New SSTable Format
• blocks: CASSANDRA-2319 Promote row index
• is related to: CASSANDRA-2812 Allow changing comparator between compatible collations