From 84f32b3969cdcc676aaee428383b34b3b67dbdf5 Mon Sep 17 00:00:00 2001 From: Gao Hongtao Date: Tue, 17 Oct 2023 13:40:03 +0800 Subject: [PATCH] Update doc for releasing v0.5.0 (#344) --- README.md | 24 ++++++++--------- docs/installation.md | 50 ++++++++---------------------------- docs/installation/cluster.md | 19 ++------------ docs/observability.md | 2 +- 4 files changed, 24 insertions(+), 71 deletions(-) diff --git a/README.md b/README.md index f23e5e3e3..212f99b2c 100644 --- a/README.md +++ b/README.md @@ -36,12 +36,12 @@ The database research community usually uses [RUM conjecture](http://daslab.seas - [x] gRPC server - [x] HTTP server -### Distributed manager (v1.0.0) +### Distributed manager -- [ ] Sharding -- [ ] Load balance -- [ ] Distributed query optimizer -- [ ] Data queue +- [x] Sharding +- [x] Load balance +- [x] Distributed query optimizer +- [x] Data queue ### Data processor @@ -54,7 +54,7 @@ The database research community usually uses [RUM conjecture](http://daslab.seas - [x] Index processor - [x] TTL - [x] Cold data processor -- [ ] WAL (v0.5.0) +- [x] WAL ### Query processor @@ -65,15 +65,11 @@ The database research community usually uses [RUM conjecture](http://daslab.seas - [ ] Parallel executor - [ ] Cost-based optimizer -### Verification +### Storage -- [x] E2E with OAP and simulated data -- [x] E2E with showcases, agents and OAP -- [x] Space utilization rate -- [ ] Leading and trailing zero (v0.5.0) -- [ ] Stability (v0.5.0) -- [ ] Crash recovery (v0.5.0) -- [ ] Performance +- [ ] Compaction +- [ ] Merge data files +- [ ] Sparse index ### Tools diff --git a/docs/installation.md b/docs/installation.md index 9e85cd650..a8d9e5425 100644 --- a/docs/installation.md +++ b/docs/installation.md @@ -122,53 +122,25 @@ $ ./banyand-server standalone The banyand-server would be listening on the `0.0.0.0:17912` if no errors occurred. -## Setup Multiple Banyand as Cluster(TBD) +## Setup Multiple Banyand as Cluster -### Setup Standalone Nodes +Firstly, you need to setup a etcd cluster which is required for the metadata module to provide the metadata service and nodes discovery service for the whole cluster. The etcd cluster can be setup by the [etcd installation guide](https://etcd.io/docs/v3.5/install/). The etcd version should be `v3.1` or above. -The standalone node is running as a standalone process by +Then, you can start the metadata module by ```shell -$ ./banyand-server standalone --data-node-id n1 -$ ./banyand-server standalone --data-node-id n2 -$ ./banyand-server standalone --data-node-id n3 -``` - -`data-node-id` is the unique identifier of the standalone node. - -The standalone node would be listening on the `` if no errors occurred. - -### Setup Role-Based Nodes -The meta nodes should boot up firstly to provide the metadata service for the whole cluster. The meta node is running as a standalone process by +Considering the etcd cluster is spread across three nodes with the addresses `10.0.0.1:2379`, `10.0.0.2:2379`, and `10.0.0.3:2379`, Data nodes and liaison nodes are running as independent processes by ```shell -$ ./banyand-server meta +$ ./banyand-server storage --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 +$ ./banyand-server storage --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 +$ ./banyand-server storage --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 +$ ./banyand-server liaison --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 ``` -The meta node would be listening on the `` if no errors occurred. +## Docker & Kubernetes +The docker image of banyandb is available on [Docker Hub](https://hub.docker.com/r/apache/skywalking-banyandb). -Data nodes, query nodes and liaison nodes are running as independent processes by - -```shell -$ ./banyand-server storage --mode data --data-node-id n1 -$ ./banyand-server storage --mode data --data-node-id n2 -$ ./banyand-server storage --mode data --data-node-id n3 -$ ./banyand-server storage --mode query -$ ./banyand-server storage --mode query -$ ./banyand-server liaison -``` - -`data-node-id` is the unique identifier of the data node. - -The data node, query node and liaison node would be listening on the `` if no errors occurred. - -If you want to use a `mix` mode instead of separate query and data nodes, you can run the banyand-server as processes by - -```shell -$ ./banyand-server storage --data-node-id n1 -$ ./banyand-server storage --data-node-id n2 -$ ./banyand-server storage --data-node-id n3 -$ ./banyand-server liaison -``` +If you want to onboard banyandb to the Kubernetes, you can refer to the [banyandb-helm](https://github.com/apache/skywalking-banyandb-helm). diff --git a/docs/installation/cluster.md b/docs/installation/cluster.md index d419c06e6..fce7dc5f6 100644 --- a/docs/installation/cluster.md +++ b/docs/installation/cluster.md @@ -1,4 +1,4 @@ -# Cluster Installation +# Cluster Installation ## Setup Meta Nodes @@ -10,22 +10,7 @@ The etcd cluster can be setup by the [etcd installation guide](https://etcd.io/d There is an example: The etcd cluster is spread across three nodes with the addresses `10.0.0.1:2379`, `10.0.0.2:2379`, and `10.0.0.3:2379`. -Data nodes, query nodes and liaison nodes are running as independent processes by - -```shell -$ ./banyand-server storage --mode data --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 -$ ./banyand-server storage --mode data --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 -$ ./banyand-server storage --mode data --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 -$ ./banyand-server storage --mode query --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 -$ ./banyand-server storage --mode query --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 -$ ./banyand-server liaison --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 -``` - -The data node, query node and liaison node would be listening on the `` if no errors occurred. - -## Mix-mode Data Nodes - -If you want to use a `mix` mode instead of separate query and data nodes, you can run the banyand-server as processes by +Data nodes and liaison nodes are running as independent processes by ```shell $ ./banyand-server storage --etcd-endpoints=http://10.0.0.1:2379,http://10.0.0.2:2379,http://10.0.0.3:2379 diff --git a/docs/observability.md b/docs/observability.md index af0c38c35..2b13589a4 100644 --- a/docs/observability.md +++ b/docs/observability.md @@ -20,7 +20,7 @@ The Docker image is tagged as "prometheus" to facilitate cloud-native operations ## Profiling -TODO: Add details about the profiling support in BanyanDB, such as how to enable profiling, available profiling tools, and how to analyze profiling data. +Banyand, the server of BanyanDB, supports profiling automatically. The profiling data is collected by the `pprof` package and can be accessed through the `/debug/pprof` endpoint. The port of the profiling server is `2122` by default. ## Tracing