[Feat] document v4 cluster size #198
Conversation
| | `baas-alpha` | 3 | 8 | 16 (at least) | Yes | 750 | High throughput and IOPS | | ||
| | `baas-zero` | 3 | 2 | 2 | Yes | 300 | High throughput and IOPS | | ||
| | `isa95` | 2 | 2 | 1 | NO | N/A | | | ||
| | `libre-core` | 3 | 1 | 2 | No | N/A | HA requires 2 pods, but 3 is to avoid hotkey issues and balance load | |
There was a problem hiding this comment.
libre-core was renamed in v4 to isa95, its line item can be removed
bpmn-engine was renamed in v4 to workflow, its line item can be removed
libre-audit -> was folded into isa95 / workflow, its line item can be removed
router is no longer required, its line item can be removed
| | `quest-db` | 1 | 4 | 8 | Yes | 250GB | High Throughput and IPOS | | ||
| | `restate` | 3 | | | Yes | 50 | High Throughput and IPOS | | ||
| | `appsmith` | 3 | 4 | | Yes | 50 | High Throughput and IPOS | | ||
| | `grafana`* | 3 | 0.5 | 2 | No | 20-50 | Storage can be in host or in object bucket. | |
There was a problem hiding this comment.
If grafana is defined below in the monitoring stack it can remove it from here
|
|
||
| For high availability, Rhize recommends a **minimum of three nodes** with the following specifications. | ||
|
|
||
|
|
There was a problem hiding this comment.
Remove 2nd new line
| | `router` | 2 | 1 | 2 | Yes | <1 | Requires volume to compose supergraph | | ||
| | `quest-db` | 1 | 4 | 8 | Yes | 250GB | High Throughput and IPOS | | ||
| | `restate` | 3 | | | Yes | 50 | High Throughput and IPOS | | ||
| | `appsmith` | 3 | 4 | | Yes | 50 | High Throughput and IPOS | |
There was a problem hiding this comment.
I'm unsure what HA appsmith involves. Need to follow up with those that know.
| | `keycloak` | 2 | 1 | 2 | No | N/A | | | ||
| | `keycloak-postgres` | 2 | 1 | 2 | No | 200 | Runs in pod with `keycloak` | | ||
| | `router` | 2 | 1 | 2 | Yes | <1 | Requires volume to compose supergraph | | ||
| | `quest-db` | 1 | 4 | 8 | Yes | 250GB | High Throughput and IPOS | |
There was a problem hiding this comment.
I'm unsure what HA QuestDB involves. Need to follow up with those that know.
| | CPU Speed (GHz) | 3.3 | | ||
| | vCPU per Node | 16 | | ||
| | Memory per node (GiB) | 32 (64 is better) | | ||
| | Persisted volumes | 12 | |
There was a problem hiding this comment.
Persisted volumes is actually higher ~16. I need to confirm what QuestDB/AppSmith HA invovles
| However, some deployments prefer to separate monitoring to its own cluster. | ||
|
|
||
| | Service | Pods for HA (replica count) | vCPU cores per pod | Memory per pod | DiskSize (GiB) | | ||
| |-------------------------|-----------------------------|--------------------|----------------|----------------| |
There was a problem hiding this comment.
Might need to checking with the devops guys if this is still applicable in the latest tempo/loki/prom distributed helm-charts.
No description provided.