Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

introduce latest cf #95

Open
wants to merge 5 commits into
base: master
Choose a base branch
from
Open

introduce latest cf #95

wants to merge 5 commits into from

Conversation

BusyJay
Copy link
Member

@BusyJay BusyJay commented Jun 5, 2022

Add a cf (column family in RocksDB) named "latest" to store the latest version of keys in MVCC.

Signed-off-by: Jay Lee <[email protected]>
Signed-off-by: Jay Lee <[email protected]>
- key set the original key without any encoding or version
- value set the same value in write cf but include the corresponding version.

For example, insert k1 with version v0 and value dummy will insert two keys
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This will double the write amplification, not only by size but also by key value pairs.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, it's addressed in the drawback section.


Because all existing cfs are updated just as before, so there are no major compatibility issues.

But using latest cf should be triggered explicitly by client. Client should ensure only when it updates all keys with latest cf will it ask TiKV to query using latest cf.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It is a bit of complicated.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Usually it won't as the client already needs to manage different ranges, just like table in TiDB or key space.


## Alternatives

unistore separates the latest version and other versions by adjust file format. So when flushing or compacting, it will make latest versions key be the first part, and the rest in the second part. This approach doesn't have write overhead, but is not backward compatible in TiKV.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How about create a history cf to store as many history version as it can, the write cf store the latest several versions, and background thread like GC thread move existed history versions to history cf.
For point get, using RocksDB's user timestamp feature to get the latest version can see, this also can prevent creating iterator.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There is no compatibility issue for this solution.

Copy link
Member Author

@BusyJay BusyJay Jun 5, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Compaction filter is not reliable, we have seen issue caused by not having compaction in time like tikv/tikv#12729. This proposal doesn't depend on the user timestamp, which is not used in production yet. And can be landed table by table instead of the whole cluster, have less impact on existing code. And this proposal should have better scan performance than relying on compaction filter.

Copy link
Member Author

@BusyJay BusyJay Jun 5, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added as an alternative with more argument.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Using history cf doesn't need the compaction filter, gc just check the history cf sst file's max ts to determine wether to drop the whole sst file.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Compaction filter is not used for dropping history sst, but for moving data from write cf to history cf.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Compaction filter is not necessary, I think version statistics-based GC is much better.


As explained, the reason why seek is necessary is because TiKV has no idea what versions are available. Otherwise it can just use `get` to query the specific key, which is a lot cheaper.

## Detailed design
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How about lightning, cdc, br, tiflash's compatibility?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

They are not part of TiKV. I would like to talk about them in another doc.


TiKV doesn't need to know all existing versions of keys. In fact, most of the time, v0 is larger than any existing versions of keys in TiKV if there is more read than write. So it should be enough to just let TiKV knows the latest version of all keys.

The RFC propose to add a new cf named "latest". When a key is inserted using transaction API, it should update write cf (and default cf) as current does. In addition, it also insert a key to latest cf with
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this double write may also double the total disk size in some scenarios.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe for those new table(new range) in the existing cluster and the new created cluster we can use the new mechanism(only write latest cf, move the old version to write cf only when the older version existed) is a better choice?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It would work, though it loses the ability to upgrade existing table to new format online.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Or we can introduce 3 formats. One is the original one, second is double write, third is move on update. They can be upgraded online one by one. For new tables, third format is used by default.

Signed-off-by: Jay Lee <[email protected]>
@BusyJay
Copy link
Member Author

BusyJay commented Jun 12, 2022

Although I updated the RFC to write latest CF first, and then move keys between CFs, I suggest to always double write in POC as it has less compatibility issues.

For example, supposing there is no key in latest cf. Inserting k1 with version v0 and value foo will insert one key:
- to latest cf, k1 -> (foo, v0 and other meta)

Inserting v1 again with version v1 and value bar will insert two keys:
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
Inserting v1 again with version v1 and value bar will insert two keys:
Inserting k1 again with version v1 and value bar will insert two keys:

Is it k1?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes.

Signed-off-by: Jay Lee <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants