Skip to content

Commit 86b48f1

Browse files
authored
update byoc aws usage based information (#560)
* update byoc aws usage based information * update mpu
1 parent 94cfc6b commit 86b48f1

File tree

1 file changed

+59
-15
lines changed

1 file changed

+59
-15
lines changed

docs/byoc.md

Lines changed: 59 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -18,6 +18,36 @@ Alternatively, users can enable two AZ for nodegroup.
1818

1919
![BYOC architecture with multi-az nodegroup](/img/byoc_architecture_white.png)
2020

21+
## Pricing Model and License Management
22+
23+
User has two options of pricing model:
24+
25+
- Bring your own license (BYOL)
26+
- AWS usage based
27+
28+
29+
### AWS Usage Based
30+
31+
When `pricingModel` is configured as `aws_usage_based`, Timeplus BYOC deployment will call [AWS Marketplace metering API](https://docs.aws.amazon.com/marketplace/latest/APIReference/API_marketplace-metering_MeterUsage.html) to report usage every hour. The MPU metrics is defined as the total number of cores deployed for Timeplus at the collect time.
32+
33+
Use helm version `10.0.8` or higher is required.
34+
35+
#### MPU Sizing Guide
36+
37+
MPU (micro processing unit) is the Timeplus usage unit for usage based pricing. Related EC2 instance type will be used for EKS nodes according to the minial MPU size.
38+
39+
| MPU Size | CPU Cores | Memory | Recommended For |
40+
| :---- | :---- | :---- | :---- |
41+
| 4 MPU | 4 | 16 GB | Development, testing, small workloads |
42+
| 8 MPU | 8 | 32 GB | Production, small workloads |
43+
| 16 MPU | 16 | 64 GB | Production, normal workloads |
44+
| 32 MPU | 32 | 128 GB | Production, large workloads |
45+
46+
### Bring your own license (BYOL)
47+
48+
When `pricingModel` is configured as `byol`, use can install Timeplus enterprice license by himself. Refer to this [doc](https://docs.timeplus.com/server_config#license) for how to manage license.
49+
50+
2151
## Prerequisites
2252

2353
### Required Tools
@@ -358,6 +388,31 @@ Redeploy:
358388
timeplus_byoc --config timeplus-stack-{cluster-name}
359389
```
360390

391+
### Scale EKS Nodes
392+
393+
By default, a 4 nodes EKS cluster will be created, use can update related configuration in cluster.yaml to change this.
394+
395+
```yaml
396+
minNodes: 4
397+
maxNodes: 10
398+
desiredNodes: 4
399+
```
400+
401+
As EKS cluster autoscaler is installed, the node will be automatically provisioned when the workloads increases and exceeds the current capability. User can set the maxNodes to have an upper limits of how many nodes can be automatically provisioned.
402+
403+
404+
### Deploy on existing Subnet
405+
406+
User can manually update the [`eks-stack.yaml`](https://gist.github.com/gangtao/57720fe5d48f7577af042c281ddb6cb5) to depend on existing subnet to create the EKS on existing subnet.
407+
408+
Replace following default values with acutally Subnet information:
409+
410+
- VpcId
411+
- PublicSubnet1Id
412+
- PublicSubnet2Id
413+
- PrivateSubnet1Id
414+
415+
361416
## Cleanup and Deletion
362417

363418
### Delete Entire Deployment
@@ -417,8 +472,11 @@ clusterName: {cluster-name}
417472
# AWS region
418473
region: us-east-1
419474

475+
# Pricing Model
476+
pricingModel: byol # support aws_usage_based or byol
477+
420478
# Helm chart version
421-
helmChartVersion: 8.0.8
479+
helmChartVersion: 10.0.7
422480

423481
# Node configuration
424482
nodeInstanceType: m5.xlarge
@@ -473,15 +531,6 @@ helmChartPath: ""
473531
customerHelmValuesFile: ""
474532
```
475533
476-
### MPU Sizing Guide
477-
478-
| MPU Size | CPU Cores | Memory | Recommended For |
479-
| :---- | :---- | :---- | :---- |
480-
| 2 MPU | 2 | 8 GB | Development, testing, small workloads |
481-
| 4 MPU | 4 | 16 GB | Production, moderate workloads |
482-
| 8 MPU | 8 | 32 GB | Production, high-throughput workloads |
483-
| 16 MPU | 16 | 64 GB | Production, very large workloads |
484-
485534
## Best Practices
486535
487536
### Naming Conventions
@@ -550,8 +599,3 @@ Users can use AWS EBS backup mechanism to backup/restore the persistent vol of E
550599
- [Use Kubernetes volume snapshots with EBS CSI](https://docs.aws.amazon.com/eks/latest/userguide/ebs-csi.html) AWS’s EKS documentation for the EBS CSI driver covers how to enable snapshot capability in EKS using VolumeSnapshotClass / CSI snapshot controller
551600
- [AWS Blog: Using EBS snapshots for persistent storage with EKS](https://aws.amazon.com/blogs/containers/using-amazon-ebs-snapshots-for-persistent-storage-with-your-amazon-eks-cluster-by-leveraging-add-ons/) Describes using CSI snapshot features to capture point-in-time copies of EBS volumes.
552601

553-
### Manage admin user credentials
554-
555-
## License Management
556-
557-
Refer this [doc](https://docs.timeplus.com/server_config#license) for how to manage license with BYOL (bring your own license) mode

0 commit comments

Comments
 (0)