timesync doc (#6783)

Signed-off-by: yefu.chen <yefu.chen@zilliz.com>
pull/7042/head
neza2017 2021-08-11 11:48:48 +08:00 committed by GitHub
parent 936c7e6b03
commit 67b3e6e2a1
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
6 changed files with 114 additions and 0 deletions

Binary file not shown.

After

Width:  |  Height:  |  Size: 15 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 48 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 55 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 84 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 197 KiB

View File

@ -0,0 +1,114 @@
# Timesync -- All The things you should know
`Time Synchronization` is the core part of `Milvus 2.0`, it affects all components of the system. This article describes the desgin detail of `Time Synchronization`.
In the `Milvus 2.0`, all events (such as `Create Collection`, `Insert`, `Search`, `Drop Collection`, etc.) have a `Timestamp` to indicate when does this event occurred.
Suppose there are two users, `u1` and `u2`, have connected to the `Milvus`, and do the following actions at their respective timestamp.
| timestamp | u1 | u2 |
|-----------|----------------------|--------------|
| t0 | create Collection C0 | - |
| t2 | - | search on C0 |
| t5 | insert A1 into C0 | - |
| t7 | - | search on C0 |
| t10 | insert A2 | - |
| t12 | - | search on C0 |
| t15 | delete A1 from C0 | - |
| t17 | - | search on C0 |
Ideally, `u2` expects `C0` is empty when it searches at `t2`, and could only sees `A1` at `t7`; at `t12` , the search from `u2` could sees both `A1` and `A2`, but only sees `A2` at `t17`. It's much easier to achieve these targets in `single-node` database. But for `Distributed System` , such like `Milvus`, it's a little difficult, and the following problems needs to be solved.
1. If `u1` and `u2` are on different nodes, and their time is not synchronized. To give an extreme example, suppose that the time of `u2` is 24 hours later than `u1`, then all the operations of `u1` can't been seen by `u2` until next day.
2. Network latency. If `u2` starts the `Search on C0` at `t17`, then how to ensure that all the `events` before `t17` have been processed. If the envents of `delete A1 from C0` has been delayed dure to the network latency, then it would lead to incorrect state: `u2` would see both `A1` and `A2` at `t17`.
`Time synchronization system` is used to solve the above problems.
## Timestamp Oracle(TSO)
Like [TiKV](https://github.com/tikv/tikv), `Milvus 2.0` provides `TSO` service, all the events must alloc timestamp from `TSO`not use local timestamp any more, so the first problem should be solved.
`TSO` is provided by `RootCoord` component, clients could alloc one or more timestamp at single request, the `proto` is defined as follows.
```proto
service RootCoord {
...
rpc AllocTimestamp(AllocTimestampRequest) returns (AllocTimestampResponse) {}
...
}
message AllocTimestampRequest {
common.MsgBase base = 1;
uint32 count = 3;
}
message AllocTimestampResponse {
common.Status status = 1;
uint64 timestamp = 2;
uint32 count = 3;
}
```
`Timestamp` is type of `uint64`, contains physical and logical parts.
This is the format of `Timestamp`
![Timestamp struct](./graphs/time_stamp_struct.png)
In the `AllocTimestamp` request, if `AllocTimestampRequest.count` if greater than `1`, then in the response, `AllocTimestampResponse.timestamp` indicates the first available timestamp.
## Time Synchronization
In order to understand the `Time Synchronization` better, firstly we need to introduce the data operation of `Milvus 2.0` briefly, taking `Insert Operation` as example.
- Users can configure lots of `Proxy` to achieve load balancing, in `Milvus 2.0`
- `SDK` could connect to any `Proxy`
- When `Proxy` receieves `Insert` Request from `SDK`, it would hash the `InsrtMsg` by `Primary key`, and then split the `InsertMsg` into different `MsgStream` according to the hash value.
- Each `InsertMsg` would be assigned an `Timestamp` before send to the `MsgStream.`
*Note: `MsgStream` is the wrapper of message queue, the default message queue in `Milvus 2.0` is `pulsar`*
![proxy insert](./graphs/timesync_proxy_insert_msg.png)
Based on the above information, we can know that the `MsgStream` have the following characteristics:
- In `MsgStream`, `InsertMsg` from the same `Proxy` must be incremented in timestamp
- In `MsgStream`, `InsertMsg` from different `Proxy` have no relationship in timestamp
The following figure shows an example of `InsertMsg` in `MsgStream`, the snippet contains 5 `InsertMsg`, 3 of them from `Proxy1` and others from `Proxy2`.
The 3 `InsertMsg` from `Proxy1` are incremented in timestamp, and the 2 `InsertMsg` from `Proxy2` are also incremented in timestamps, but there is no relationship between `Proxy1` and `Proxy2`.
![msgstream](./graphs/timesync_msgstream.png)
So the second problem has turned into this: after reading a message from `MsgStream`, how to make sure that all the messages earlier than this timestamp have been consumed. For example, when I read a message , whoes timestamp is `110` and produced by `Proxy2`, from `MsgStream`, but the message ,whoes timestamp is `80` and produced by `Proxy1`, is still in the `MsgStream`, what shoudl I do on this status?
The following graph shows the core logic of `Time Synchronization System` in `Milvus 2.0`, it should solve the second problem.
- Each `Proxy` will periodically reports the latest timestamp of every `MsgStream` to `RootCoord`, the default interval is `200ms`
- For each `Msgstream`, `Rootcoord` finds the minimum timestamp of all `Proxies` on this `Msgstream`, and inserts this minimum timestamp into the `Msgstream`
- When the consumer reads the timestamp inserted by the `RootCoord` on the `MsgStream`, it indicates that the messages eariler than this timestamp have been consumed, so all actions that depend on this timestamp can be executed safely
- The message inserted by `RootCoord` into `MsgStream` is type of `TimeTick`
![upload time tick](./graphs/timesync_proxy_upload_time_tick.png)
This is the `Proto` that usecd by `Proxy` to report timestamp to `RootCoord`:
```proto
service RootCoord {
...
rpc UpdateChannelTimeTick(internal.ChannelTimeTickMsg) returns (common.Status) {}
...
}
message ChannelTimeTickMsg {
common.MsgBase base = 1;
repeated string channelNames = 2;
repeated uint64 timestamps = 3;
uint64 default_timestamp = 4;
}
```
After inserting `Timetick`, the `Msgstream` should looks like this:
![msgstream time tick](./graphs/timesync_msgtream_timetick.png)
`MsgStream` will process the messages in batches according to `TimeTick` , and ensures that the output messages meet the requirements of timestamp. For more details, please refer to the `MsgStream` design detail.